We’ve been discussing pension obligation bonds (POBs) for the last couple of years and it seems as if folks are listening! According to a WSJ article that was published during the weekend, there have been 72 POBs issued in 2021 compared to the annual average of 25. In addition, more POB money has been raised this year than in any year during the last 15 years. We think that a POB can dramatically improve a plan’s economics provided that the proceeds are not used within the plan’s current asset allocation.

According to the WSJ article, the average interest rate for a municipal POB is 3%, while the average pension plan is striving for a 7% ROA. If the potential arbitrage is the driving factor in the decision to issue these bonds – good luck! Certainly, the historically low-interest rates help but remember that we are at historic levels for equities and the fundamental support is eroding. A dramatic decline in the equity markets would undercut the benefits of this issuance. We’ve seen this story unfold numerous times and it is what led the Center for Retirement Research at Boston College to issue their initial POB study in 2009. Their conclusions were not very supportive of using POBs to support DB plans.

There are potentially huge savings by issuing POBs in this environment, but plans need to be smart. Defease your pension liabilities chronologically as far out as possible using the POB proceeds. This allows your current assets and future contributions to be used for alpha-generating purposes, as they can now grow unencumbered as they are no longer a source of liquidity to meet benefits and expenses. Our work in this space has shown that plans can dramatically improve their plan’s funded status, stabilize contribution expenses, and in most cases reduce the target return on asset assumption (ROA). We’d be happy to produce a Custom Liability Index (CLI) highlighting the impact of your proposed POB on your plan’s future economics. Don’t hesitate to reach out for some help.

We, at Ryan ALM, are big supporters of Pension Obligation Bonds (POB) provided that the proceeds are used to defease the pension system’s liabilities chronologically for as far out as the POB allocation will permit. We are also encouraging the use of this funding strategy because of the historically low interest rate environment (reason #1). But there are other reasons, too. We remain very concerned about the underlying fundamentals of the US equity market and the extended blow-out performance for the last 10 years that would suggest (regression to the mean). As we wrote yesterday, the dividend yield on the S&P 500 (1.28% as of 9/1) has only been lower in 1999 (1.17%) and 2000 (1.22%).

For the 10-years ending 12/31/2009, the S&P 500’s total return (including dividends) was -0.95%! We aren’t suggesting that we are on the verge of a similar result, but given the 11 basis points difference in yield, we are intimating that US equity returns are not going to be close to historic averages for the next 10-years. Here are some #s to chew on:

S&P 500 (total returns through 7/31/21)

1-year   35.5%

3-years 18.2

5-years 17.4

7-years 14.7

10-years 15.4

20-years 8.8

Since 1871 to present: 9.1%

Reversion to the mean suggests that the 8.8% 20-year S&P 500 return generated through the period ending 7/31/31, would only produce a 2.6% return for the next 10-years given that it achieved a 15.4% return during the first 10-years of that period. Furthermore, I do believe that 8.8% is a reasonable target to shoot for over 20-year periods given the 9.1% annualized return since 1871. If you believe that a 10% annualized 20-year return is more appropriate because something has changed in the markets to support this higher target, a 10% annualized return for 20-years ending 7/31/31 would mean that the next 10-years produces a 4.85% annualized gain. Still no where near “average” long-term results or what is forecast for many DB plan asset allocations.

Why a POB? If plan sponsors are to receive considerably less from their alpha assets during the next 10-years, they will likely need to contribute much more than they have been to date. This greater contribution negatively impacts the sponsoring entity’s ability to support other important programs or initiatives. By issuing a POB at this time and using the proceeds to defease plan liabilities through a cash flow matching process, you are dramatically changing the plan’s economics, while securing the promised benefits for 10-years or more. In addition, you are providing those alpha assets with time to weather potentially rocky markets, as they are no longer a source of liquidity to meet benefits and expenses.

As stated previously, there is regression to the mean tendencies within our markets. You don’t get the types of returns that we’ve enjoyed recently without impacting (stealing from) future returns. There are strategies that can be utilized to help reduce the impact of weak markets on pension plan funding. Taking advantage of historically low interest rates to issue a POB and using those proceeds to pre-fund your plan makes sense to us. Defeasing the plan’s liabilities with those proceeds makes even greater sense. DB plans have benefited from historic returns recently. It is time to rethink your asset allocation for the next 10-years and beyond. We can’t afford as a nation to have more DB pension systems shuttered and plan sponsors can’t afford to see contribution expenses continue to rise. It is time to act.

There once was a time when investors demanded more from corporate America than just the opportunity to invest in companies with the hope of a return on that investment. They demanded and were given significant dividends. It was not unusual for the dividend yield on the S&P 500 to be greater than the yield on corporate bonds because of the risk that the equity investor was taking. In fact, it wasn’t until 1958 that the US 10-year Treasury Note had a yield that eclipsed the S&P 500’s dividend yield. Why the change?

In 1871, the dividend yield on the S&P 500 was 5.49%. It would remain robust for many years to come peaking at >10% in 1917 and 9.72% in 1931. It wasn’t until 1961 that we first had the S&P dividend yield fall below 3% (2.98%). It wasn’t until 1997 that we had the first dividend yield on the S&P 500 fall below 2%. The yield would subsequently bottom at 1.17% in 1999.

The dividend yield briefly rose to above 3% in 2008, but it has since plummeted to today’s level at 1.28% (9/1/21). Dividends have always played a significant role in the total return of the S&P 500 throughout its history. Furthermore, the level of the dividend yield has been a great predictor of future returns. We all know how the decade of the 2000s performed following the S&P’s lowest dividend yield. What does that suggest for today’s level and the next 10-years?

According to the folks at Advisor Perspectives, for an average holding period of 1 year, dividends accounted for 27% of total returns for the S&P 500 since 1940. Over a 10-year period the contribution of dividends to the total return rises to 48% and with a 20-year holding period dividends account for roughly 60% of total returns. Incredible! Worse than just the decline in the overall dividend yield is the fact that dividends are no longer as sacrosanct as they once were. We witnessed this quite dramatically during the second quarter of 2020 when many companies cut or eliminated their dividends in their response to Covid-19 related events. In fact, only 76% of S&P 500 companies paid a dividend in 2020 compared to 95% in 1980. Given how important dividends are to the total return of the S&P 500 over time, why would you ever elect to invest solely on the potential of an investment and not the certainty of receiving a significant down-payment each and every quarter?

In our recent post, “Right Idea, Wrong Implementation” we discussed the fact that using STRIPS in lieu of coupon bonds would neither reduce the cost of de-risking nor work more effectively than a cash flow matching portfolio. We believe that the true pension objective is all about cash flows…asset cash flows versus liability cash flows. The use of US Treasury STRIPS in lieu of coupon bonds is a very expensive implementation.  Just how expensive?

We produced two Custom Liability Indexes (CLI) as of June 30, 2021 – one using STRIPS and the other coupon bonds. We have a representative defined benefit liability stream that is $1 billion in future value liabilities and has payments going out to 2102. When using U.S. Treasury STRIPS to implement a de-risking strategy the present value cost to defease that liability is $695.2 million. The yield on that portfolio is 1.77% and the modified duration is 16.3 years. When implementing a defeasance strategy using our Liability Beta Portfolio (LBP) invested in U.S. investment grade corporate bonds, we see a cost savings versus the STRIPS implementation of $142.2 million! The yield on our portfolio is 3.6% and the modified duration is 13.3 years. The cost to defease $1 billion in pension liabilities using our cash flow matching approach saves the “client” $446 million (44.6%).

As a reminder, the funding cost savings are realized immediately upon implementation allowing the client (and their consultant) to use those savings within the alpha or performance portfolio to hopefully improve the probability of achieving the return on asset (ROA) objective. Oh, and by the way, a cash flow matching portfolio, which uses interest, principal, and reinvested income to meet the future benefits is the most efficient implementation. For example, a $10 million per year projected benefit schedule would require a $500 million bond portfolio at a 2% yield to fund benefits through income. Our LBP cash flow matching could fund these same projected benefits for only $84 million.

I’m happy to share with you today a wonderfully written article by Eleanor Laise, Reporter, Barron’s Group, who effectively captures the human toll inflicted by MPRA legislation that saw 18 multiemployer pension plans slash promised benefits to their participants. Highlighting the conversation in Eleanor’s article is Carol Smallen, who I had the pleasure of first speaking with in August 2018 and whose story I shared on several occasions within this blog. Carol’s story is heart-wrenching, but regrettably, not unique! Importantly, Eleanor also introduces us to several other individuals who have been severely impacted by developments within a subset of the multiemployer pension plan universe.

When we, as an industry, discuss issues related to DB pension plans it is easy to get lost in the debate surrounding various investment theory/strategies, and often neglect the true reason why it is so important that these issues are addressed – the plan participants, who are our relatives, friends, and neighbors. These hard-working Americans went to their jobs each and every day with the expectation and understanding that they were entitled to a pension upon retirement that was going to be based on years of service and other elements. Why was it okay that the promise that they were given was not kept? We need to keep fighting for these individuals and for all American workers, who should have access to a defined benefit plan. Defined contribution plans were intended to be supplemental to DB plans, and not Social Security.

There recently appeared an article in FundFire that suggested that plan sponsors, particularly corporate sponsors, were using their allocation to fixed income inefficiently. We absolutely agree! However, we find that the suggested implementation cited in that article to be just wrong. The article stated that corporate pension plans had roughly 40% in fixed income, which a consultant from a leading firm said was too much. Again, we agree. This consultant went on to say that plans should use US Treasury STRIPS in lieu of coupon bonds, which would allow them to put far fewer $s to work. This is incorrect math and needs to be tested. The pension objective is all about cash flows… asset cash flows versus liability cash flows.

US Treasury STRIPS are highly volatile and expensive. They performed well during the bond market’s lengthy bull market (since 1982), because they have longer durations than coupon bonds. Since we are near historic lows in interest rates now a trend toward higher interest rates (which most economists predict) would produce an opposite effect. Moreover, STRIPS do not secure benefit payments, as they have been stripped of their income component and they certainly are not low risk.

Does it really make sense to use STRIPS now? We would also suggest that the true pension plan objective is too “secure” the promised benefits in a cost efficient manner with prudent risk.  This is best accomplished through cash flow matching liabilities with coupon bonds. The difference in cost versus STRIPS could be close to 1% per year of liabilities (a 25-year benefit payment schedule = 25% cost savings). The higher the funded ratio the more the plan sponsor should allocate to defeasance through fixed income securities. We recommend that Retired Lives should be defeased as much as possible since they are the most certain, imminent, and important liabilities.

Do you want to improve the efficiency of your asset allocation? Would you like to put fewer assets into fixed income in this low-interest-rate environment? Adopting a cash flow driven investing (CDI) approach will accomplish your objectives. As the example below reflects, traditional bond management (TBM) requires an allocation of $500 million to fund $10 million in annual benefit payments in this 2% interest rate environment. This is very inefficient. By adopting a CDI approach, the fund can accomplish the objective of funding $10 million in benefit payments with ONLY $84 million. The additional $416 million can now be used in other asset classes to support alpha generation.

The allocation to fixed income is now 83.2% less than using a traditional bond manager. In a CDI implementation the $10 million in benefits is achieved through the use of principal, income, and reinvested income. In addition, bond math suggests that the longer the maturity and the higher the yield the lower the cost. Our cash flow matching process is a cost optimization model that will take advantage of both of these mathematical principles by biasing our portfolio to longer maturities. Thereby we can partially fund the next nine years of benefit payments with the income from a 10-year maturity at 3% yield versus a 1-year Bill providing the investor with 7 basis points and lower yields for the 1-9 year benefit payments.

A CDI strategy is the most cost efficient implementation in accomplishing the primary objective of securing the pension system’s promised benefits. Only an insurance company annuity can provide the same certainty of securing benefits, but at a much higher cost (@ 4% premium). We agree that pension plans should get smarter about the asset allocation that they adopt. It needs to be driven by the funded status and not the return on assets assumption. If plans were to start doing this, I don’t believe that there would be any debate as to how plans would approach their fixed income allocation.

I was extremely pleased to be asked to participate in a Barron’s Live interview today. Eleanor Laise and I discussed ARPA and other matters related to DB plans. With regard to ARPA we discussed the SFA calculation and discount rate issues. Despite some concerns about both, we at Ryan ALM have determined that the SFA working in conjunction with the legacy assets and future contributions can accomplish the primary objective set by Congress to pay benefits and expenses for the next 30-years, while still maintaining a surplus to meet future liabilities.

In addition, we talked about the state of public DB plans, including the use of Pension Obligation Bonds. Please don’t hesitate to reach out to me if you have any questions or comments related to what was discussed today.

We’ve written from time to time about individual pension plans whose unique situation has surprised us. Here is another example of a plan and the current asset allocation that just floored us! Here are the details: Closed plan, 150% funded based on the ROA and 95% funded using ASC 715, no more contributions, benefits can’t be enhanced, and the sponsor(s) can’t recapture any of the surplus. Given these characteristics one would assume that the asset allocation would be incredibly liability focused. Yet, that couldn’t be further from the truth. This plan is invested in 70% equity and equity-like product! This situation is outrageous.

Again, there is NO upside for anyone – neither the participant nor the plan sponsor. There is only downside risk, especially as equity market valuations and the underlying fundamentals seem extraordinarily stretched. In this particular case we were approached by the plan’s actuarial firm because they, too, recognized the incredible risk being taken in the management of these assets. Since this plan has won the battle… they have the potential to SECURE all of the promised benefits with little to no risk by cash flow matching assets to the liability cash flows. Should the equity markets fall, they have no way of recouping those losses because the fund no longer receives any contributions. They’d have to wait for the markets to recover the losses, but would they have the time? Remember a 25% equity correction requires a 33.3% recovery and a 50% correction requires a 100% recovery.

Through our analysis we are able to show that ALL of the benefits can be secured through cash flow matching. In fact:      

 • We can secure the benefits at a $6.2mm or 28.68% funding cost savings

• Leave a $4.8 mm surplus or 20%

• Reduce management fees by $100,000 per year or 50 bps.

This is one of those “no-brainer” moments that has one just shaking their head. How could this situation exist? Are we so focused as an industry on the return on asset assumption (ROA) that we’ve forgotten pension basics? Remember, the primary objective in managing a pension plan is to SECURE the promised benefits. Everything else is a distant second. I’m not sure how this situation will unfold. I hope that we will be given an opportunity to help this plan secure those promised benefits. Unfortunately, this situation is not unique, especially among smaller DB plans. As fiduciaries we have a responsibility to do what is best for the plan and the participants. Can anyone say that is true in this case?

I’ve been as much a critic of the PBGC’s Interim Final Results (IFR) as anyone. I was disheartened by how the PBGC decided that the Special Financial Assistance (SFA) should be calculated. That said, struggling multiemployer pension systems are still getting new found money that will enhance the funded status. Although the amount of the grant may be less than what was originally expected or intended by Congress through this legislative effort, we at Ryan ALM have determined that this SFA program can achieve success, but it very much depends on how the SFA and legacy assets are invested.

Many folks, including me, have been focused on the disconnect between the discount rate of 5.5% (3rd segment + 200 bps) and the potential yield (roughly 2%) from investment grade bonds within the SFA. This difference does create a funding gap between the cost to fund benefits and expenses for 30-years and the amount received, but it isn’t fatal. Many comments received by the PBGC during the 30-day comment period focused on the appropriateness of mandating a very conservative investment grade bond portfolio while this gap exists. But we believe that securing the SFA assets is paramount, while mirroring the intent of Congress.

A leading actuarial firm with a particular expertise in multiemployer plans provided us with an example of a hypothetical pension fund that would qualify to receive an ARPA grant based on the characteristics of this plan’s profile. In their example, the “client” had $100 million in beginning assets and liabilities of $256.6 million in liabilities discounted at 5.5% for a funding deficit of ($156.6 million). The funded ratio was only 39.0%. Based on the PBGC’s SFA calculation, this plan would be eligible for $180.7 million in Special Financial Assistance, which is forecast to be received in 2024 when the current assets will have declined to $63.4 million. Furthermore, they estimated a 2% return on SFA assets and a 6.75% return on legacy assets. Given these expectations the plan would likely be in a $28.1 million deficit by 2052 supporting the calls for changes to either the legislation or the PBGC’s guidelines or both.

Here’s the good news. Despite the appearance of a funding gap because the discount rate is much greater than the potential return on investment grade bonds, funding of benefits and expenses is a function of the SFA and legacy assets working in harmony. Ryan ALM is proposing that the SFA assets defease as many of the liabilities as possible once the grant is received (2024 in our example). By defeasing the promised benefits and expenses within the SFA bucket until those assets are exhausted, our approach fulfills the goal of Congress to “secure” benefits. Furthermore, the legacy assets and future contributions can now be managed with a slightly more aggressive risk profile as they aren’t needed for cash flow for the eight years from 2024 to 2031 that the SFA assets defease liabilities. Importantly, bonds are no longer needed within the legacy asset allocation. Rotating away from bonds to more equity-like products will increase the potential ROA from the existing legacy portfolio.

We have modeled a number of scenarios by tweaking the actuarial firm’s expected return on the SFA and legacy assets. For instance, an investment grade corporate bond portfolio used to defease the plan’s liabilities will provide a return somewhere in the area of 2.5% to 3% narrowing the gap to the discount rate and not the 2% used in the original scenario. Because we have eliminated bonds from our legacy assets during the 2024-2031 period in which we only use the SFA to pay benefits and expenses, the legacy portfolio should be able to achieve a return greater than the 6.75% used to calculate the SFA. In the scenario that we feel is most reasonable, we used a 2.5% return on the SFA, the 6.75% for the first three years (2021-2024), a 7.5% ROA for the legacy assets from 2024 to 2031, and a very modest 7.15% return from 2032 to 2051.

In this scenario, we are able to fully fund all future benefits and expenses for 30 years, while beginning 2052 with $77.4 million in assets to meet future liabilities. The other scenarios that we modeled used similar inputs and produced results in every case that had 2052 beginning with at least $42.6 million in assets available to meet future liabilities. I think that our inputs are extremely reasonable given the 30-year investing horizon. Furthermore, we have not had to dramatically increase risk in the legacy portfolio or the SFA portfolio to achieve these terrific results. Again, it is unfortunate that the calculation to determine the SFA was so conservative, but at the end of the day the grant money goes a long way to helping secure the promised benefits for plan participants, many who have endured tragic cuts for several years now. How the SFA assets are implemented is crucial to this program’s success. Having the SFA and the legacy assets working in harmony to achieve these positive results is critical.  It is imperative that the SFGA assets stay true to their objective of defeasing and funding benefits and expenses for as far out as possible. This requires a cash flow matching strategy with fixed income. To be clear, matching the 5.50% discount rate is NOT the objective of the SFA assets.

Lastly, I read a comment on the PBGC website that claimed that the legacy assets would need to achieve a 12%+ return in order for all the benefits and expenses to be paid. Nonsense! We have shown that a 39% funded plan (prior to getting the SFA) can achieve success even with a 7.5% return on legacy assets combined with only a 2.5% return on SFA assets. We’re happy to share our analysis with you. Don’t hesitate to reach out.

There are many definitions of risk, but the one that we think is appropriate for pensions is that risk is the UNCERTAINTY of achieving the objective. In the case of a defined benefit pension plan, risk is not the volatility of returns, but the uncertainty of paying the promised benefits! There is nothing more important than building a corpus that can accomplish this objective with great certainty. The promise, as you know, is the plan’s scheduled benefit payments (liabilities) and they must be measured, monitored, and managed on a regular basis.

The American Rescue Plan Act (ARPA), signed into law by President Biden in March 2021, provides a lifeline to struggling multiemployer pension plans that are designated as in Critical and Declining status. The legislation calls for the US Treasury to provide the necessary funds to the PBGC which will then send payments to these plans through grants. The grant money, aka Special Financial Assistance (SFA), “shall be such amount required for the plan to pay all benefits due during the period beginning on the date of payment of the special financial assistance payment under this section and ending on the last day of the plan year ending in 2051, with no reduction in a participant’s or beneficiary’s accrued benefit as of such date of enactment”. In other words, secure the promised benefits for a 30-year period – no games!

How do you secure these benefits? The PBGC and the legislation have stated that the SFA should be invested in investment grade bonds as the means to “secure” these promised benefits. Why bonds? They are the only financial instrument that has a known cash flow that can be used to meet the pension plan’s obligations. Any other asset class or financial instrument will create uncertainty as to their cash flows and/or terminal value.

The PBGC presented their Interim Final Rules (IFR) on July 9th, which included a 30 day comment period. Not surprisingly, the PBGC received 101 submissions from individuals (mostly union members), unions, Congress, and investment advisory organizations. The comments covered a variety of topics identified in the legislation including how the SFA is to be calculated, the discount rate used (3rd segment (PPA) + 200 bps), eligibility, and the most common which dealt with the approved list of investments for the SFA. One firm, a leading fixed income shop providing ALM solutions articulated perfectly the goal of the legislation by stating “we need to uphold the spirit of the Butch Lewis Act (BLA). Moreover, the SFA needs to be based on “pillars of certainty”.” AMEN!

Plan participants in these struggling plans, and especially those that have already seen benefit reductions through MPRA (18 funds), want the promise that was made to them to be returned and secured for the next 30-years. They no longer want their benefits subjected to the whims of the market. I certainly don’t blame them. Everyone should take the time to read some of the horror stories on the Teamsters’ Facebook page that speak to the pain inflicted on individuals through these benefit cuts. Plan participants, through no fault of their own, have lost homes and much more, as benefits have been slashed by as much as 50% or more – horrible.

We hope that the PBGC continues to support the idea that the SFA should only be invested in investment grade bonds (IG) (with <5% in high yield as a result of downgrades after the IG bond was owned) and that those bonds be used to cash flow match SFA assets to the promised benefits and expenses (liabilities) securing the promise that has been made, while fulfilling the intent of the legislation. These plan participants have already lived through years of uncertainty. Creating an investment program that doesn’t secure the benefits would be unfair and harsh. Any investments where the cash flows are uncertain are risky assets. To repeat the intent of the ARPA legislation: the SFA needs to be based on “pillars of certainty”. Risky assets in the SFA bucket should not be allowed under ARPA.