P/Es and Pension Funding Ratios

Abstract

Some evidence supports the intriguing conjecture that P/Es in the U.S. market may decline in times of both significantly lower, as well as significantly higher, real interest rates. The P/E response pattern would then resemble a tent that angles downward at both ends. For pension liabilities defined in real terms, very low real rates would then lead to a clifflike falloff in the funding ratio from the decline in equity valuations combined with surging liability costs. This article explores the risk implications of such a low-rate scenario and the equity valuations that could give rise to such a tent pattern.

A number of historical studies have exhibited generally declining P/Es for U.S. stocks as a function of higher nominal interest rates. A different perspective emerges when real (inflation-adjusted) rates are substituted for nominal rates. We provide a histogram based on monthly samples for the 1978–2004 period to study the P/E pattern related to real rates. The P/Es were computed from 12-month forward earnings projections and the real rates were derived by subtracting concomitant inflation from 10-year U.S. Treasury rates. A tentlike pattern emerges in which, in contrast to the typical decreasing pattern of P/Es with nominal interest rates, P/Es decline in times of both significantly lower, as well as significantly higher, real interest rates. When real rates are used, the highest P/Es lie within a “sweet spot” of 2–3 percent and then fall off for both higher and lower rate levels.

In this article, with due recognition of the many empirical limitations, we proceed on an admittedly conjectural basis to explore the potential implications such a tent pattern has for the asset/liability management of pension funds and for some aspects of asset valuation in general.

A potential explanation for the tent pattern relates to the interaction of real rates and growth prospects. For real rates in the 2–3 percent sweet spot, economic and profit growth can be presumed to be reasonably normal, reflecting a comfortable balance between funding needs and capital availability. As growth and demand for funds push real rates beyond 3 percent, however, valuations begin to be impaired by the increasing cost of funds. This reasoning would explain the P/E decline with higher interest rates—whether nominal or real.

For the infrequent events when real interest rates fall significantly below 2 percent, a feasible explanation for the lower P/Es is more challenging. Circumstances can certainly be envisioned, however, in which low real rates are associated with poor economic conditions and dour prospects for future profit growth. In such a situation, money may be readily available but P/E valuations could still be severely depressed by a mixture of dismal growth prospects, elevated risk prospects, and perhaps, a reduced tolerance for market risk. Although such a hypothetical situation is grim, these conditions are not unknown within the recent history of global markets.

A P/E tent pattern would have particularly intriguing implications for the asset/liability management of defined-benefit (DB) pension funds. Consider a pension fund that can be simplistically modeled as having 12-year duration relative to real rates. Now, suppose that, with real rates at 3 percent, this liability is 100 percent funded with a portfolio consisting of 60 percent equity and 40 percent 5-year-duration bonds. Suppose also that the tent pattern is a snapshot of the movement of equity prices as real rates change.

On the one hand, as real rates move to levels higher than the 2–3 percent sweet spot, the 60 percent equity component of the asset portfolio would decline (in accordance with the right side of the tent). The 40 percent bond component would also decline, so the total portfolio would be under considerable stress. On the other hand, at rates lower than the 2–3 percent sweet spot, the equity component would again decline but there would be some offset from increasing bond values. The net effect would be a somewhat flat tent.

In contrast to the asset/liability offset seen under rising rates, the scenario of falling rates would lead to a severe deterioration in the funding ratio driven by a “perfect storm” of asset deterioration and rising liability values. Thus, the overall pattern for the funding ratio would be a surprisingly high degree of stability at higher interest rates but a horrendous falling off at lower rates.

The key point is that the “left-hand scenario” below the sweet spot may represent the ultimate “black hole” for investors subject to some form of long-term liability. Such an environment, especially if persistent over time, would exert a gravitational pull to produce a host of adverse events—low-growth prospects, reduced risk tolerance, a move to lower-risk assets, increased risk premiums, adverse correlations across alternative asset classes, reduced annuity value per investable dollar, deteriorated funding ratios, a pullback of automatic-rebalancing strategies (together with a loss of volatility smoothing), and so on. That is the bad news.

The good news is that such events are rare and that, for the most part, real interest rates and the associated market conditions are localized within or near the sweet spot. In addition, when events push toward the extreme right or left sides of the tent diagram, natural counterbalancing forces of recovery and/or societal intervention act to speed a return back toward the sweet spot. In particular, the right-hand scenario, with its excessively high cost of capital, is inherently self-correcting. The more troublesome left-hand scenario is apparently extremely rare and rarely persistent.

Notes

¹ Private communication at a Q-Group meeting in October 2006.

² As pointed out by Waring and Siegel (2007) in this issue, even pension funds that pay only nominal benefits in retirement may have a real-rate component in their liabilities because of the correlation between wage growth and systemic inflation.

³ See, however, Leibowitz and Hammond (2004).