Friday, November 04, 2022

Rudy Fichtenbaum: STRS needs to change its investment philosophy and move toward index investing.

 Through the STRS Looking Glass


Rudy Fichtenbaum

November 3, 2022

Last month, the day before the Board meeting, I sent an email to Board members in preparation for the Investment Committee meeting. On the agenda for that meeting was a discussion of STRS’s investment philosophy. Staff prepared materials with Callan’s assistance in which they asserted that our philosophy supports active investment. STRS’s “belief in active investment” is based on the claim that active investment outperforms passive investment. The investment staff and our consultants constantly talk about the value added by active investing.

The only verifiable benchmark that STRS uses is for U.S. equities. Beyond that all the other benchmarks are custom, and none of them get audited. The international equities are 50% hedged; without daily data on hedging, you cannot verify performance. Fixed income uses daily weights that only STRS has and cannot be verified. Real estate uses an index that does not include leverage. For alternatives, STRS has been using its own performance until this year. Look at the reports for Cliffwater, and you will see that they never beat the Russell 3000 +1% for private equity and Russell 3000 -1% for Opportunistic Diversified, which used to be the benchmark. Now STRS uses the Cambridge “indices”, which are not real indices because, for example, there is no data publicly available on say the largest 1000 private equity deals. What is being constructed by Cambridge is an “index” based on peer reporting.

I attached a spreadsheet to my aforementioned email to Board members that used calendar year data for the S & P 500 and a Bond Portfolio which uses 10-year Treasurys, Baa Corporate Bonds and 2-month T Bills with various weights for the bonds. I noted the sources of my data and explained in a fair amount of detail how I constructed 6 different 60/40 stock-bond portfolios using three different mixes of bonds. Each portfolio started with a 60-40 mix of stocks and bonds using three different mixes for bonds. Three portfolios started at 60-40 and changed through the year, ending at 70-30, which is about where STRS is today in terms of risk. The other three just stayed at 60-40.

Over 30-years, STRS loses to all six of these portfolios. The losses range between 44 and 94 basis points (bps) on an annual basis. (A bps is 1/100th of a percent, so 44 bps is 0.44%.) That is real money. The bottom line is that between 1992 and 2021, STRS did not beat a 60/40 portfolio of stocks and bonds i.e., a real passive benchmark. Moreover, even when the returns of these six portfolios are adjusted for risk, all six still outperformed STRS.

In 1992 STRS had about $30 billion in assets. My spreadsheet also showed the money lost over 30 years, all other things equal, was between $48 billion and $109 billion. Of course, I told the Board that I realized that all other things had not remained equal i.e., the amount of money we have in assets is a function of the rate of return as well as the burn rate -- the amount by which payments to retirees exceeds contributions. In fact, I would argue that the burn rate is our biggest problem. So, the calculation of money lost is meant to illustrate the magnitude of the losses.

The bottom line is STRS between 1992 and 2021 did not beat a 60/40 portfolio of stocks and bonds i.e., a real passive benchmark. Moreover, each of these portfolios have risk adjusted returns that exceed that of STRS.

Like most pensions, STRS beats the benchmarks it constructs. But it does not beat a real benchmark, i.e., a passive benchmark that is investable and transparent.

This is a classic example of the principal-agent theory in economics where staff and consultants benefit from active investing by creating benchmarks that they can beat, allowing staff to get bonuses while consultants get their contracts renewed. They rely on complexity and the fact that most of us are not investment experts. I believe most members of the Board have good intentions. Unfortunately, as the old saying goes, the road to hell is paved with good intentions.

I am not alone in believing that STRS does not beat a passive benchmark.

Richard Ennis just sent me a paper entitled "Lies, Damn Lies and Performance Benchmarks: An Injunction for Trustees.” It will appear on LinkedIn, SSRN (an outlet for scholarly research before it appears in scholarly journals) and eventually appear in the Journal of Investing. For those of you not familiar with Richard Ennis, here is a short bio that appears on his blog.

Richard M. Ennis, CFA, managed money at Transamerica and pioneered quant investing in the early 1970s. He helped create the field of institutional investment consulting at A.G. Becker & Co. Richard co-founded EnnisKnupp, the first consultancy to be recognized as a professional services firm. During his career Ennis received lifetime achievement awards from CFA Institute and Investment Management Consultants Association. His research won Graham & Dodd and Bernstein Fabozzi Jacobs Levy Awards. He edited Financial Analysts Journal. He and his wife, Sally, retired to Sanibel Island, Florida, upon the sale of his firm to Aon in 2010.

One of the main points that Ennis makes in this paper is that most pensions report outperforming their benchmarks, but they do not use real benchmarks to measure their performance. They use custom benchmarks that exhibit benchmark bias. In his paper he explains a passive benchmark must be investable and fully transparent.

Richard Ennis has also done an analysis of STRS’s performance constructing benchmarks used in the article "Cost, Performance, and Benchmark Bias of Public Pension Funds in the United States: An Unflattering Portrait”. In this article Ennis constructs benchmarks for each pension he looks at using three indices: U.S. Stock, non-U.S. stock, and an aggregate U.S. bond index. The methodology he uses was developed by William Sharpe, Emeritus Professor of Finance and Stanford and winner of the 1990 Nobel prize in economics. Although STRS OH was not included in the article, in an email to me he confirmed that he had looked at STRS performance for 10 years and found that STRS underperformed a passive benchmark by 81 bps.

Finally, my email stated that STRS has not had a successful active management program, which is not surprising given the overall track record of active managers. Without some sort of an information edge, beating real benchmarks is basically flipping a coin, except you must pay to play – which is why most active managers lose to an index.

I concluded that STRS needs to change its investment philosophy and move toward index investing.

You would think that after receiving this email that there would have been a robust discussion of my spreadsheet and email. And yet the content of my email and spreadsheet was not discussed at all. Not one Board member said, “your analysis is wrong” or “there is a mistake in your spreadsheet.”

Instead, a majority of Board members focused their attention on whether my sending this email to the Board was a violation of the open meetings law. After a lot of back and forth I think the conclusion was that if members did not engage in discussion via email, there was no violation of the open meetings law. So, then the discussion turned to saying, but there could have been a violation if members had responded, although nothing in my email invited a response. The whole point of the email was to provide food for thought so we could have a robust discussion of our investment philosophy. Then I was admonished by Board members for not having sent my email to the Chair.

The whole experience is akin to Alice in Wonderland. Under the status quo, we have a pension where there is no prospect in my lifetime of restoring a permanent COLA for retirees and active members, and no chance of returning to 30 years for unreduced benefits. But don’t worry -- the staff will continue to get their bonuses for beating the benchmarks they create.

Change requires 6 votes. That means that if you want a COLA and you want a chance of being able to retire after 30 years with unreduced benefits, then you need to convince an existing board member to do the right thing and vote for what is in the members’ best interests or elect a new Board member for the active seat in 2023.

Dr. Rudy Fichtenbaum is Professor Emeritus of Economics at Wright State University. He is an elected member of the STRS Ohio Board, filling a retiree seat since September 2021. He and Board member Wade Steen have been outspoken in pushing for reform on behalf of both active and retired teachers, especially in their opposition to Board members who have followed the dictates of another group instead of advocating solely in the interest of STRS stakeholders. As of September 1, 2022, there are three additional Board members who strongly support their efforts: Liz Jones (retired), Julie Sellers (active) and Steve Foreman (active). All three were elected to replace incumbents in the spring 2022 STRS Board election.

Larry KehresMount Union Collge
Division III
web page counter
Vermont Teddy Bear Company