In this article we consider the effects of defined benefit plan liability settlement (via a lump sum or annuity distribution), mark-to-market DB accounting, and PBGC premium reduction on corporate earnings, and how the one-time reduction in corporate earnings that often results from a settlement should be viewed.
We begin with a brief review of DB accounting. We then discuss the decade-long move of some firms to mark-to-market pension accounting to reduce earnings drag, their related move to an LDI pension asset strategy to reduce future earnings volatility, and the implications of these decisions for plan settlement/de-risking/PBGC premium reduction strategies. We conclude with a discussion of a mirror of that process – that firms that have converted their asset portfolio to an LDI strategy may now be in a position to undertake plan settlement/de-risking/PBGC premium reduction strategies and mark-to-market accounting with little or no long-run earnings implications.
Accounting for pensions
Generally, the cost of a defined benefit plan benefit – which, by its nature, generally will not be paid until sometime in the future – is reflected on the sponsor’s income statement based on a set of assumptions (most critically, interest rates, return on trust assets, and expected mortality). To the extent that experience differs from these assumptions, an “experience” loss/gain results. Under standard accounting practice, these losses/gains are generally only recognized to the extent that they exceed a “corridor” equal to 10% of the greater of the plan’s projected benefit obligation (PBO) or plan assets and, to the extent they exceed the corridor, are then amortized over the expected future working lifetime of active participants.
One reason for this treatment is that the smoothing of performance that results eliminates market volatility (“noise”). In an equilibrium state, e.g., losses from interest rate declines will be cancelled out by gains from interest rate increases. But in present conditions – a 40-year decline in interest rates – this approach has for many firms simply built up losses that will, ultimately (because, ultimately, expense must equal cash cost), have to be recognized at some point as a charge against income. Thus, among other things, these accumulated pension losses represent a drag on future earnings.
Sponsors may use an alternative to this method of recognizing experience gains/losses, “marking to market” plan performance each year. Under this approach, when interest rates go down, the experience loss is recognized as a charge against current year income. If, in a subsequent year, interest rates go up, the gain is also currently recognized.
Some sponsors have, over the last decade, elected the mark-to-market (MTM) approach, generally to reduce the future earnings drag.
Accounting and settlement
We have written a number of articles on the utility of reducing DB participant headcount, settling plan liabilities through the payment of lump sums or the purchase of annuities, and thereby reducing (in many cases, significantly) Pension Benefit Guaranty Corporation premiums.
One impediment, for some sponsors, to settling plan liabilities is that, as a necessary element of the settlement process, a share of unrecognized losses based on the proportion of total liabilities settled must be currently recognized in income, generating (sometimes significant) losses.
With respect to this issue, two observations: As indicated above, for many sponsors, recognition of these pension-related losses has become a feature, not a bug, as they represent a drag on future earnings. And, quite aside from a firm’s preference for a “corridor” vs. MTM approach to recognition, the reduction of PBGC premiums represents a net gain to the firm.
Reducing PBGC premiums is necessarily positive
The latter point needs to be emphasized. Generally, the different accounting approaches simply determine when long-run interest rate losses will be recognized. Which approach is used does not change the net effect on long-run corporate income. This is why the market has (generally) reacted positively to the adoption of MTM pension accounting.
It can be argued that, if interest rates “go back up,” then a shift to MTM accounting will have had the result of introducing unnecessary volatility (noise) into the income statement. It could also, however, be argued that, for many companies, it is no longer realistic to believe that interest rates will go up so much that accumulated pension-related interest rate losses will be wiped out.
But – whatever a firm’s view on the merits of these different pension accounting methods – PBGC premiums function more or less as a tax: a net loss to the firm. Reducing them will necessarily improve corporate income (over time).
Reducing volatility under MTM vs. future interest rate increases
Some firms that convert to MTM also reallocate their asset portfolio to adopt an LDI (liability driven investment) strategy – e.g., investing in duration-matched fixed income securities. This approach will (obviously) reduce future earning volatility, as, e.g., in the context of a decline in interest rates, asset gains offset losses resulting from increases in liability valuation.
This feature of MTM strategy does present a “real money” risk: If the firm has shifted to an LDI portfolio and interest rates subsequently go up, then the firm has experienced an “opportunity” loss. That is, e.g., if the firm had waited it could have recognized an experience gain with respect to its liabilities associated with the increase in interest rates without the offsetting loss on an LDI portfolio.
This point simply underlines the fact that any decision to adopt an LDI strategy must be made in the context of the firm’s overall view as to the trajectory of future interest rates and the performance of alternative portfolio strategies (if the firm believes it can come to such a view with reasonable confidence).
Reversing the causal arrow: LDI and settlement leading to adoption of MTM
To summarize what we have said so far: some firms have adopted MTM to reduce the drag on future earnings represented by accumulated, unrecognized interest rate-related pension losses. That decision simply changes the timing of recognition. Let’s call that “Step 1.” In connection with the decision to go to MTM, however, the firm may also shift to an LDI strategy, to reduce future earnings volatility. Let’s call that “Step 2.” Unlike Step 1, Step 2 does present a real (as opposed to timing-only) risk: if interest rates go up, then the firm experiences what we’re describing as an “opportunity” loss – but for the shift to LDI, the plan would have made money (e.g., in the form of a reduced liability valuation) on the increase in interest rates.
Many firms, today, however, have already taken Step 2 (shifting to an LDI portfolio allocation) without having taken Step 1 (moving to MTM). Thus, while they have had to confront the “real money” risk (that interest might go up), they have not had to confront the recognition timing issue.
In this situation, settlement (distribution of a lump sum or annuity liquidating the plan’s liability to a participant) – which in some respects might be thought of as the ultimate liability driven investment – forces the recognition issue and may lead to a move to MTM accounting more generally. And (to some extent repeating what we said above) that decision to settle is supported by the following points:
The change to MTM only affects timing – it does not affect the long-run real cost of the plan.
MTM accounting (and the associated current recognition of accumulated interest rate losses) may be viewed by the market as a feature, not a bug.
The real risk associated with a move to an LDI strategy has already been taken.
The gain from settlement – a reduction in PBGC premiums (associated with the reduction in headcount) – is a real gain to the firm.
In this context, and, as we said, reversing the causal arrow, firms considering settlement of some plan liabilities may also want to consider a move to MTM accounting.
* * *
We will continue to follow this issue.