Pensions Finally Come of Age

I recently celebrated the 35 year anniversary of my Bar-Mitzvah. It was a well-earned coming of age, but it also pegs my birthdate in the eyelet of a modern historical crossroads. An unlikely chain of events was set in motion that would ultimately culminate in an historic bailout of an inexorable institution that symbolized, perhaps, the most enduring human value of the 20th century, the pension plan promise.

On the evening of August 8, 1974, after an acrimonious 2-year national scandal culminating in the release of the Watergate tapes, and facing impeachment hearings, President Richard Nixon announced his resignation to the US public over national airwaves. While those who lived through it knew Nixon primarily for that Watergate scandal, those of us born afterwards have come to know the 37th US president as the man who ended the Gold Standard, leaving a legacy of Fiat money for their descendants to support with their future productivity. I was born days after his resignation and I’m told that my incubation was filled with my mothers prayers that he would be gone by the time I arrived.

A few weeks later and amidst my parent’s relief over Nixon, the US would announce, possibly, the most ambitious entitlement program in the history of the world, The Employee Retirement Income Security Act, otherwise known as ERISA. One of the fun aspects of the 1970s was that the laws were actually named for what the law was intended to do. ERISA sought to provide a FDIC-like insurance on a workers pension, should his employer go bankrupt and be unable to pay it. In the event of a bankruptcy, the government would assume the pension liability. This act created a mega-industry of accountants, lawyers, actuaries, money managers, union negotiators, and regulators. I entered the industry as an actuary in 1996 when I, once again, coincided with American economic history and worked for AT&Ts pension department. The department was managing the forced divestiture of the AT&T monopoly and quickly became an actuarial consulting firm that would employ close to a thousand people.

When I began working on pensions in 1996, the size and impact of plans was not measured in dollars but in lives. The pension was the responsibility of Human Resources, not the Chief Financial Officer. The yield on the 30-year Treasury was 7% and the financial impacts of pension liabilities were inconsequential. This would quickly change after the bursting of the 2000 dot.com bubble, an event that was called “The Perfect Storm” when both the equity markets and interest rates declined, taking pension funding in the wrong direction. It is here where I’ll explain that the pension liability is calculated by adding up the present value of all of the future annuities it promises to pay out, and that when interest rates decline, the value of the liability increases, i.e. you need to fund the liability with more money today because you expect to earn less interest towards what will be needed to pay off the eventual pension.

Companies used the pension as a both a retention tool to keep employees when they wanted them and as a termination tool that allowed them to provide incentives to make large workforce reductions. When pensions had a surplus, companies would often enrich the promise as a negotiation tool instead of giving wage increases. When rates were high, the value of these incentives were so small that they seemed costless. The “Perfect Storm” of 2000-2002 exposed these liabilities as material and began to shift the corporate governance of pensions from the Human Resources office to the Chief Financial office.

An example of the lack of financial understanding, even in large systems, can be seen by the Pennsylvania State Employees Retirement System (SERS ). In 2000 (Actuarial Report), the plan covered over 100,000 employees (not including the equivalently large teachers plan called PSERS). The plan had a reported liability of $20B against reported assets of $26B and appeared to be significantly overfunded. In 2001, the system decided to increase the annual pension for the majority of current and future employees from 2% per year to 2.5% per year. An individual’s annual benefit is defined as their average salary times their years of service times this percentage. To illustrate., an employee who made 50K per year with 20 years of service would have their benefit increase from 20K/year to 25K/year. Increasing the promised benefit of each participant in this way effectively increased the liability by 25%. The sponsors of the plan essentially spent the 2000 surplus on this increase. But was that surplus really a surplus? I would give a resounding no.

The interest rate that was used to calculate the liability was an astonishing 8.5%. This was a level that hadn’t been seen in 10 years (the 30 year Treasury was 6% in 2000). Had the more current discount rate been used, there would have been no surplus. Under 2021 treasury rates, the increase alone is costing today’s Pennsylvania taxpayers an unimaginable sum of money. The intuitive way to understand this is that the pension managers in 2000 expected to earn an 8.5% on their assets but today, they are only earning 3% on those assets. That is 5.5% less per year being earned on the $26B portfolio for however long Pennsylvania plans to pay these benefits. Remember that this is only just one plan in one state but the practice was mainstream and nothing, not even the financial crisis experienced in 2001, could deter such practices.

One thing that every pension expert understood in the ‘90s was that pensions were regressive and that governments providing tax incentives or credit guarantees were actually providing for individuals who aren’t saving at the expense of those who are. It was a canonical assumption that a pension system would never be bailed out by a government, even though the Pension Benefit Guaranty System (PBGC) provided exactly that to a bankrupted company. Traders would refer to the protection as the “PBGC Put” which actually provides an incentive to companies to maximize the risk they take in the management of the pension. The methods in which the assets and liabilities are disclosed in financial statements reflect the summation of these worldviews. Asset returns are smoothed over periods as long as 5 years, while liability changes are smoothed out over even longer periods of time, for example, the future working lifetime of the currently employed participants.

By the time the assets and liabilities had fully reflected the 2000-2002 perfect storm, the economy was getting ready for another crisis. I was advising funds in an industry practice that would eventually be known as “Liability Driven Investing” (LDI). The idea behind LDI is that the return on the pension assets isn’t as important as the return relative to the liability. If the S&P500 goes down by 20%, but your investment manager only went down by 19%, they would be lauded as heroes even though the plan is left now with a large hole to fill. Moreso, if rates decreased by 100bps - the liability would increase by around 12% - so the plan should want either an investment or a hedge that paid off that 12%. The idea could not overcome the established financial reporting and incentives already in place and did not catch on. My colleagues and I were regularly dismissed from boardrooms, in some cases with laughter. It was here where I realized that the pension was doomed to an eventual risk management failure and I left the industry wanting no part of it.

As one might guess, the Great Financial Crisis of 2008 was not good for pensions. It was everything the Perfect Storm was but orders of magnitude greater. Now rates were heading below 4% and liabilities were continuing to compound future increases. I watched from a different industry, fairly certain that some pension would ultimately require a government bailout. That day, surprisingly, never came. While the pension system is a collection of guaranteed failures, its difficult to know when the day comes where the failure materializes. One thing that did change after the GFC was that more and more pension plans began to embrace using LDI in their portfolios. This foreshadows the impending reckoning for a few reasons: 1) The damage is already done and rates are already close to zero. The fire insurance was bought after the match was struck; 2) Regulation post GFC made liquidity a larger and more immediate concern, namely the risk of a sharp rate increase that can cause bond losses.

We saw the first sign of the failure on September 26, 2022 when the Bank Of England announced a “temporary program” to buy bonds. This amounts to a bailout, particularly under the backdrop of record inflation in the UK and the unravelling of the British Pound, which sent gilt yields soaring on the evening of the 24th. One must ask, incredulously, why is a rising yield a problem? It’s an event so rare under anyone’s career in managing a pension that risk managers seemingly don’t even consider the scenario. A large spike in yields punishes the returns of the bond portfolio, and the longer (or more liability matched) the bonds, the larger are the losses. The details get complicated but I’ll spend one more paragraph taking this home.

LDI managers, such as Blackrock, L&G, and Scrhoders, use interest rate swaps to hedge themselves of the P&L risk associated with investing in long duration bonds. There are two problems with the recent rise in yields - and the bailout addresses the 2nd one.

• Problem #1: The sharp rise in yields causes losses because LDI managers hedge using fixed receiver swaps. Margin on these losses must be posted daily to the extent the swaps are centrally cleared, which in 2022 is likely to be the majority.

Problem #2: Failure to post the margin will result in the liquidation of the swaps, leaving Blackrock, et al. without a hedge at the exact time that they are likely to need it most.

Pensions around the world are coming of age in the collateral damage of worldwide inflation and the boom-bust cycles of central monetary planning. The 20th century pension promise has yet to read from the Torah to lead its congregation, and convince them that their future is in capable hands. It has yet to tell the world that it can outlive even the first generation of lives who benefited from it. The idea of “too big to fail” institutions has seemingly overshadowed the ideals of a retiree society. It may be that the original condition for the promise, which was a government in control of fiat money, was the limiting factor of its longevity. In order to financially plan for the unexpected mortality of people, we might need to do so with a money that is, itself, immortal.