Analyst Sees ‘Moral Hazards’ in GASB Pension Accounting Standards

The National Conference on Public Employee Retirement Systems (NCPERS) has published a detailed new white paper, “The Case for New Pension Accounting Standards.”

Tom Sgouros, a research associate in the computer science department at Brown University and former senior policy adviser to the Rhode Island General Treasurer, authored the peer-reviewed white paper for NCPERS. He argues the existing accounting rules for public defined benefit pension funds as formulated in the various statements of the Governmental Accounting Standards Board (GASB) “provide a misleading picture of the health of a pension system and a poor guide to decisions by policy makers.”

According to Sgouros, the GASB rules furthermore create “a degree of moral hazard for stakeholders,” whereby the consequences of important decisions are not felt for years, possibly decades, after the decisions are made.

“Following the rules closely means hiding some risks and exaggerating others, attending to unimportant details while ignoring important assets, and basing momentous decisions about the distant future on layers of guesses while delaying the consequences of poor decisions until the decision makers are long gone,” Sgouros says. 

As outlined in the white paper, the current GASB accounting approach involves calculating a net pension liability by subtracting the assets on hand from the estimate of total liability. Plan sponsors must present the net pension liability in a statement of net position for the employer, and future contributions cannot be counted against current liabilities. This means effectively that the promise of future payments has no value.  Other elements of the GASB approach include a limitation on the discount rates used to calculate liabilities, and the fact that all assets are counted as equal in value if their present market values are equivalent. Under this system, the funded ratio becomes the key indicator of pension health, and there is essentially no acknowledgement of the economic strength or weakness of the plan sponsor in measuring the health of the plan.

Looking across these various elements, Sgouros presents a series of alternative approaches that he believes could be more effective. He emphasizes that the goal of these recommendations is the same as the goal of the GASB rulemaking committees—to preserve the valuable institution of the pension plan, and to make clear how best to manage these plans so as to keep them strong and solvent for this and future generations of public employees and citizens.

On the point of calculating a net pension liability by doing a simple subtraction of the assets on hand from the estimate of total liabilities, Sgouros suggests this approach is inherently flawed in that it combines low-accuracy numbers on the liability side with high-accuracy numbers on the assets side. Plan sponsors should not expect to get anything but a very rough estimate of their funded status in this manner, he argues, and as such, this approach is unsuitable as a basis for making important decisions.

“The calculation is a useful exercise for any system, but differences cannot be precise, so important decisions must not depend on that precision as they currently do,” Sgouros writes.

Sgouros argues that pension-sponsoring employers simply should not be required to include pension liability in their statement of net position.

“As we see in case after case (Detroit, Stockton, Connecticut, etc.), including this vast but very uncertain number in the net position increases pressure on policy makers and guarantees it is used inappropriately to make policy decisions,” Sgouros suggests. “It is not a debt known with anywhere near the precision of accounts payable, for example, and combining the two merely creates inaccurate financial statements.”

While not all future contributions should be counted against current liabilities, Sgouros says, some future contributions should be. His argument is based on the importance of accounting clarity.

“A public pension system is an ongoing, permanent, enterprise,” he writes. “Employees who support their predecessors will have successors to support them in turn. Introduction of a standard cost is an alternative way to provide the necessary clarity while responsibly allowing future payments to pay for the future retirement benefits of current employees.”

According to Sgouros, it would further be wise for the system to revert to the requirement to divulge discount rate in use when liability estimates are made, rather than prescribing discount limitations.

“Decreasing the importance of the total liability calculation makes the chosen discount rate less important,” he writes. “Estimates of future liabilities must be made responsibly, but the discount rate limitations effectively increase the size of the liability, increasing the pressure to adopt risky funding policies.”

The white paper suggests that plan sponsors’ use of “normal costs” in allocating 100% of the projected cost of a single individual’s pension to that specific individual leaves them prone to making errors when assessing the financial position of the entire system.

“Normal costs cover a fraction of a single individual’s pension expense, while a standard cost covers the remaining expense of keeping the system running,” the white paper explains. “Normal costs are estimates, prone to errors. By introducing a standard cost to cover part of the expense of maintaining the system, the accounting can accommodate the collective nature of a pension system and provide estimates of the whole system cost with greater confidence.”

Sgouros then suggests that asset values should be weighted in inverse proportion to their risk. The current standards require that all assets are counted as equal in value if their present market values are equal.

“It is almost tautologically true that a fund with a risky capital structure is likely to be unable to cover its liabilities even if the current value of the assets equals the value of the liabilities,” Sgouros writes. “This is precisely what it means to have a risky capital structure. Funds should be evaluated with risk at the forefront, not tucked inside an estimated rate of return.”

Sgouros concludes by arguing that the use of a “depletion date estimate using risk-weighted assets” should become the key method for indicating the fiscal health of a public pension—and that the economic strength of a plan sponsor is actually very important to understanding whether a pension system is in crisis or not.

The full text of the analysis is available for download here.