Tag Archives: Pension Finance

Discount rates and pension plans: 98 percent of economists agree….

I realize that 98 percent sounds impossibly unified around any subject. But I find this recent survey by the University of Chicago’s IGM Economics Experts panel especially compelling. According to their survey of 39 professional economists, 98 percent agree that public sector plans understate pension liabilities and costs by using high discount rates.

Here’s the question as stated:

Question A: By discounting pension liabilities at high interest rates under government accounting standards, many U.S. state and local governments understate their pension liabilities and the costs of providing pensions to public-sector workers.

The response:

49 percent strongly agree

49 percent agree

Who was surveyed?

39 economists – from across the field –  including Richard Thaler (Chicago), Jose Scheinkman (Princeton), Robert Hall (Stanford), Austen Goolsbee (Chicago, and former advisor to President Obama), Barry Eichengreen (Berkeley), Claudia Goldin (Harvard), Alberto Alesina (Harvard) and Daron Acemoglu (MIT).

You can check out who was surveyed and their academic credentials at the site.

This principle concerning the valuation of pension liabilities is not very controversial (or even interesting) for economists as M. Barton Waring has noted in, Pension Finance. It only remains controversial among actuaries and policymakers/pension plan analysts and advisors in this one corner of the world: U.S. public sector pension plans. It is partly a matter of professional training. Economists and actuaries are using different toolkits to evaluate plans. (There are notable exceptions, see, Gold and Bader). It is also a matter of the implications of what happens when governments use discount rates more in line with the guaranteed nature of public plans. Lowering discount rates increases the necessary contribution.

But if governments are serious about offering these plans as guaranteed to retirees, then they should be especially interested in valuing them accurately.



Pension Reform in Pennsylvania and a book recommendation

Last week I testified before the Pennsylvania General Assembly on their pension reform efforts. In my testimony I covered the valuation problem. While the state reports an unfunded liability of $39.5 billion. I calculate it is $116 billion. Like many other state and local pension systems, the mis-valuation problem caused Pennsylvania lawmakers to undertake a series of measures over the past several decades that further weakened the system. These include benefit enhancements when the market was booming and creating a “funding collar” which capped annual contributions to the pension plan effectively pushing that bill forward. to the present

Pennsylvania has run out of time and costs are rising rapidly. In my testimony I cite the work of M. Barton Waring, whose book, Pension Finance, I highly recommend as an analysis of why economic valuation matters to the very existence of defined benefit plans. Without it, plans are subject to often questionable and arbitrary accounting choices. His analysis is well-grounded, consistent with theory and cogently explained.

He states his findings as a series of propositions that should guide how defined benefit plans are structured, valued and funded.

Proposition 1: Measures of the pension plan based on conventional accounting methods will always follow measures based on economic accounting sooner or later, even with a lag. The accounting will follow the economics sooner or later.

I think that proposition can been seen in the funding schedule for Pennsylvania’s two main pension plans: SERS and PSERS. As a result of artificially capping the state’s annual contribution to the plan, future contributions are slated to increase by 267 percent in the next five years. (And that is an underestimation since it is working off of the misvalued liability). Lawmakers know there is a serious hole ahead and as Waring’s book shows that is something that could have been avoided if plans had used economic valuation from the start.