Over the past few years, the average funded status of Taft-Hartley pension plans has improved significantly, reaching the highest levels since the 2008 financial crisis. While some of this improvement stems from the Special Financial Assistance Program (SFA) under the American Rescue Plan Act (ARPA), strong equity market performance has also played a role.

With plans in stronger funded positions, as shown in Figure 1, now may be an opportune time to consider “getting off the roller coaster” by de-risking the pension plans. This means shifting toward more stable investment strategies to preserve funded status gains. However, timing this transition can be challenging as de-risking a Taft-Hartley plan is different from the well- established de-risking strategies for a single-employer corporate pension plan. Here, we explore how trustees can capitalize on the current opportunity to implement a strategy that improves long-term sustainability. 

Figure 1: Milliman Multiemployer Pension Funding Study

Source: Milliman Multiemployer Pension Funding Study

View accessible version of this chart.

How did we get here?

Traditionally, Taft-Hartley pension plans have followed a “going concern” investment philosophy — targeting long-term returns aligned with the assumed return on assets (used to value liabilities) while managing risk exposure. While this strategy can lead to robust  gains in strong markets, the pension roller coaster illustrates that plans are exposed to significant downside risk during market downturns, often making it difficult to recover without external support. The issues were perpetuated as plans became more mature, which increased the occurrence of negative net cashflow where more money is paid out to retirees than is contributed to the plan.

ARPA’s SFA program marked a turning point. Plans receiving assistance are now required to invest the majority of those funds in high-quality, investment- grade fixed income assets. This approach provides more predictable cash flow and improves benefit security for retirees for decades to come. Although only underfunded plans receiving SFA are subject to the requirement, the structure has sparked broader interest among trustees in applying similar strategies across all Taft-Hartley plans, not just those receiving aid. 

What can we do?

A key risk management tool is de-risking with a custom liability-driven strategy, in particular one focused on cash flow matching. This approach aligns a portion of the plan assets — specifically investment grade bonds — with the timing and amount of expected retiree payments.Most Taft-Hartley pension plans are mature, with a large share of liabilities attributable to retirees already receiving benefits. These liabilities tend to have shorter maturities/durations and are therefore more efficient to hedge using a cash flow-matched fixed income strategy.

For the sample plan illustrated in Figure 2, the average maturity of the retiree liabilities is half that of the active population. By constructing a diversified portfolio of individual bonds that mature in line with the scheduled benefit payments, the plan makes sure funds are available precisely when needed leading to reduced funded status volatility.

 Figure 2: Sample Projected Benefit Payments

Source: PNC

View accessible version of this chart.

How much de-risking is appropriate?

There is no one-size-fits-all answer. The optimal level of derisking depends on many factors, including fiduciary risk tolerance, plan status, plan maturity, funded status, expected investment return assumptions and expectations for future funding. A prudent first step is to shift the existing fixed income investments to a liability- matching framework. If the fixed income allocation is small relative to the total liability, plans can consider matching a specified portion of future retiree cashflows (e.g., up to 10 years). From there, plans can adopt a glidepath strategy that has a higher allocation to return- seeking (i.e., equity-like) assets at lower-funded levels and gradually shifts toward liability-matching assets as the funded status improves.

A reasonable target when the plan reaches full funding is to allocate enough to cover all retiree liabilities using a liability-matching strategy. For instance, the plan shown in Figure 3 that currently has 31% allocated to fixed income assets would shift the initial 31% to liability-matching strategy, and further move toward 70% liability match, corresponding to the share of liabilities for retirees in pay status as the plan ages. This strategy minimizes shortfall risk in the plan while preserving some exposure to growth assets to cover the longer-term liabilities of active participants.

Figure 3: Sample Asset Allocation Progression

Source: PNC

View accessible version of this chart.

What's next?

De-risking is not just a tactical shift, it’s a strategic evolution toward long-term stability. Before implementing changes, plan trustees should conduct a comprehensive asset liability study to understand the impact on plan metrics, particularly the expected return on assets (which directly affects liability valuations for Taft-Hartley plans. With the recent higher interest rate environment, many plans can de-risk without significantly reducing expected returns. But each plan’s situation is unique, and careful analysis is essential.

Once a de-risking strategy is implemented, reporting requirements should evolve to focus on asset performance relative to liabilities. This shift in focus promotes a comprehensive view of plan health, ultimately providing greater retirement security for participants.

For plan trustees considering how to stabilize the pension, now is the time to evaluate options. A thoughtful de-risking approach could be the key to leaving the pension roller coaster behind — for good.

For more information, contact Kimberlene Matthews, FSA, EA, CFA, Managing Director, Pension & Enterprise Solutions, at PensionEnterpriseSolutions@pnc.com.

 


Accessible Version of Charts

Figure 1: Milliman Multi-employer Pension Funding Study

Date Surplus/(Shortfall) Billions Funded percentage
12/31/2007               (73,145,444,266) 85.2%
3/31/2008               (98,586,470,138) 80.1%
6/30/2008            (110,588,114,506) 77.8%
9/30/2008            (139,129,764,874) 72.3%
12/31/2008            (201,407,000,338) 60.1%
3/31/2009            (238,316,283,929) 53.2%
6/30/2009            (199,682,328,181) 61.0%
9/30/2009            (164,347,811,002) 68.1%
12/31/2009            (158,550,239,200) 69.5%
3/31/2010            (150,604,665,992) 71.2%
6/30/2010            (178,833,040,490) 66.0%
9/30/2010            (158,514,566,709) 70.1%
12/31/2010            (140,800,893,572) 73.5%
3/31/2011            (133,481,939,036) 75.1%
6/30/2011            (137,056,228,482) 74.5%
9/30/2011            (183,377,109,995) 66.0%
12/31/2011            (164,075,700,523) 69.8%
3/31/2012            (145,037,429,174) 73.5%
6/30/2012            (156,095,103,384) 71.6%
9/30/2012            (148,084,424,379) 73.7%
12/31/2012            (157,550,377,473) 72.4%
3/31/2013            (138,276,387,788) 76.0%
6/30/2013            (142,313,870,637) 75.4%
9/30/2013            (131,649,996,652) 77.4%
12/31/2013            (111,980,014,549) 80.9%
3/31/2014            (119,314,676,938) 79.7%
6/30/2014            (108,104,883,802) 81.7%
9/30/2014            (120,263,201,440) 79.8%
12/31/2014            (117,199,750,866) 80.4%
3/31/2015            (115,457,348,347) 80.8%
6/30/2015            (124,693,623,104) 79.4%
9/30/2015            (159,140,103,487) 73.9%
12/31/2015            (150,929,169,709) 75.3%
3/31/2016            (152,211,732,400) 75.4%
6/30/2016            (150,404,118,772) 75.8%
9/30/2016            (141,865,432,296) 77.5%
12/31/2016            (145,715,187,428) 77.0%
3/31/2017            (133,769,112,496) 79.1%
6/30/2017            (125,277,879,416) 80.5%
9/30/2017            (124,233,866,251) 81.1%
12/31/2017            (110,022,325,643) 83.3%
3/31/2018            (123,898,675,774) 81.3%
6/30/2018            (124,638,619,623) 81.3%
9/30/2018            (123,504,054,969) 81.7%
12/31/2018            (176,166,449,702) 74.0%
3/31/2019            (137,223,191,935) 79.9%
6/30/2019            (123,772,128,613) 82.0%
9/30/2019            (134,827,894,614) 81.0%
12/31/2019            (107,488,292,000) 84.9%
3/31/2020            (200,104,496,199) 72.1%
6/30/2020            (133,413,057,187) 81.5%
9/30/2020            (128,874,270,914) 82.3%
12/31/2020               (90,771,039,174) 87.6%
3/31/2021               (80,886,710,400) 89.0%
6/30/2021               (59,822,133,615) 91.9%
9/30/2021               (80,793,354,502) 89.3%
12/31/2021               (68,542,657,707) 91.0%
3/31/2022               (91,996,862,229) 88.0%
6/30/2022            (154,201,046,623) 80.0%
9/30/2022            (188,021,974,894) 75.9%
12/31/2022            (165,577,524,056) 78.9%
3/31/2023            (110,454,695,973) 86.1%
6/30/2023            (101,615,670,766) 87.2%
9/30/2023            (121,422,183,790) 84.9%
12/31/2023               (87,364,818,855) 89.2%
3/31/2024               (59,073,573,316) 92.7%
6/30/2024               (57,659,080,484) 92.9%
9/30/2024               (12,503,339,865) 98.5%
12/31/2024               (22,725,438,614) 97.2%

Source: Milliman Multiemployer Pension Funding Study

Figure 2: Sample Projected Benefit Payments

  Active Deferred Retired Total
2022 3,418,162 209,307 67,503,885 71,131,353
2023 6,429,602 341,154 65,056,894 71,827,650
2024 9,051,574 514,849 63,065,843 72,632,266
2025 11,646,765 704,699 61,146,943 73,498,407
2026 14,032,968 860,608 59,188,089 74,081,664
2027 16,155,435 1,058,261 57,283,860 74,497,555
2028 17,977,433 1,226,410 55,418,860 74,622,704
2029 19,538,556 1,402,945 53,552,649 74,494,150
2030 21,212,030 1,583,697 51,620,372 74,416,099
2031 22,531,935 1,766,722 49,635,485 73,934,142
2032 23,704,044 1,970,279 47,597,248 73,271,571
2033 24,740,725 2,119,702 45,502,533 72,362,960
2034 25,815,732 2,248,990 43,360,038 71,424,760
2035 26,803,233 2,346,208 41,178,786 70,328,228
2036 27,662,146 2,458,385 38,967,961 69,088,493
2037 28,460,220 2,548,754 36,736,733 67,745,707
2038 29,209,961 2,612,066 34,494,399 66,316,426
2039 29,856,044 2,665,611 32,250,573 64,772,227
2040 30,402,636 2,717,729 30,015,145 63,135,510
2041 30,836,345 2,746,820 27,798,297 61,381,462
2042 31,223,854 2,784,540 25,610,448 59,618,841
2043 31,527,484 2,797,096 23,462,269 57,786,849
2044 31,746,072 2,813,090 21,364,686 55,923,848
2045 31,842,493 2,807,037 19,328,751 53,978,280
2046 31,839,255 2,807,039 17,365,554 52,011,849
2047 31,722,754 2,768,706 15,486,069 49,977,528
2048 31,525,881 2,702,191 13,700,944 47,929,016
2049 31,213,782 2,630,778 12,020,272 45,864,833
2050 30,784,691 2,553,638 10,453,063 43,791,392
2051 30,253,165 2,465,988 9,006,850 41,726,003
2052 29,640,102 2,385,938 7,687,226 39,713,265
2053 28,898,175 2,289,308 6,497,456 37,684,938
2054 28,081,939 2,180,760 5,438,229 35,700,927
2055 27,197,735 2,066,037 4,507,482 33,771,255
2056 26,220,256 1,949,403 3,700,524 31,870,182
2057 25,166,955 1,824,222 3,010,279 30,001,457
2058 24,082,566 1,701,570 2,427,790 28,211,926
2059 22,953,947 1,577,146 1,942,774 26,473,867
2060 21,789,302 1,457,823 1,544,180 24,791,304
2061 20,608,820 1,339,828 1,220,762 23,169,410
2062 19,408,431 1,225,759 961,529 21,595,718
2063 18,210,145 1,116,614 756,104 20,082,864
2064 17,024,370 1,013,160 594,991 18,632,521
2065 15,864,945 915,619 469,732 17,250,296
2066 14,731,455 824,241 372,996 15,928,691
2067 13,635,270 739,011 298,582 14,672,863
2068 12,581,116 659,865 241,386 13,482,367
2069 11,572,174 586,669 197,309 12,356,152
2070 10,610,540 519,245 163,131 11,292,916
2071 9,697,016 457,399 136,377 10,290,792

Source: PNC

Figure 3: Sample Asset Allocation Progression

 

Current

Retiree Liability Match

Future Stage – Fully Funded

Return Seeking

69%

69%

30%

Fixed Income

31%

0

0

Liability Match

0

31%

70%

Source: PNC