Good, But Not Great!

While many private sector DB pension systems have reduced risk in their portfolios as they approached full-funding, many public and multiemployer pension systems have taken on more risk trying to enhance returns to help close their pension deficits. Both public and multiemployer plans have strategically allocated significant assets into both equities and alternatives since the Great Financial Crisis, and the total combined exposure is now greater than it was prior to the 2008 meltdown. For those plans riding greater exposure to equities and alternatives, 2019 must be a phenomenal year. Everything is up! Just look at the S&P 500, which is up a very strong 18.5% for the first 6 months.

The markets have rewarded plan sponsors for their aggressive asset allocation, but not to the extent that most sponsors and their consultants would have had you believe. How is that possible? Well, plan liabilities have grown rapidly, too. Ryan ALM has produced performance numbers for a generic asset allocation of 60% S&P 500 (18.5%), 30% Bloomberg Barclay Aggregate Index (6.1%), 5% MSCI EAFE (14.5%) and cash (1.5%) the combination produces a very solid, and ROA beating, 13.8% year-to-date return. But, managing a pension plan isn’t about beating the ROA, but eclipsing plan liability growth, which is up a robust 10.1% YTD (according to Ryan ALM) when using the U.S. Treasury STRIPS curve. If one were operating under FAS 158 accounting liability growth would be even stronger at 14.3% YTD or 0.5% greater than the generic asset allocation cited above.

Why has liability growth been so robust? Most plans have a duration on their liabilities somewhere between 10-15 years. Long U.S. interest rates have plummeted since November 2018. In fact, as of today (7/8), the U.S. 10-year Treasury’s yield is down 121 bps since 11/8/18, while the U.S. 30-year Treasury’s yield has fallen 100 bps from its peak on 11/2/18.

Why is this important? Pension liabilities are bond-like (when marked-to-market) and the present value of liabilities grows as yields fall. For a plan with a liability duration of 12 years, liabilities will have grown somewhere between 12-15% since the latest peak in yields just on price movement alone.

Unfortunately, this follows the very poor fourth quarter of 2018 when asset growth underperformed a generic pension plan liability by more than 13%. I suspect that most plan sponsors are feeling great about recent asset performance, but when overlaid onto plan liabilities the story isn’t nearly as robust. Managing a DB pension without knowledge of how the opponent is doing (plan liabilities) is like trying to play a sport when you only know your score. It is impossible to adjust your strategy confidently without greater transparency on how your liabilities are performing.

Advertisements

Is Your Systematic Manager Overfitting?

With the explosion of data and vast computing capabilities, machine learning and artificial intelligence are all the buzz these days. However, a serious consequence of this is the ease of producing a wonderful looking backtest that fails to deliver when put into practice. In statistical parlance, this is called overfitting, or a situation where researchers have merely captured noise in data. This is believed to be largely responsible for the claim that “most claimed research findings in financial economics are likely false.”

An outstanding paper written by Spring Valley Asset Management gives context to this claim by demonstrating that seemingly inconsequential changes to an investment model can introduce substantial risks when put into practice. It then explores a novel approach to mitigate these risks. Whether you design, or allocate to, systematic investment strategies, the concepts described in this paper are of immense importance. I hope that you’ll find this paper as educational as I did.

 

We Don’t Have The Luxury of a Clean Slate!

An acquaintance of mine recently shared with me an article from The Hill, titled, “Multiemployer Pension Bailout Plan Is Fatally Flawed”. Renowned economist, Olivia Mitchell, Economics professor at the Wharton School of the University of Pennsylvania, wrote the article. Much of what she highlights as issues pertain to the broader multiemployer pension space and are not relevant to the argument as to whether or not H.R. 397 should or shouldn’t be supported. Unfortunately, we are not working with a clean slate at this time and as a result, we cannot wipe away years of actions and decisions that have contributed to today’s pension crisis. If we could, I would agree with many of her suggested actions.

H.R. 397 (aka the Butch Lewis Act) would provide a lifeline to more than 120 Critical and Declining (C&D) multiemployer plans that provide (or will) benefits to more than 1 million Americans. A collapse of these plans and their subsequent move to the Pension Benefit Guaranty Corporation (PBGC) would provide pennies on the dollars to these pensioners. Fortifying these plans through the proposed loan program would keep taxpayers from having to support a very expensive pay as you go social safety net. Furthermore, Ms. Mitchell complains that some of these plans may not be able to pay back the loan in 30 years or may have to renegotiate the terms. Are you kidding me? Countries, corporations, and individuals renegotiate terms of their loans all the time. Why should this be a basis for sabotaging this legislation?

In addition to her concern about a potential default 30 years from now, she also raises several other concerns. Ms. Mitchell claims that the legislation doesn’t prohibit benefit enhancements. She is wrong. Section 4, Loan program for multiemployer defined benefit plans, subsection B Loan terms, paragraph 3A as a condition of the loan, the plan sponsor stipulates that— (A) except as provided in subparagraph (B), the plan will not increase benefits, allow any employer participating in the plan to reduce its contributions, or accept any collective bargaining agreement which provides for reduced contribution rates, during the 30-year period. Furthermore, paragraph B has to do with reinstating benefit cuts in plans that have filed for relief under MPRA.

The legislation contemplates a new course for the plans and their sponsors. Any plan taking a loan must defease the retired lives and the terminated vested liabilities. This reduces risk in the process. Furthermore, the determination of the loan amount is predicated on a true mark-to-market of those liabilities. U.S. Treasury STRIPS will be used to determine the true liability. The defeasance strategies that will be considered use in two of the 3 cases Corporate bonds which significantly out-yield Treasury STRIPS providing for additional assets that can be used to meet future liabilities (Active Lives).

Importantly, the defeasance of retired and terminated vested liabilities addresses the near-term liabilities and as such, extends the investing horizon for the balance of the Active Lives. This extended time horizon is an important investing tenet. The predictability of returns over a 15-year or longer time horizon will allow the remaining assets to capture the liquidity premium that exists in equities and alternative assets.

Again, if we had a clean slate and we were creating a multiemployer pension system starting from day one we would agree with her that all liabilities should be priced at a true mark-to-market rate, that PBGC premiums be greater, that benefits not be enhanced until the current promises are met, and that exiting companies pay their fair share of what is owed and not leave the liability to the surviving sponsors, but alas we are where we are and protesting that this legislation doesn’t address these issues is a non-starter.

We cannot kick this pension crisis down the road any longer. Furthermore, in an attempt to address this situation we must not enact legislation that harms currently “healthy” plans by increasing their costs either through a significant reduction in the discount rate currently used or a dramatic increase in PBGC premiums. We have an opportunity through the Federal Government to provide a lifeline of 30-years. Much can happen during that time period. Nothing good happens if we let these C&D plans collapse, causing bankruptcies and reduced pension benefits under the PBGC. If you are concerned about the taxpayers then doing nothing puts them on the hook to support the federal social safety net far sooner than if we try a loan program.

 

Has It Been A Change For The Better?

The primary objective in managing a defined benefit pension is the securing of the promised benefits. Asset allocation plays a significant role in accomplishing that objective. However, there are at least two schools of thought regarding how one should manage to that objective.

According to U.S. Federal Reserve data, in 1952 the average U.S. DB pension plan had 96% of their assets in fixed income and cash equivalents. Interestingly, U.S. interest rates were not too dissimilar to where they are today. In fact, the U.S. 10-year Treasury yield was at 2.68% at the beginning of the year in 1952, which was only 3 bps different from where it started this year (2.71%). Funded status wasn’t particularly strong in those days, but minimum-funding standards hadn’t been established and wouldn’t be until the passage of ERISA in 1974.

The more conservative asset allocation created a greater dependence on contributions to fund the promised benefits, but significantly less volatility associated with the plan’s asset allocation made for greater certainty of what those contributions would look like from year-to-year!

Today, DB plans have a far greater exposure to non-fixed income through considerable investments in equities and alternatives. This significant movement was done primarily to enhance the long-term investment returns achieved by the plans making contributions a less significant part of the plan’s funding.  However, the greater use of these higher risk investments has injected significant volatility into the process and has created far greater uncertainty surrounding annual contributions, especially in light of two devastating corrections in the last 18 years.

Perhaps an allocation of 96% to fixed income wasn’t the right answer in 1952, but approaching asset allocation with a gambler’s mindset of letting it ride doesn’t seem to be working either. For those of us who have been around since the late ’70s and early ’80s, you may recall the use of dedication and defeasance strategies that removed significant risk from plans by effectively matching plan liabilities and assets.

As we’ve traveled across the country discussing enhanced asset allocation strategies we’ve highlighted the need for plan sponsors to once again become more focused on their plan’s liabilities to help guide and inform investment structure and asset allocation decisions. Our approach consists of allocating a portion of plan assets to a cash flow matching bond portfolio in order to meet near-term liabilities chronologically (retired lives). The remaining assets are going to be invested with a more aggressive risk profile given that the investing horizon has been extended allowing for more time for these assets to capture their liquidity premium.

1952’s asset allocation might not be appropriate for today given constrained budgets, but living with considerable volatility of returns in a 4.5% risk premium environment doesn’t seem right either. Adopting our model will help to stabilize the funded status and contribution expense while taking every opportunity to reduce unnecessary risk in the process.

De-Risking Isn’t Necessarily An End Game

Mercer’s study on pension de-risking that was done in conjunction with the PBGC and a plan sponsor advocacy group highlighted the following. “This de-risking trend, backed by pension industry data collected in the first phase of this study, highlighted that de-risking is not viewed solely as a means to exit the defined benefit pension system in the short-term; rather it is a practice embraced by plan sponsors across the board, including those heartily committed to maintaining ongoing pension plans.”

One respondent even took this a step further to suggest that it was the entities’ responsibility to evaluate these de-risking tactics to make the most appropriate economic decisions for their organizations.

“I think it’s our job to always continue to look for ways to transfer risk and to minimize the risk to make the best economic decisions both for the company and for our retirees.”

As we’ve discussed many times in this blog, a DB pension plan shouldn’t be focused exclusively on generating the highest return. The primary objective should be about SECURING the promised benefit at the lowest cost. DB pension systems should consistently seek opportunities to reduce risk in their plans, but they need more tools to accomplish this objective.

I mentioned at the Opal Public Fund Forum in January (Phoenix, AZ) that a generic asset allocation had outperformed generic plan liabilities by 11.7% for the 9 months ending September 30, 2018.  An incredible achievement as both assets rose, while US interest rates did, too. However, because plan sponsors rarely see how their plan’s liabilities are performing outside of receiving their annual actuarial report they didn’t appreciate this significant advantage.

Regrettably, the fourth quarter’s weak asset returns and subsequent rally in U.S. rates (U.S. Treasury 10-year bond yield peaked at 3.24% on 11/8) saw the 11.7% advantage disappear, and become a -1.7% relationship by year-end. This is a perfect example of a wasted opportunity to reduce risk while stabilizing both contribution expense and the funded status of the plan.

Many plan sponsors and their consultants believe that de-risking any portion of their plan impairs their ability to achieve the ROA, especially in this low-interest rate environment. I would counter that claim by suggesting that the significant ups and downs created by a traditional asset allocation places the plan in greater jeopardy than one that de-risks when given the opportunity, which will produce a more stable funded status and contribution expense. De-risking is going to be different for every plan, as plan liabilities are like snowflakes. There is no one-size-fits-all approach.

Why DB? This is Why!!!

There are people who have argued with me that there is no retirement crisis. Really? There are those that claim that because only about 40% of the private sector had been covered by DB plans at their peak that today’s 14% is no big deal. Really?  Let’s take a look at the latest information from the U.S. Census Bureau, and you tell me that there isn’t a crisis.

While the 401k is one of the “best” available retirement saving options for many workers, only 32% of Americans are currently investing in one, according to the U.S. Census Bureau. That is shocking given the number of employees who have access to one, which is estimated at 59%.

According to a Personal Capital-sponsored study conducted by ORC International, nearly 40% of Americans have saved nothing for retirement and 56% have accumulated less than $10,000. Are they foolish, guilty of consumerism, lazy, etc.? Heck no! In many cases lower- and middle-income Americans just don’t have the financial wherewithal to set aside a portion of their compensation to fund a retirement vehicle. Furthermore, why do we think that it is good policy in the first place to expect a significant majority of our population to become portfolio managers? It is NOT an easy problem for many of us to fund, manage, and then disburse a retirement plan.

Just how bad is the problem?

<25 $4,154.00
25-34 $22,256.00
35-44 $61,631.00
45-54 $116,699.00
55-64 $178,963.00
65+ $196,907.00

*Source: https://pressroom.vanguard.com/nonindexed/How-America-Saves-2017.pdf

These numbers are pitiful especially when one realizes that life expectancy for the average female is 86.6 years, while it is 84.3 for men. When asked how much one should have saved to have a “comfortable” retirement, 40% of Americans indicated that at least $1 million was necessary. Given that the average 65 year-old has accumulated <$200,000, I would suggest that there is a significant shortfall.  Just how are these retirees going to “live” on 5% of their meager balances combined with Social Security, which pays an average of just over $16,000? You barely can live in the Northeast as a homeowner and pay your property taxes with that small savings.

Defined Benefit plans are not perfect, but they are much better for the average American than a defined contribution plan. First, they are professionally managed. Furthermore, one cannot take an early withdrawal or a loan. They can’t stop contributing and benefits are usually paid out on a monthly basis in the form of an annuity. DC plans were NEVER intended to be anyone’s primary retirement vehicle. They were set up to be supplemental retirement vehicles for high earners. Where did we go off course?

The Benefits of Bailouts

We reported yesterday that the TARP and Fannie and Freddie bailouts have resulted in more than $107 BILLION in profits for the U.S. Treasury. In addition to this profit, bailouts have been beneficial in that they preserved companies and jobs, wages and tax revenue, and pensions while protecting investors in bonds and stocks of those companies.

With regard to the preservation of pensions, and specifically DB pension plans, Diane Oakley, former Executive Director for the National Institute on Retirement Security (NIRS), spoke to the importance that benefits received by the participants play in contributing to a healthy economy. According to the NIRS, 40 million Americans were covered by a private (both single and multiemployer) pension plan in 2016, including more than 13.5 million retirees. The 13.5 million beneficiaries received just over $200 billion in benefits equating to just under $15,000 per participant (multiemployer recipients got about $12,000 on average).

With regard to the impact of just the multiemployer payouts on the economy, the NIRS has determined that “246,324, jobs were attributable to direct impacts (direct spending by retirees), 122,978 to indirect impacts (spending by merchants on businesses further up the supply chain), and 173,566 through induced impacts (additional jobs supported when employees whose jobs are tied to direct and indirect spending rounds spend their paychecks). These jobs collectively paid out an estimated $27.9 billion in labor income.”

The NIRS analysis also calculated that the $41.8 billion in benefits received in 2016 stimulated the economy to the tune of $89 billion in economic output. Lastly, the tax revenue produced by this economic activity is roughly $14.7 billion, with $8.4 going to the Federal government.

It is clearly past the time to put ideologies aside and get something implemented before the pension crisis magnifies and a solution becomes impossible. We need to protect our retirees before we not only damage their financial well being but the health of the U.S. economy in the process.