Complex Public Pensions Have Lost Sense of Purpose
Lengthy piece but I think there’s much to be said…
Criticism of public pension funds is widespread. Over the past few years, alarm bells have been sounding louder and louder. Are pensions a ticking time bomb? Are “lofty” return assumptions going to result in reduced retirement benefits? How much are taxpayers on the hook for? Does politically charged governance lead to irresponsible investment decisions?
These questions are fair, and each worthy of debate. For this post, though, I’d like to explore a much deeper issue…
Have some public funds lost sight of their purpose?
The Importance of Purpose
As a financial planner, I constantly preach the importance of attaching purpose to investments. In their book Personal Benchmark, Charles Widger and Daniel Crosby wrote:
By attaching purpose (the “Why?”) to our investment strategies, we identify what matters to us personally. We establish our own “Why?,” making it almost effortless to do the right thing in our investing and in our lives. This makes us more productive and our lives more meaningful.
This brief, simple statement says so much. Personally, our “Why’s” are things like; to fund our grandchildren’s education, to plan a legacy, or, to retire abroad. These are our goals. The reasons why we work, save, and invest.
Although managing a public pension fund is undoubtedly different than developing a personal investment policy, “the Why?” still matters. For these funds, the “Why” isn’t unique, and can be summarized by two basic goals:
- Help reduce the unfunded liability of the system.
- Provide a base income to qualified employees whose earning capacity has been reduced by age or disability.
Pension investing can be thought of as retirement planning on a very large scale. It is a perpetual long-term investment meant to provide retirement income for thousands of workers.
I live in a small town on the south shore of Massachusetts and for the past three years, I’ve served on my town’s finance committee. As my first foray into municipal policy, it has been a great education. Last year, I became interested in how the town’s pension liability payments were calculated. I wanted to better understand how those payments might be affected by changes in financial markets. The county I live in manages its own pension investments, but my research involved digging into the Massachusetts state fund.
Personally, I haven’t had a ton of exposure to institutional investing. Maybe that’s a good thing. A different perspective never hurts. What surprised me the most about the Massachusetts fund was its complexity. The fund’s purpose is fairly straightforward; ensure pensioners a satisfying retirement. Did long-term investment planning on a large scale have to be significantly more complicated than on a smaller, individual level? I don’t think so.
I’m not suggesting that managing a $62 billion fund is as simple as opening a Betterment account (although I think there’s some validity to a similar approach). There are governance and structural issues attached to managing massive sums of money that individuals don’t have to deal with. However, when we consider the stakeholders and the purpose of the pension investments, complexity is not a necessity.
Complex markets do not require complex solutions. The complex approach Massachusetts takes comes with enormous and largely unnecessary costs to the retirees and taxpayers. To better illustrate my point, I’d like to compare two public pension investment systems. Nevada and Massachusetts. Based on outcomes, Massachusetts might want to rethink its strategy. Let’s dig in…
Before we start, it is important to understand the differences between the two systems.
In Massachusetts, individual municipalities, counties or organizations manage their own retirement systems. They are each responsible for the administration of benefits, recordkeeping, and investment management. For those units not interested in managing their own investments, the state will do it for them. The Massachusetts Pension Reserves Investment Management Board (PRIM) acts as an asset manager for public pension dollars (the fund they manage is known as PRIT). It is responsible for managing over $60 billion throughout the state, including assets of Massachusetts Teachers and State Employees Retirement systems.
The Nevada Public Employee Retirement System (NVPERS) handles the retirement system for all public service employees. It is responsible for benefits administration, recordkeeping, and management of investments. Unlike Massachusetts, there is not a separate body solely dedicated to managing the pension investments.
For purposes of this discussion, the comparisons I draw are only related to the investment management functions of each pension system. After contrasting their stated goals, organizational structures, investment styles and returns, I’ll return to “purpose,” and why I think Massachusetts (and others) has lost its way.
The goals of both pension funds are essentially the same. My only thought on goals is a subtle one. Embedded in each system’s investment policy is a description of its “Mission.” Nevada, an integrated system, stresses the importance of its role to the pensioners. Massachusetts makes no such reference. Yes, this observation is overtly cynical, but it’s worth mentioning.
Provide public workers and their dependents with a retirement program that provides a reasonable base income for retirement or for periods where a disability has removed a worker’s earning capacity.
Encourage those workers to enter into and remain in government service for such periods of time to give public employers and the people of the State of Nevada the full benefit of their training and experience.
PRIM’s mission is to provide a professional investment service that maximizes the return on investment within acceptable levels of risk by broadly diversifying its investment portfolio, capitalizing on economies of scale to achieve cost-effective operations, and providing access to high quality, innovative investment management firms, all under the management of a professional investment staff and members of the PRIM Board and its Committees.
See the difference? The importance of attaching real, tangible purpose to an investment policy shouldn’t be understated. Nevada’s mission statement focuses on the stakeholders. Massachusetts’s mission focuses on the portfolio.
The structures of PRIM and NVPERs couldn’t be any more different. Again, keep in mind that PRIM is dedicated to managing the investments, not administering the entire retirement system.
There is not a typo in the table above. NVPERS employs one (1) person on its investment staff; Steve Edmundson. You might have seen him recently profiled in the Wall Street Journal (What Does Nevada’s $35 Billion Fund Manager Do All Day? Nothing).
While the organizational chart for investment management in Nevada is pretty simple, the one in Massachusetts (below) is not.
Boards and Committees
Steve Edmundson is responsible to a board and an advisory committee. Those bodies, however, have responsibilities related to the entire retirement system. They are not specific to investment oversight. Massachusetts is a different story. Not only does each retirement system within the state have its own governance bodies, but PRIM (investment only) does as well.
Operating costs are considered employee compensation, professional fees, and occupancy expenses. Nevada reports operating costs for the entire system and does not report granular expenses at the investment management level. The cost of operating the entire system in 2015 was $9.6 million. We know from the WSJ article that Steve Edmundson’s salary was $127,000 last year. If we estimate another $100k to account for Steve’s office space and any professional services he requires, investment related operating costs come in around $225,000. Admittedly, this isn’t a very scientific calculation, but likely a reasonable estimate.
With PRIM, we know exactly what investment related operating costs are. The $10.1 million is comprised of wages and office space for 37 employees and any related professional service fees. The executive director alone, Michael Trotsky, pulls in an annual salary of $395,000, more than 3 times that of his Nevada counterpart.
Both PRIM and NVPERS state in their annual reports and investment policies that asset allocation is responsible for the majority of their returns. Nevada favors low cost, market indexing. Massachusetts mirrors the traditional Endowment Model, allocating more money to active managers and alternative investments.
Policy Asset Allocation
Nevada segments its portfolio into five asset classes. Massachusetts uses fifteen. There’s no need to debate the specific asset weightings. I simply want to draw attention to the reoccurring theme. Nevada = Simple. Massachusetts = Complex.
Unsurprisingly, more complex investment strategies require lengthier reports. Nevada’s annual investment report is neatly tucked away in 13 pages of the NVPERS 2015 Comprehensive Annual Report (CAFR). The 2015 Annual Report for PRIM is 114 pages. Nevada’s Investment Policy Statement is 9 pages long. PRIM’s Investment Policy Statement is 23 pages long.
The difference in investment expenses (defined as: fees paid to outside managers, custodial/brokerage costs, and consulting fees) and the sheer number of outside money managers used is wild! Here’s a closer look at the managers (Nevada the smaller list):
Massachusetts manages 1.7x more money than Nevada, but requires:
- 37x more employees
- 19x more money managers
- 8x more investment expenses
Not exactly an efficient use of scale!
The Score Card
If investment returns favored complexity, most criticism would be unwarranted. If Massachusetts had spent billions of dollars in recent decades delivering spectacular investment returns to pensioners, I probably wouldn’t be writing this post. But…I am writing this post.
On all reported timeframes except 1-Year, Nevada’s simple investment policy has bested returns of PRIM. Over the last 10 years, Nevada has outperformed Massachusetts by 84bps. That equates to a 10% difference in investment returns. Past performance is not indicative of future returns but, extrapolating those numbers would result in a 16% difference over a 20 year period and a 24% difference over a 30 year period. Compounding makes little numbers add up quickly.
Benchmarking & Performance Reporting
Another odd difference between the two funds relates to how they each report performance. While Nevada reports fund returns net of fees, Massachusetts does all performance reporting gross of fees. In order to find a net return number for Massachusetts, I had to dig through their annual reports and calculate it myself . To be blunt, there is no value in reporting investment returns without accounting for expenses. As a performance metric, “gross of fees” is meaningless.
The seemingly inexplicable practice of ignoring cost also carries into how PRIM benchmarks their portfolio. From their policy statement:
The investment policy benchmark enables PRIM to compare the PRIT Fund’s actual performance to a passively managed proxy and to measure the contribution of active investment management and policy implementation.
Benchmarks are usually assembled and tracked using unmanaged indexes. Unmanaged indexes do not have investment costs. Because managed portfolios do have investment costs, comparing them to an unmanaged index isn’t exactly fair. We can reasonably assume, however, that a passively managed index costs significantly less than its active counterpart. Without much effort, the cost of owning a passive investment can be estimated. Once the cost is known, benchmarking performance (net of fees) vs a passive alternative (net of fees) is viable, and appropriate. Simply put, PRIM should be much more transparent in reporting actual investment returns.
Getting Back To Purpose
As previously stated, the goal of the pension fund managers is to provide for the retirees of their system. To achieve this goal, they have to attain a certain rate of return over long periods of time. Returns don’t have to be smooth, but risk should be a consideration. Today, the average return assumption for most funds in the country is 7.62%. The chart below shows the distribution of return assumptions amongst the 127 public funds surveyed by the National Association of State Retirement Systems earlier this year . A long-term return of 7-8% is the goal for the vast majority of public retirement systems.
What Steve Edmundson and NVPERS seem to have figured out, is that achieving long-term returns of 7-8% shouldn’t be extraordinarily difficult. In fact, one of the most respected asset-allocation models, the traditional 60/40 stock/bond portfolio, has historically handled the task quite well. There hasn’t been one decade in the past 60 years where the average 30-year rolling returns fell below 8% (h/t to Jake @Econompic for the data).
The risk assumed by this allocation (as represented by volatility and Sharpe ratio) has also been well within the parameters of what the Massachusetts fund targets . If you’re interested in a more in-depth look at 60/40, Ben Carlson has a couple good write-ups (The Real Risk to a 60/40 Portfolio and 60/40 Return Expectations).
Don’t mistake this example as me advocating for a 60/40 pension model (it would probably work fine), but rather evidence supportive of the idea that simple approaches such as 60/40 or Nevada’s model, can, and do work.
Know Your Clientele
Perhaps the most interesting quote I’ve read from Steve Edmundson is the following:
“The trick in investment management is knowing your clientele. While Nevada has casinos and gambling, the people of Nevada, particularly the (165,000) members of our system, actually have a very conservative posture. We have molded our investment portfolio to match that.” 
Steve understands the purpose of Nevada’s pension investment fund and that the path to achieving its goal doesn’t have to be complicated. If market returns meet the goals of the pensioners, why build incredibly complex investment organizations to manage portfolios in an attempt to outsmart the market? Complexity masked as sophistication is not what people need.
Avoiding the Rube Goldberg Machine
A Rube Goldberg machine is a contraption, invention, device or apparatus that is deliberately over-engineered to perform a simple task in a complicated fashion. The expression is named after American cartoonist and inventor Rube Goldberg. I can think of no better metaphor to describe what Massachusetts and many other pension funds across the country seem to be doing.
Hedge funds, “portfolio completion strategies”, and value-add, non-core, real-estate investments are completely unnecessary to meet the needs of public funds. Massachusetts spends hundreds of millions of dollars each year by taking the investing scenic route. They are Sunday driving with the retirees’ money. Hunting mosquitos with bazookas. Hammering nails with a sledgehammer. The idioms are endless. Selling people what they don’t need has long been a problem in the finance industry. The needless pension complexity may be the single greatest example that.
With Nevada, we see a very clear example of prudent investment management that has a tight grasp on purpose. It is simple and effective. Leonardo da Vinci said, “Simplicity is the ultimate sophistication.” Massachusetts and other funds across the country should take those words to heart. Eliminate the fuss, remember “the Why?”, and never forget purpose. It’s time to leave the Rube Goldberg machines behind.
A Few Notes:
- I focus on Massachusetts but complexity is by no means unique to this fund. Complex endowment approaches have become widely adopted in the past two decades. Complexity seems to be the rule rather than the exception.
- One downside of complexity that I did not mention, involves added headaches and governance issues. Being under the constant scrutiny of a public microscope, public funds are forced to answer a lot of questions. Do you think the general public has an easy time understanding why their retirement money is invested in distressed debt and multi-stage venture capital? If they don’t understand it, they are less likely to trust it. A lack of public trust can create big problems for investment managers.
- In 2013, PRIM kicked off Project SAVE, an initiative to understand, analyze and evaluate its expense structure. I applaud PRIM’s efforts to “recommend the ideal configuration of activities to optimize costs to benefit the plan participants and the taxpayers”  and hope the outcome extends beyond fee negotiations and into simplifying the investment model. There is much room for improvement.
- The battle for basis points is troubling. Even if PRIM and similar funds manage to match or slightly exceed market returns, we shouldn’t ignore the economic cost of doing so. As stated, PRIM employs really smart people. Those really smart people hire lots of really smart money managers. If, for a fraction of the cost, taxpayers and retirees get nearly identical results, what is the cost to society? Anonymous blogger @Jesse_Livermore of Philosophical Economics has raised this issue in the past (Index Investing Makes Markets and Economies More Efficient). Would we be better off if these highly-talented people were deployed in the workforce doing something else? Are valuable resources being wasted? This is a much larger issue that I’ve found myself thinking about more and more lately.
Nevada PERS Investment Information available at:
Massachusetts PRIM Investment Information available at:2015 A Comparative Analysis of Investment-Related Expenses, Massachusetts PERAC  Investment expenses for PRIM are typically listed in the “Management Costs” section of their annual reports.  NASRA Issue Brief: Return Assumptions, February 2016  PRIM targets: standard deviation: 12.6%, Sharpe .40, FY 2015 PRIT Fund CAFR  A Straightforward Structure, Markets Media 2014  PRIM Investor Conference, May 2013