Get Over It: 5% Is All You Can Expect
Dave Lawson on why most other forecasts are bunk.
January 26, 2012
This time of year, we are overloaded with forecasts and prognostications about the upcoming year. Global growth will be x%, equity markets will be up y% with lots of volatility, etc, etc. The amount of time and energy that is spent on forecasts that have very little use to any real investor is mind boggling.
It isn’t that forecasting returns is a useless exercise. The problem is that the time horizon forecasted is too short to be of much use. CNBC and Bloomberg are focused mostly on days, weeks and maybe months at the longest. The long term forecasts directed at institutional investors typically have an annual time horizon. These forecast documents can be interesting to read, but annual is still too short a time horizon for the information to be put to any productive use.
Five years or ten years: now those are useful time horizons. Not necessarily useful from the perspective of tactical asset mix and beating benchmarks, but as inputs to decision-making on more important issues. A realistic and informed projection of returns for the next decade can be a useful support for decisions about plan design, funding and also asset mix and investment strategy.
In terms of funding and plan design, sustainability and affordability are top of mind issues. Whether discounted at bond yields or expected total fund returns (a debate for another day), the principle question is the same: Will the contributions and investment returns provide enough money to pay pensions, or whatever liability the fund is obligated to pay?
When I look at longer term return forecasts, I am more inclined to give credit to those based on fundamentals. I have less time for forecasts based on opinions, historical returns or a historical equity risk premium. With fixed income, the current yield to maturity is the logical starting point and one can reasonably expect the yield to mean revert to something close to long term nominal GDP growth over time.
Equity return expectations can be built based on current yield, potential GDP growth based earnings growth and reversion to a mean P/E ratio. This type of analysis, performed rigorously by industry participants, comes out with 10-year returns of 2% to 3% for fixed income and 7% to 8% for equities. A 50:50 asset mix gives an expected return of 5% nominal.
Two things jump out from this analysis:
1. 5% nominal may not be enough return under current scheme arrangements.
2. Equities should outperform bonds by a wide margin over the next decade.
The first challenge is acceptance: get over it. 5% nominal is a reasonable expectation. If it is not enough to pay the bills, then something has to change. Pay out less, collect more, and/or take more risk.