I had the opportunity during the weekend to read the Comprehensive Annual Financial Report (CAFR) for a municipality, which will remain unnamed. These reports provide a lot of really good information.  I always find the Retirement Systems and Deferred  Income Plans section to be particularly interesting, and this report didn’t disappoint.

I was heartened to see that this municipality was participating in a state-sponsored defined benefit system that was actually using a discount rate of their liabilities (5%) that wasn’t equal to the state’s return on asset assumption (7%), and was truly more representative of a legitimate discount rate in today’s interest rate environment.

Where I take umbrage with this analysis is in the long-term “forecast” of asset class returns by either the asset consultant or actuary or both.  I had to laugh when reviewing the data because this (or these) entity (entities) had used 10.63% to represent the expected return for “debt-related private equity”. Buyouts/venture capital had a forecasted long-term estimate of 13.08%. In fact, 13 of the 15 asset class return forecasts were to the second decimal. Are you serious?

It is highly unlikely that these forecasters will get the first number to the left of the decimal place correct let alone get their prediction anywhere close to two decimal places to the right! Was the use of two decimal places supposed to make this a more legitimate exercise to prove that they had some mystical power to forecast what others couldn’t?

Again, managing a pension plan shouldn’t be about creating a return that exceeds some return on asset assumption (ROA), and in this case 7%. Managing a DB plan should be about exceeding liability growth, as that is the only reason why a plan exists in the first place. But, the industry spends nearly all of its time trying to build a better mousetrap when allocating assets and creating an investment structure. Everyone would be much better off in the long run, and there would be far less guesswork if the plan’s primary objective was the plan’s specific liabilities.

Furthermore, the return forecasting exercise, which was for the periods ending both June 30, 2017, and June 30, 2016, had the consultant/actuary use Global Debt ex U.S. as one of the asset class categories that had a 5% target allocation. The long-term forecasted return was -2.5%. Why then would you have this “asset” in your fund?

Finally, can we please stop making this process much harder than it needs to be? Can we stop playing games with long-term forecasted returns to two decimal places as if we have real knowledge of how markets will perform during the next 10-, 20-, or 30-year period? Can we finally agree that the only reason a pension plan exists is to fund a promise that was made to that entities employees? If we can agree to that, can we agree then that the true objective shouldn’t be a made up ROA, but that plan’s specific liabilities?