The Lost Decade

Oleg Gershkovich

Oleg Gershkovich, A.S.A., M.A.A.A.

LDI Strategist and Senior Actuary

Brett Cornwell

Brett Cornwell, CFA

Client Portfolio Manager, Fixed Income

What can corporate plan sponsors learn from the “L” shaped recovery of funded status since the Great Financial Crisis?

Lessons Learned: Don’t Be Held Hostage by Glide Path Triggers

  • Despite record-breaking returns in equity and fixed income markets and significant contributions, the funded status of S&P 500 pension plans, in aggregate, exhibited an “L” shaped recovery following the 2008 crisis.
  • The main culprit was plan sponsors’ decision to wait at the gates of their glide paths for interest rates to rise—they never did.

Avoid Yesterday’s Mistakes: Seize the Opportunity of Today

  • At the end of December 31, 2019, the funded status of pension plans was at a relative high since the 2008 crisis, approaching levels last seen during the 2013 “Taper Tantrum” when a sharp spike in rates caused funded ratios to increase, in aggregate, by 12% (from 77% to 89%).
  • Looking back to 2013, many plans missed the opportunity to de-risk, forcing sponsors to offset the resulting erosion in funded status with increased pension contributions partly facilitated by tax reform.
  • Given the first quarter COVID-19 market dislocation, plan sponsors who did not lock in their 2019 funded status gains could be back to pre-2017 levels.
  • Looking ahead, we encourage plans to seize the current opportunity. Sponsors do not need to wait for glide path triggers to de-risk—the use of treasury futures can hedge interest-rate risk without tying up growth seeking assets.
  • Rates can still go lower and at these levels, a smaller drop in rates can deteriorate funded status by the same amount as a larger drop in rates did 10 years ago (when rates were +300 basis points higher) due to convexity.
Figure 1. The “L” shaped recovery of funded status despite $498bn in contributions
Figure 1. The “L” shaped recovery of funded status despite $498bn in contributions
Funded Ratio: S&P 500 companies; aggregate pension plans

As of March 31, 2020. Source: S&P and Voya Investment Management. Information shown reflects publicly available information for the 158 of the 500 S&P 500 companies that maintained a defined benefit pension plan, have a fiscal year end of December 31, and provided consistent historical data going back to December 31, 2007.

The Last Decade, A Lost Decade

Bookended by two “once in a generation” crises, the last decade was full of unexpected surprises, including a double bull market for both equities and fixed income. However, despite record-breaking returns in equity and fixed income markets, the funded status of S&P 500 pension plans, in aggregate, exhibited an “L” shaped recovery.

To answer the question of why funded status did not keep up with soaring equity and bond returns, we analyzed the publicly available information published in the annual reports of companies in the S&P 500 as of December 31, 2019. To be included in our study, the company needed to maintain a U.S. defined benefit pension plan, have a fiscal year end of December 31, and provide consistent historical data going back to December 31, 2007. This screen resulted in 158 corporations. We aggregated their assets, liabilities and benefit payments to analyze the group as a whole. The combined group has assets and liabilities totaling $1.3tn and $1.5tn, respectively. The group experienced a 30% drop in funded status following the 2008 crisis, and as Figure 1 shows, the funded status never returned to pre-crisis levels.

Despite half a trillion dollars in contributions over the last 12 years, the group’s funded status recovered only one third of its original 30% drop immediately after the 2008 crisis.

Perhaps the most striking revelation in our analysis is the amount of pension contributions made over the decade. The group of 158 pension plans contributed $498bn since the 2008 crisis and $250bn since the Taper Tantrum. However, by the end of 2019, despite half a trillion dollars in contributions over the last 12 years, the group’s funded status recovered only one third of its original 30% drop immediately after the 2008 crisis.

Despite Market Tailwinds and Sizeable Contributions Funded Status Never Recovered—How Is That Possible?

Following the 2008 crisis, many plan sponsors, like the rest of the investment community, assumed interest rates would rise and maintained their equity exposures in an attempt to earn their way out of low funded status levels. Instead, discount rates dropped 300 basis points and liability growth, on average, outpaced stock returns despite one of the greatest equity bull markets in history.

The mismatch between liability growth and plan sponsors’ expected asset returns underscores that the view (and hope) for rising rates was implicit in sponsors’ assumptions.

While rates were declining, the expected return on asset assumption (EROA) also declined by 200 basis points—100 basis points less than the decline in rates (Figure 2). This mismatch underscores that the view (and hope) for rising rates was implicit in sponsors’ assumptions. Furthermore, the portion of the allocation that was attributed to fixed income was not necessarily duration matched to the liability, which introduced curve risk.

Figure 2. Discount rates and EROAs decline over 300 and 200 basis points, respectively, during the last 12 years
Figure 2. Discount rates and EROAs decline over 300 and 200 basis points, respectively, during the last 12 years

As of December 31, 2019. Source: 10k’s, FTSE and Voya Investment Management.

Service Cost Accruals Were a Hurdle Despite Their Gradual Decrease Over the Decade

More plans were open and accruing at the beginning of the prior decade, which also created headwinds. However, by the end of the decade, service cost as a percentage of liabilities was halved from 2.4% as of December 31, 2007 to 1.2% as of December 31, 2019 (Figure 3).

Figure 3. Service cost as a percentage of PBO
Figure 3. Service cost as a percentage of PBO

As of December 31, 2019. Source: 10k’s, Voya Investment Management. Information shown reflects publicly available information for the 158 of the 500 S&P 500 companies that maintained a defined benefit pension plan, have a fiscal year end of December 31, and provided consistent historical data going back to December 31, 2007.

While the decline in service cost (as a percentage of liabilities) alleviated some of the pressure for plan sponsors, it was not significant enough to drive the performance of funded status during the decade. Furthermore, contributions are the appropriate analog to service cost accruals, and the $498bn in contributions made by the S&P 500 plans during this time frame significantly exceeded the $268bn of service cost accruals over the same time horizon (Figure 4).

Figure 4. Contributions more than offset benefit accruals
Figure 4. Contributions more than offset benefit accruals

As of December 31, 2019. Source: 10k’s, Voya Investment Management. Information shown reflects publicly available information for the 158 of the 500 S&P 500 companies that maintained a defined benefit pension plan, have a fiscal year end of December 31, and provided consistent historical data going back to December 31, 2007.

COVID-19 Global Pandemic, Enter Stage Right

In the first quarter of 2020, the COVID-19 crisis drove the group’s aggregate funded status down to 81%, 3% shy of the 78% low at the end of 2008. However, the COVID-19 crisis only caused a 7% drop in funded status. For context, recall that during the GFC, funded status fell by 30%. The difference can be attributed to the meaningful shift to LDI and long duration fixed income that sponsors implemented in the latter half of the decade leading up to COVID-19. Unlike the “Taper Tantrum” period (2012 to 2013), when the group’s aggregate fixed income allocation went down 2 percentage points (from 37% to 35%), from 2017 to 2019 there was a 10% increase to fixed income (Figure 5).

The timing of sponsors’ shift towards fixed income proved to be fortunate. Without the increase to fixed income, funded status could have declined 9% to 12% due to the COVID-19 crisis.

Much of the shift towards fixed income in recent years was due to securing accelerated contributions that were precipitated by the 2018 tax reform. In fact, the percentage of corporations with fixed income allocation greater than 50% grew from 5% to 37% from December 31, 2007 to December 31, 2019. The timing of sponsors’ shift towards fixed income proved to be fortunate. Without the increase to fixed income, funded status could have declined between 9% and 12% due to the COVID-19 crisis.

Figure 5. Fixed income allocation increased in recent years
Figure 5. Fixed income allocation increased in recent years
Fixed income allocation (LHS) and funded status (RHS)

As of March 31, 2020. Source: S&P and Voya Investment Management. Information shown reflects publicly available information for the 158 of the 500 S&P 500 companies that maintained a defined benefit pension plan, have a fiscal year end of December 31, and provided consistent historical data going back to December 31, 2007.

What if Sponsors Could Do It All Over Again?

What if plan sponsors were given a crystal ball? We already know that they should not have waited for glide path triggers to de-risk, and an analysis of different asset allocations help reinforce this point. With further investigation, we present the base case scenario in Figure 6 showing the funded status, assets and liabilities of the pension plans in the S&P 500.

If the last decade taught us anything, it’s that sponsors do not need to wait for glide path triggers to de-risk; they can use Treasury futures to increase their interest-rate hedging immediately without encumbering assets.

To determine the efficacy of different asset allocations, we back-tested the assets using two investment strategies that are polar opposites. In one strategy, we invested 100% of the assets in U.S. equities (S&P 500 Index). In the other strategy, 100% of the assets were invested in Barclays Long Credit Index. In both cases, we adusted the results to reflect the actual historical benefit payments and contributions. As Figure 6a highlights, the 100% equity strategy underperformed the actual assets of the S&P 500 pension plans until 2013 and underperformed long credit until 2017. In fact, the equity strategy resulted in assets greater than long credit strategy in only 3 out of the last 12 years (2017, 2018 and 2019).

Over the last decade, if the 158 plans invested 100% of their assets in a long credit strategy, they would have achieved a similar level of return as investing 100% of their assets in an equity strategy—but with far less volatility.

Figure 6. Base Case: Assets, liabilities and funded status for the S&P 500 pension plans
Figure 6. Base Case: Assets, liabilities and funded status for the S&P 500 pension plans

As of March 31, 2020. Source: Standard & Poors and Voya Investment Management. Information shown reflects publicly available information for the 158 of the 500 S&P 500 companies that maintained a defined benefit pension plan, have a fiscal year end of December 31, and provided consistent historical data going back to December 31, 2007.

After 2013, there is a rapid acceleration in the performance of the equity strategy until the end of 2019, albeit not without volatility, such as 2018. Then of course, the 100% equity strategy is decimated by the COVID-19 crisis. As a result, the fixed income and equity strategies arrive at a similar asset level as of March 31, 2020. While this is not necessarily surprising given the two bookends of unprecedented “once in a generation” crises, it is also worth noting that the 100% long credit strategy acheived a similar level of asset return with far less volatility.

In the world of pension investing, assets do not exist in a vacuum so in the next stage of analysis, we review funded status percentage based on the back-tested assets (Figure 6b). Here, the differences in volatility between the two approaches are also apparent. The funded status of plans invested in the 100% long credit strategy stays within a narrower range of 26%, while the plans invested in the 100% U.S. equity strategy exhibited a range in funded status movement of 74%. Actual funded status had a maximum range of 28%.

Not hedging interest-rate risk inherently expresses a directional view on rates—plan sponsors are not compensated for this directional view.

As previously mentioned, pension contributions exceeded service accruals during the last decade, a dynamic that was a powerful component in preserving funded status levels. To highlight this impact, in Figure 6c we removed the amount of pension contributions that exceeded service costs. As Figure 6c highlights, by the end of the decade, excess pension contributions improved funded status by 25% for the long credit strategy and 35% for the equity approach. Absent these contributions, actual funded status would have exhibited a steadily decreasing trend to 50% funded (ignoring funding rules which would have forced contributions even with MAP-21 Relief).

Figure 6a. Asset Growth: U.S. equities vs. long credit (back-tested results)
Figure 6a. Asset Growth: U.S. equities vs. long credit (back-tested results)

As of March 31, 2020. Source: Standard & Poors and Voya Investment Management. Information shown reflects publicly available information for the 158 of the 500 S&P 500 companies that maintained a defined benefit pension plan, have a fiscal year end of December 31, and provided consistent historical data going back to December 31, 2007.

Figure 6b. Funded Status Volatility: U.S. equities vs. Long Credit (back-tested results)
Figure 6b. Funded Status Volatility: U.S. equities vs. Long Credit (back-tested results)

As of March 31, 2020. Source: Standard & Poors and Voya Investment Management. Information shown reflects publicly available information for the 158 of the 500 S&P 500 companies that maintained a defined benefit pension plan, have a fiscal year end of December 31, and provided consistent historical data going back to December 31, 2007.

Figure 6c. Contributions Save the Day/Decade: Funded status plummets without contributions
Figure 6c. Contributions Save the Day/Decade: Funded status plummets without contributions

As of March 31, 2020. Source: Standard & Poors and Voya Investment Management. Information shown reflects publicly available information for the 158 of the 500 S&P 500 companies that maintained a defined benefit pension plan, have a fiscal year end of December 31, and provided consistent historical data going back to December 31, 2007.

The Future Is Not What It Used to Be

Key takeaways from our analysis of the last 12 years:

  • Rates did not rise as sponsors had hoped, a trend that created a significant mismatch between liability expectations and actual liabilities (liabilities ended the period ~1.5x bigger than 2008).
  • Contributions in excess of service cost accruals were $231bn. To put this in perspective, this is almost 25% of the initial $1tn asset value of the 158 pension plans included in this analysis.
  • Setting an EROA does not guarantee the actual return experience of plan assets. Had the S&P 500 group of plan sponsors realized their expected asset return, plans would have been $128bn better funded.
Figure 7. Unexpected losses due to missed forecasts totaled $824bn
Figure 7. Unexpected losses due to missed forecasts totaled $824bn

As of December 31, 2019. Source: Standard & Poors and Voya Investment Management. Information shown reflects publicly available information for the 158 of the 500 S&P 500 companies that maintained a defined benefit pension plan, have a fiscal year end of December 31, and provided consistent historical data going back to December 31, 2007.

Head for the Exits: Buyouts Are Born and Momentum Expected to Continue

While the decade may have been “lost” in terms of funded status improvement, for a growing number of corporations there has been a dramatic shift to the ultimate de-risking: transferring pension obligation to insurance companies.

The lack of interest-rate hedging and delayed rotation into fixed income, despite significant contributions, postponed pension risk transfer solutions for many plan sponsors who were keen to reduce the pension plan liability once their funded status improved. However, many transactions were completed. This is rapidly evolving into a mature market since the watershed moment in the fourth quarter of 2012 when both General Motors and Verizon annuitized their defined benefit plan obligations (~$35bn combined in plan assets). There has been a lot of momentum here, as shown in Figure 8, with over $160bn of liabilities novated to insurers since 2012 and more deals are likely on the way. During 2018, FedEx delivered $6bn of pension annuities to MetLife. Pension buyout activity was strong during 2019, with $28bn in annuities transferred to insurance companies – that’s 5% more than 2018.

Included in the 2019 transactions was Lockheed Martin Corp., which completed a $1.9bn retiree carve out in the fourth quarter. Another notable transaction during 2019 was Bristol-Myers Squibb Co.’s complete termination of its $3.8bn pension plan with a full transfer to Athene Annuity and Life Co. in the third quarter. While we may see similar full terminations in the coming years, we expect most buyouts to be predominantly retiree-only transactions.

For those plan sponsors who were able to complete a retiree-only transaction, it is now more critical than ever to carefully manage the remaining plan, which typically has a longer duration and a complicated participant profile. By carefully managing the remaining plan and not repeating the mistakes of the prior decade, plan sponsors are better prepared to complete more transactions in the years to come.

Figure 8. Annuitizations continue unabated
Figure 8. Annuitizations continue unabated

Source: Secure Retirement Institute, Fourth Quarter 2019 U.S. Group Annuity Risk Transfer Study.

Looking Ahead: Even in a Post COVID-19 World, It’s Always a Good Time for Meaningful De-risking

Pension plans are not hedge funds. Their objective function (when sufficiently funded) should be geared towards solvency and liquidity—not maximizing returns. In fact, qualified plans in the U.S. operate as nonprofit insurance subsidiaries since an overfunded position is subject to an excise tax (assessed by the IRS to limit profits from pension fund raids that were rampant during the 1980s), while underfunded plans incur onerous PBGC premiums and mandated contributions. In short, there is an asymmetric payoff when it comes to taking equity risk. Plans receive limited upside when equities rise and incur all of the downside when equities underperform.

There is an asymmetric payoff when it comes to taking equity risk in pension plans. Plans receive limited upside when equities rise and incur all of the downside when equities underperform.

Turning our attention to fixed income allocation, not hedging interest-rate risk, inherently expresses a directional view on rates for which plan sponsors are not compensated. While contributions were pouring in over the last decade, plan sponsors were slow to de-risk as they waited for rates to rise. Instead, rates dropped 300bps during that time. The truth is sponsors do not need to wait for glide path triggers to de-risk; they can use Treasury futures to increase their interest-rate hedging immediately without encumbering assets.

The prior decade and the recent COVID-19 market dislocation is a reminder that reliance on equities and ignoring interest-rate risk comes at its own peril. Interest-rate risk can be painful and overpowering even during times in which stocks and bonds rise. Moreover, fixed income garnered a better risk-adjusted return even during the last 20 years. While this is no guarantee of future performance, it is worth noting that from January 2000 through April 2020, annualized returns for the S&P 500 were 5.5%, which trailed the Barclays Long Credit return of 7.7%. Furthermore, Barclays Long Credit exhibited a lower volatility of 9.3% versus 15.0% for the S&P 500.

Why take chances with equities when fixed income is a better match to liabilities? If sponsors fear that bonds are overpriced, consider that a rising rate environment typically results in less borrowing and less spending, which leads to lower earnings and ultimately lower stock valuations.

Clearly, difficult allocation decisions need to be made. In the interim, using Treasury futures is a simple step sponsors can take to meaningfully reduce interest-rate risk without forsaking allocations to growth assets. This simple strategy reduces sponsors’ reliance on the increasingly elusive glide path trigger and enables them the opportunity to conduct a liability aware review of their strategic asset allocation.

Disclaimer

Voya Investment Management has prepared this commentary for informational purposes. Nothing contained herein should be construed as (i) an offer to sell or solicitation of an offer to buy any security or (ii) a recommendation as to the advisability of investing in, purchasing or selling any security. Any opinions expressed herein reflect our judgment and are subject to change. Certain of the statements contained herein are statements of future expectations and other forward-looking statements that are based on management’s current views and assumptions, and involve known and unknown risks and uncertainties that could cause actual results, performance or events to differ materially from those expressed or implied in such statements. Actual results, performance or events may differ materially from those in such statements due to, without limitation, (1) general economic conditions, (2) performance of financial markets, (3) interest-rate levels, (4) increasing levels of loan defaults, (5) changes in laws and regulations, and (6) changes in the policies of governments and/or regulatory authorities.