Padded Public Pensions

In the Friday, May 21, 2010, issue of The New York Times , there was a front-page story by reporters Mary Williams Walsh and Amy Schoenfeld entitled “Padded Pensions Add to New York Fiscal Woes.” The reporters highlighted the fact that many financially strapped New York State cities are saddled with pension costs far in excess of what their financial experts estimated when the pension plan provisions were put into place.

Unfortunately, this is an all-too-familiar story: a governmental entity that irresponsibly agreed to rich pension benefits to allow government workers to retire very young, receive an exceptionally high percentage of their pay, and have taxpayers feel the financial burden decades later. However, the example provided relative to Yonkers, New York, is especially outrageous.

Final Year Overtime

A pension is usually determined by multiplying a percentage determined by the number of years of employment by a number called the “final average earnings.”

Many governmental pension plans calculate final average earnings to which a pension plan percentage is calculated in terms of pay earned in the last year of a worker’s employment. The percentage calculation is usually in excess of what private sector workers get. The final year is always the highest year of pay for government workers, and the calculation of earnings based on the final year’s W-2 income invites abuse. Most private plans prevent abuse by averaging the highest five years of earnings.

Many government workers volunteer for more overtime work and pay in their last year of employment than at any other point in their career. As a result, one retired Yonkers worker, whose base pay is $74,000 a year, started receiving a $101,233 pension when he retired at age 44. To put age 44 retirement into perspective, the worker’s life expectancy at retirement is around 81, which means that Yonkers taxpayers will carry him at his inflated pension for another 37 years, after he worked 20 years.

Overtime for Moonlighting

Moreover, many cities and towns stuck the taxpayers with pension obligations for work performed for private firms. One example was the pension obligation for overtime work police officers performed as flagmen for Con Ed, the large private utility. Con Ed paid the wages and work-related expenses for the police officers, but all the overtime pay went into the workers’ earnings base for pension plan calculations.

“Low-Ball” Estimates

One comment from a spokesperson for the mayor of Yonkers may have been the most outrageous of all. The individual was quoted as saying that pension cost estimates were “often lowballs,” presumably so the city could agree to get stuck without arousing attention from the public. Throughout the article, there were several points at which it was clear that estimates were wildly low of their eventual cost. For example, according to the article, Yonkers city officials were told, and communicated to the public, that the richer pension formula for police would cost $1.3 million a year, but the yearly cost is now $3.75 million and rising.

One fundamental problem with government today is that we have excessive transparency for issues that do not matter, and no transparency on issues like pension and retiree medical costs that have huge financial implications. Some of the assumptions on government pension liability are transparent:

Financial reports typically reflect the discount rate used on future liabilities and the investment return assumptions;
They identify the life expectancy tables used; and
They also tell us whether pension benefits are subject to cost-of-living adjustments.

What is not transparent are the following:

How numbers underlying the assumptions are derived;
How the cost-of-living adjustments are calculated; and
How compensation increases are determined.

The pension padding results from a total control breakdown on compensation increases in the last year of employment. Employees who are part of a pension system that calculates pay according to the compensation obtained in the last year of employment will become expert at manipulating the system to maximize their compensation.

Cost-of-Living Adjustments

Cost-of-living adjustments, which are included in almost every government pension plan, prevent a government from growing its way out of the pension liability problem. If the pension accounts for 10% of the government’s budget, and the economy grows by 5%, the pension liability will likely grow, after cost-of-living adjustments, at a rate comparable to the 5%, thereby negating the benefit of the growth.

Remedies

What can we do?

Stop the bleeding. Governments should radically renegotiate these pension obligations for new hires, and for those far away from retirement. At Pitney Bowes, we amended our defined benefit pension plan in 1997 to eliminate unsustainably rich benefits, but we provided a transition plan to protect those within a few years of normal retirement age.
Increase transparency. Calculations should be shared with the public. These calculations involve a certain amount of strongly supported scientific analysis, like life expectancy. However, pension calculations also require plan administrators to estimate investment return on pension assets, which involves a huge amount of judgment. Government plan administrators routinely overestimate investment returns.
Modify cost-of-living adjustments. Government retirees typically have health plans protected from inflation, and are usually not saddled with tuition increases and increases in costs of shelter. In fact, many economists point out that both the absolute costs of living and the rate of increase in these costs is lower for the elderly (other than the cost of health insurance prior to age 65 when they do not have access to either Medicare or a retiree medical program through their employer). Any cost-of-living adjustment should be modified to reflect this.
Change the actuarial estimates by requiring actuaries to factor in behavioral responses to pension plan provisions. Common sense tells us that if someone can volunteer for overtime pay in one year to increase their income for the next 35 years, they will do that. However, actuaries not only do not take this “gaming of the system” into account, they are probably not allowed by traditional actuarial pension calculation principles to do so. They are required to take voluntary and involuntary employment turnover into account, but, for whatever reason, they leave voluntary overtime decisions out of the calculation. This has to change.

This pension calculation issue illustrates a much bigger problem with government: the boring and highly technical work on budgets and cost forecasting is the most important and the least understood part of government. There are many examples of flawed government cost forecasting that have profound implications:

The Congressional Budget Office (“CBO”) costing of the recent health insurance reform legislation, which is already believed to have been $115 billion too low;
The flaws in pension forecasting across all levels of government. I saw this first hand relative to the U.S. Postal Service, where the federal Civil Service Retirement System overestimated postal service pension liabilities by $78 billion until the industry and the Postal Service found the problem in the Fall of 2002.
The CBO assumption that preventive health practices and services would save no money in the Medicare system. This may be one of the most monumentally stupid and consequential failures of the budget scoring system. The CBO only looked at selective preventive screenings and looked out only three years (even though it forecast the rest of the legislation out 10 years), and did not take into account benefits for any other part of government beyond that covered by the legislation.

This defies common sense, but it had profound consequences. It caused growth-killing taxes to be added to the health insurance legislation to make the budget appear to balance. It reduced the potential pool of money that could have gone to prevention. It also reduced the opportunity for other vital investments in health care system transformation.

We have to get these flawed systems fixed, or else government will make some very big mistakes in its attempt to be stewards of taxpayer and bondholder money.