I’m talking about a largely overlooked side effect of the Federal Reserve’s Operation Twist, its program to stimulate the economy by reducing long-term interest rates. Assuming that those rates actually go down and stay down, one of the unintended consequences will be to increase the stated cost of pension obligations, an insight for which I have to thank Jack Ciesielski of the Analyst’s Accounting Observer, a research service for investors.

You remember pensions, right? That benefit that increasingly few American workers have? An employer promises today to make payments to retirees, present and future, for the rest of their lives.

You determine today’s cost of tomorrow’s pension by using what’s known as the discount rate. You figure out what you’ll pay in the future, then discount it to figure out how much the obligation amounts to in current terms.

Let’s say you’ve promised to pay someone $10,000 25 years from now, and you use a discount rate of 5.4%. Today’s cost of the obligation is $2,685. But let’s say that instead, you used a discount rate of only 3.88%. Today’s obligation is $3,861, which is more than 40% higher.

I didn’t pick those numbers out of a hat — who could? I got them from Ciesielski, who says that at year-end 2010, the 271 S&P 500 companies that still have pension programs and report on a calendar-year basis used a median discount rate of 5.4% to calculate the obligations’ present value. This means that half the companies used a higher rate than that, and half used a lower rate.

An aside: I’m not going to deal with the impact that a lower discount rate would have on already hard-pressed state and local government pension funds. Corporate pension numbers have at least some basis in reality, but many government pensions are in la-la land.  Their discount rates are a subject for another day, or perhaps never.

Back to the main event. Between year-end 2010 and Sept. 30, the key long-term rate — the yield on 30-year Treasury bonds — fell by 1.52% (to 2.92% from 4.44%). Subtract that 1.52% decline from the year-end 5.4% discount rate, and you get 3.88%. That’s how I got those numbers.

Pensions are much more complicated than my simple example, and consist of a complex mix of obligations, not just one big amount due in 25 years. Ciesielski estimates that lowering the discount rate by 1.52% would increase the cost of the companies’ pension obligations by about $228 billion, to $1.582 trillion compared with the year-end figure of $1.354 trillion.

And remember that Operation Twist had barely gotten underway by Sept. 30. If it pushes down long-term Treasury rates even more, the difference at year-end 2011 will be more than 1.52%.

Ciesielski, who’s lived through more than a few up-and-down rate cycles, doesn’t expect companies to cut their discount rates by anything resembling the full drop in the 30-year Treasury rate. “Discount rates are very sticky when long rates are on their way down,” he quips, but when rates rise, which decreases the stated cost of future obligations, “they’re well-lubricated.”

But even if discount rates drop by less than 1.52%, it would increase companies’ stated obligations by a substantial amount, affecting their balance sheets and possibly their credit ratings. I suspect that many companies will tell auditors that the Fed’s action is out of the ordinary, and try to diminish the role that long Treasury bonds play in calculating pension discount rates.

None of this is to pass judgment on Operation Twist. Even though I’m a bit dubious about what the Fed is doing, just as I was dubious about so-called Quantitative Easing 2, the Fed, unlike much of the rest of the government, is trying to do something to stimulate the economy rather than standing on the sidelines, wringing its hands.

But Operation Twist’s impact on the stated cost of pension obligations shows that the world is often much more complicated than it seems to be. And that clouds with silver linings are still clouds.