Piling on the Pension Pain

Dan Borenstein of the Contra Costa Times has done a lot of great reporting on public pension abuses, but it looks like the beat has finally driven him over the edge. How else to explain his Sunday column, in which he says that, to protect our children from pension risks, we must spend less on their schools?

Of course, he doesn’t quite say that in so many words. The column is an attack on the assumed rate of return used by the state’s pension funds. Borenstein charges that the assumed rates are, first, “speculative” and, second, “overly optimistic.”

The first complaint is true but hardly the knock on CalPERS he supposes it to be. When it comes to knowing where the S&P 500 and bond prices will be 30 years from today, we are all speculating. The future movement of financial markets is unknown and unknowable. The best any long-term investor can do is make reasonable assumptions that draw on past performance and extrapolate from current conditions. Which is exactly what the pension funds do.

Borenstein’s second complaint contradicts his first. If the future is unknowable, how can he say that the 7.5 percent assumed rate of return used by CalPERS is “overly optimistic?” Might it just as likely be pessimistic or right on target?

Like many pundits, he seems shaken by the havoc that struck financial markets in the Great Recession. But as the economist Dean Baker points out, three years of returns don’t tell you much about 30 years, the horizon for retirement investing. In 1934, five years into the Great Depression, after the stock market had cratered and long bond yields had fallen to levels people had never experienced, many of them decided they would never again see the investment returns of the past. Yet the next 30 years worked out pretty well for anybody who didn’t put money into a mattress.

Borenstein’s assertion that the assumed return is “overly optimistic” doesn’t have any particular grounding in history or arithmetic. It’s a guess that, if taken seriously, implies decades of economic stagnation around the world. The right time to complain about “overly optimistic” assumptions would have been in 2000 when, with stocks priced at dizzying multiples to earnings, California boosted pensions on the promise that investment returns would pay the cost of the higher benefits.

But let’s assume Borenstein’s right and do what he proposes: set the assumed rate of return at “more conservative current bond yields.” Current bond yields are about 1.5 percent for 10-year Treasuries, and 2.5 percent for 30-year bonds. What would happen?

As Borenstein notes, “The higher the rate, the less employers and employees must contribute now to fund future retirement benefits. But if projections don’t pan out, the shortfall must be paid off solely by future taxpayers.” The reverse is also true. The lower the assumed discount rate, the more current taxpayers will have to pay (save) and the more future taxpayers will receive in a windfall if actual investment returns prove better than assumed.

To better understand how the math works, let’s think about the decisions individuals must make in doing their own retirement planning.

Let’s imagine a 35-year-old male worker earning $50,000 a year, about double the median wage. Our worker reads up on investment advice and decides that assuming a long-term compound return of 7 percent a year—well within the assumptions used by college endowments, private pension funds, hedge funds, managers of assets for charities and other institutions, and also in line with the historical performance of diversified portfolios—is reasonable. He also assumes a 3 percent rate of inflation, which his paycheck will match. He does the math and finds that by putting aside 10 percent of his paycheck in a retirement plan (well above the current average 401(k) contribution of 8.2 percent) he can reach his retirement goal: to accumulate sufficient assets by age 65 to purchase an annuity that, combined with Social Security, will provide him with about 90 percent of his final income, enough to maintain his middle-class standard of living in retirement.

This decision comes with risks and uncertainties, of course. Investment returns might fail to reach his assumed level, leaving him with less income in retirement. He may hit his retirement date at a moment when annuities are expensive and his accumulated assets won’t buy as much retirement income as he had hoped. On the other hand, he may see his paycheck rise higher than he’d expected, giving him greater retirement savings.

But suppose our worker reads Borenstein, who tells him that it’s “crazy” and “overly optimistic” to assume a 7 percent return. According to Borenstein, the sane thing for him to do would be to base his retirement saving on current bond yields. Under that assumption (let’s make it 2 percent for the sake of simplicity), putting aside 10 percent of his earnings won’t cut it. At age 65 our worker’s assets would replace only 17 percent of his final salary, not enough for retirement.

In other words, if he believes Borenstein, he will need to save much, much more. To reach the same level of monthly income in retirement as he had previously assumed, he would need to put aside 34 percent of his paycheck each month. While he’s working he’ll have so little disposable income he’ll never be able to buy a house, send a kid to college, or take a vacation. A comfortable retirement will be assured, but he won’t have much of a life.

The same would be true if California took Borenstein’s advice to use a risk-free rate of return to set pension contributions. It would eliminate any risk that the next generation would have to pick up some of the cost of today’s services. But it would make life today more miserable.

What’s been lost in the increasingly politicized discussion about assumed rate of return is that the assumption doesn’t actually affect the costs of pensions.

The pension promises to current workers have been made. They are what they are, unless courts change their rulings about vested contractual rights. And the investment returns of the pension funds will be what they turn out to be. The prison guard who has been promised 90 percent of his highest year of pay at age 50 for the rest of his life will get the same check each month whether CalPERS earns 2 percent or 7.5 percent.

The assumed rate of return affects only the timing of who pays the bill for that pension. A state is not like a private company, which is at risk of folding and leaving behind an inadequately funded pension plan. California and its taxpayers are going to be here to pay their promises for many decades to come. If the assumed rate subsequently proves to have been too optimistic, some of today’s pension cost for that prison guard will be pushed into the future; if it’s too pessimistic, today’s taxpayers will end up paying more than they should. If it’s set as low as Borenstein wants, required pension contributions would triple.

And who would that hurt most? Since a majority of state spending goes to support education and health care for young people, the biggest losers would be the next generation. Convinced that he can read the future path of markets, Borenstein is implicitly telling children that, “in order to protect your adult pocketbooks from any risk of paying part of the costs of the pensions of your teachers or professors, we must put more current dollars into pension funds and fewer into your schools and colleges.”

Now that’s crazy.