Industrials

Nir Kaissar is a Bloomberg Gadfly columnist covering the markets. He is the founder of Unison Advisors, an asset management firm. He has worked as a lawyer at Sullivan & Cromwell and a consultant at Ernst & Young.

It’s time for General Electric Co. to do some soul-searching about its pension problem.

As Bloomberg News reported last week, GE’s pension was underfunded by a staggering $31.1 billion at the end of 2016 -- the biggest shortfall among S&P 500 companies.  

So far, GE seems to be pointing fingers at everything but itself. Company spokeswoman Jennifer Erickson has attributed the pension predicament to the 2008 financial crisis and subsequent low interest rates.

In fairness, GE’s pension was in good health before the financial crisis. It was overfunded every year from 1999 to 2007, and GE’s surplus was $15.2 billion at the end of 2007. But in 2008, the pension portfolio tumbled by roughly 28 percent, and suddenly it was underfunded by $6.8 billion.

Turning Point
GE's pension fell deeper into the red after the 2008 financial crisis
Source: Bloomberg

That’s when GE made some classic blunders. First, it panicked when markets declined and sold its risky assets when it should have hung on to them -- or bought more of them. GE allocated 80 percent of its pension portfolio to risky assets during the boom years leading up to the crisis from 2003 to 2006. The decline in the value of those assets in 2008 reduced GE’s risk allocation to 68 percent. But after the recovery in 2009, GE lost its nerve and sold some risk assets. By the end of 2010, GE’s allocation to risk was 66 percent, which is roughly where it remains today.   

All Shook Up
GE lost its taste for risk after the 2008 financial crisis
Source: Bloomberg
Note: Risk assets are company stock, equities, alternative investments, real estate and other, as reported by GE.

GE also blundered by chasing alternative investments after the crisis. From 1999 to 2008, the pension had no alternative investments. But by the end of 2009, GE had allocated 14 percent of its portfolio to alternatives.

It’s easy to see why alternatives were appealing at the time. Alternatives held up far better than the market during the crisis, in large part because of their ability to short stocks. The HFRI Fund Weighted Composite Index was down 19 percent in 2008, while the S&P 500 was down 37 percent, including dividends. Overseas stocks fared even worse.

Traumatized by the crisis and dazzled by alternatives, GE sold more of its beaten-down risk assets to make room for alternatives -- a classic case of looking in the rear-view mirror instead of the windshield. The S&P 500 has returned 18 percent annually since March 2009 through May, while the HFRI Index has returned just 6.2 percent. Overseas stocks, too, have outpaced the HFRI Index by a wide margin.

GE’s biggest blunder, however, predates the financial crisis. A critical assumption in every pension plan is the expected return from the pension’s portfolio. The higher the expected return, the less the company must contribute to its pension to meet future obligations, and vice versa.

In 1999, GE assumed that its pension portfolio would return 9.5 percent annually. At first glance, that seems like a reasonable assumption. GE’s pension portfolio is highly correlated with a 75/25 portfolio of U.S. stocks and bonds, as represented by the S&P 500 and long-term government bonds. That correlation was 0.92 between 1999 and 2016.

This 75/25 portfolio returned 9.4 percent annually from 1926 to 2016, including dividends -- the longest period for which returns are available.

But the devil is hiding in that return. It happens that the two decades before GE chose its expected return of 9.5 percent in 1999 included one of the biggest bull markets in history. From 1981 to 1998, the 75/25 portfolio returned 16 percent annually. Before that, it had returned half as much, or 8.3 percent annually, from 1926 to 1980.

Given the moment, the prudent assumption in 1999 would have been that returns would be lower over the next two decades. And that’s exactly what happened. That 75/25 portfolio returned 6.4 percent annually from 1999 to 2016, far lower than GE’s expected return of 9.5 percent. GE’s portfolio returned roughly 6 percent annually over that period.

In Step
GE's pension portfolio has been highly correlated with a simple portfolio of U.S. stocks and bonds
Source: Bloomberg
Note: The returns for GE's pension portfolio are calculated as the fair value of plan assets divided by the difference between that value and the actual return on plan assets, all as of year end.

GE has since tempered its expectations. It now assumes a 7.5 percent annual return. But that may still be too high. Long-term government bonds currently yield 2.2 percent to 2.7 percent, depending on maturity. The S&P 500’s earnings yield is roughly 4 percent, based on 10-year trailing average positive earnings. Earnings yields are higher for overseas stocks, but even so, it’s hard to cobble together an expected return of 7.5 percent.

Lower Expectations
GE has muted its expected return over time, but it's probably still too high
Source: Bloomberg

I suspect GE knows all this, which points to a bigger problem than basic arithmetic. A lower expected return would require GE to increase its pension contributions. That would strain GE’s financial condition in the near term -- and by extension its stock price. That’s not what shareholders want to hear. 

But GE has little choice. The longer it puts off the hard decisions, the costlier its pension problem will become. The market will eventually acknowledge that reality, and perhaps it already has. Since Bloomberg reported GE’s pension woes last Friday, its stock is down 3 percent through Tuesday, while the S&P 500 is up 0.2 percent.

The right answer is simple. The only question is whether GE has the stomach to acknowledge it.

This column does not necessarily reflect the opinion of Bloomberg LP and its owners.

To contact the author of this story:
Nir Kaissar in Washington at nkaissar1@bloomberg.net

To contact the editor responsible for this story:
Daniel Niemi at dniemi1@bloomberg.net