The Highway Bill “Pension Gimmick:” A Primer
July 15, 2014
The United States Highway Trust Fund is set to run out of money in less than a month. Congress has been working to address this problem; some long-term solutions have been proposed, but the House and the Senate appear to be converging on a short-term fix to the problem.
Both the Senate and House versions of the bill raise revenue through a provision labeled as Pension Fund Stabilization. The meat of this proposal is a change in how employers are required to value their pension liabilities.
In effect, “pension smoothing” limits the amount of contributions a company makes into its pension in the near term. This increases a company’s taxable income in the near term, which increases short term tax revenue.
This is one of the more confusing things Congress does. This post is intended to explain where this provision comes from, what its effects are, and why on earth it ended up in a highway bill.
What Does The Provision Do?
The provision changes the way that single-employer pension plans value their liabilities – that is, the worth of all of the benefits an employer has promised to its workers. Specifically, this proposal makes changes to the discount rates that they use when reporting their liabilities to the IRS.
Discount rates are the way people value payments in the future. For example, a pension fund might have promised a benefit payment of $1 a year from now. That isn’t worth as much as $1 immediately. So the pension fund might use a discount rate of 4 percent per year, and value that liability at 96 cents instead. With a discount rate of 6 percent, the liability would be worth only 94 cents. In other words, the higher your discount rate, the lower your valuation of the future liability.
Discount rates for single-employer pension funds are determined by the interest rates on an index of corporate bonds. Since interest rates can change over time, the pension funds use three different rates to discount liabilities: one for the next five years, one for the fifteen years after that, and one for all years after the first twenty.
The pension smoothing provision included in the highway bills effectively raises these discount rates, lowering the measured value of future liabilities.
Where Does This Idea Come From?
The rule extends the sunset of an older policy that was designed to gradually phase out. The policy was first included in the Moving Ahead for Progress in the 21st Century Act (MAP-21) – also a transportation bill – in 2012.
The unique conditions of the 2008 recession created problems for pensions. Interest rates of all kinds collapsed – and therefore, the IRS discount rates began to fall as well. The following chart shows the three discount rates for single employer plans (known as “segment rates”) falling over time.
Falling discount rates meant that liabilities rose – as established above, the less you can discount your future liabilities, the more they cost. The higher valuation of liabilities required employers to significantly step up their pension contributions to match those liabilities. That was problematic in a time when those employers were most struggling to balance their budgets.
To offer employers some relief from the requirements, MAP-21 mandates that the IRS-required discount rates stay in a “corridor” close to their 25-year average. For example, under MAP-21, annual plan filings for 2012 used rates no more than 15 percent different from the 25-year average. Since the 25-year average was considerably higher than the low current market rates, plans used discount rates of 85 percent of the 25-year average instead of the amounts in the chart above.
The MAP-21 corridor was intended to widen until it became of limited relevance, slowly phasing out:
- 90 percent to 110 percent for 2012
- 85 percent to 115 percent for 2013
- 80 percent to 120 percent for 2014
- 75 percent to 125 percent for 2015
- 70 percent to 130 percent for 2016 or later
In other words, the bill gave pensions a short-term adjustment period to slowly become accustomed to lower discount rates (and the corresponding higher valuation of liabilities.) But the bill also seemed to endorse the idea that over the long term, discount rates should reflect something more like market rates of interest.
The provision in this month’s highway bill would simply extend the timeline by three years, adding to the “short term” and delaying the “long term” to 2019.
How Does The Provision Raise Revenue?
The short answer is that it doesn’t. The provision gets a positive revenue estimate within the ten-year budget window, but not overall. Here is the estimate of the Senate bill’s revenue from the pension change, from the Joint Committee on Taxation (JCT):
The JCT’s estimate has a clear pattern to it: it raises revenues in the short term, but that positive revenue becomes negative a few years down the line. If the graph were extended a few more years, the years after 2024 would also have negative revenue effects. But those years fall outside of the ten-year budget window for scoring bills, so they aren’t counted. The essence of the trick is that all of the positive years are counted, but not all of the negative years.
Why do the Revenues Follow That Pattern?
A pension plan measures its assets and liabilities each year. If its assets are less than the present value of its liabilities, then it has to contribute extra money to make up for the shortfall, in order to make sure it has enough to pay its benefits.
With the higher interest rates allowed from the interest rate stabilization, the liabilities are determined to be lower than they would be without the interest rate stabilization. Because of this, the pension funds end up contributing less money in the short term – but more money in the long term.
Pension contributions are tax deductible, and the Joint Committee on Taxation assumes that any money not contributed to pension funds will be retained as taxable corporate profits. So when we have fewer pension contributions, we have more taxes paid, and when we have more pension contributions later, we have fewer taxes paid.
Because of the glacial pace of pension funding, many of the contributions that make up for the previous lack of funding fall outside of the ten-year budget window. That means that the corresponding tax deduction also falls outside of the window, giving the provision a positive revenue “score,” when it is in fact approximately budget-neutral.
This explains how interest rate stabilization can be used as a gimmick. If a congressman has a policy idea that costs more money than it brings in, he can propose to “pay for it” with the pension smoothing – which appears revenue-positive, but is actually only revenue-neutral. This budgeting trick has now appeared in several tax proposals.
When Have We Seen This Proposal Before?
This gimmick has showed up in at least three major attempts at legislation recently. So far, none of them has passed. Each time it appeared, it was condemned by a variety of policy analysts across the political spectrum.
In October 2013, lawmakers tried to use it as an offset for the repeal of the medical device tax. Chye-Ching Huang of the Center for Budget and Policy Priorities wrote a piece identifying it as a “timing gimmick.” She wrote that the proposal was not budget neutral: “If coupled with repeal of the medical device tax, such a package would raise deficits by billions of dollars each year in the long run.”
In early 2014, lawmakers tried to use it as an offset for the extension of unemployment insurance. The Committee for a Responsible Federal Budget wrote that lawmakers “should avoid budget gimmicks like pension smoothing that only appear to save money, while actually driving up our deficits.” Romina Boccia at the Heritage Foundation also took issue with claims by Senator Jack Reed that the proposal would reduce the deficit, calling the claim “bogus.”
And now, in mid-2014, the proposal is being used to fund highways. At the New York Times, Josh Barro observed that this agreement “only goes to show that politicians in both parties have settled on an insane definition of ‘fiscal responsibility.’”
At the Tax Policy Center, Professor Len Burman had even harsher words: “THIS $6.4 BILLION REVENUE PROVISION RAISES NO REVENUE OVER THE LONG RUN!!!…” He continued: “There is a sad symmetry to this gimmick. Congress is basically allowing companies to do what our current legislators do—postpone funding their financial obligations.”
Is it Good Policy?
The merits of different discount rates are a debate for actuaries – a lively debate, even, to the extent that actuaries can be lively. Professionals have differing opinions on how the pension funds should account for future liabilities. Only recently, in the Pension Protection Act of 2006, did the segment rates come into being. In other areas of pension law, different rates are used – for example, the expected return on the assets as determined by a financial advisor. We should expect that debate to continue among pension professionals.
However, it is abundantly clear that this proposal is no longer being used with the best interests of pensioners in mind. The merits of interest rate corridors should be left to people who actually care about interest rate corridors, not people who want to make unrelated policy ideas appear budget-neutral.
Furthermore, using pensions to “pay for” a brief Highway Trust Fund reprieve is a particularly poor policy idea, for some additional reasons. It violates the user-pays principle, which implies that the Highway Trust Fund should come primarily from highway users, not unrelated provisions. There is no meaningful way that pension policy should be tied to highway policy. It also is part of a general worrying trend of short-term patches from Congress, and extensions of “temporary” provisions that slowly become permanent.
If this unwise patch makes it through Congress in place of a permanent solution, the interest rate corridors will live another three years, and the Highway Fund will live another few months, modestly delaying an inevitable recurrence of the same issues. So it goes.