Compare 10-Year Projections Using Historical and Forecasted Returns

Institutions generally agree that total returns over the next 10 years will be lower than the long-term historical level:

  • 5.42% mean for US large-cap stocks vs 11.92% from January 1987 – April 2018
  • 3.19% mean for Aggregate US bonds vs 6.07% from January 1987 – April 2018.

Forecasted returns used here are averages of forecasts by BlackRock, State Street Global Advisors, JP Morgan, Bank of NewYork/Mellon, Callan Associates (pension consultants) and Research Affiliates.

Considering mean return history or forecasts is not adequate for setting portfolio expectations, because future results have a wide spread of possibilities around the mean due to the impact of volatility (often made worse by investors selling in panic at bottoms and re-entering late in Bulls).

This table shows simulated probabilities for a $1,000,000 bonds or stocks portfolio at the 10th, 25th, 50th, 75th and 90th percentile probabilities (covering 80% of likely outcomes, but still leaving 10% more extreme possibilities at either end of the spectrum undefined).

(click images to view full size)

This table shows four common allocations: 40/60 (conservative balance), 50/50 (allocation Vanguard uses in their target date funds for investors age 65 just beginning retirement), 60/40 (traditional balanced fund allocation) and 70/30 (aggressive balanced allocation).

If the institutions are correct in their assumptions, you should expect lower returns, and lower cumulative values for your portfolio over the next 10 years. The differences in cumulative portfolios per million Dollars of starting capital between simulations based on historical data and forecasted data are in the hundreds of thousands of Dollars.

For example, per $1,000,000 for a 50/50 portfolio allocation at the 50th percentile simulation probability, you should expect an inflation adjusted (real) portfolio value at the end of the next 10 years to be about $624,000 smaller (about 35% less) based on forecasted returns and volatility rather than based on historical returns and volatility. Maximum drawdowns are expected to be similar.

This table shows the actual total returns of US large-cap stocks, US aggregate bonds and nine allocation levels between them over a variety of periods all ending at 12/31/2017.

This table based on daily prices shows the rolling period price returns (not total returns) ending on all market days since the beginning of 1928. All the data is for actual results — no theory or hypotheticals here. There were many very good and very bad rolling period returns.

Bottom line – a simplified look at historical mean returns all ending on a recent day, and not understanding how volatility creates a wide spectrum of possible outcomes and occasional Maximum Drawdowns is not a safe way to look at what may occur in the future. Your allocation decision is critical – more critical than your choice of specific securities – in determining the likely range of return outcomes and the severity of likely Maximum Drawdowns.

Presuming you make reasonable choices of securities and have a diversified portfolio, those decisions will have far less impact on your overall outcomes than your Own / Loan / Reserve allocation decision.

And, whatever your Own / Loan / Reserve allocation decision, the predominate institutional opinion is that returns are likely to be lower over the next 10 years than the last 10, 20 or 30 years.

As you can see in the simulation tables, the lowest projected returns are also paired with the largest Maximum Drawdowns. Minimizing Maximum Drawdown exposure is almost synonymous with maximizing return. Selecting a static allocation is implicitly selecting a likely Maximum Drawdown exposure.

There are two ways to minimize Maximum Drawdowns, not involving derivative products or shorting:

  • Select a more conservative allocation
    • Requires lower overall return expectations
    • Rebalancing maintains risk level, but does not increase return
  • Shift between more aggressive and more conservative allocations as risk levels change
    • Requires a rational signal system for increasing and decreasing risk exposure
      • Trend following approaches are superior to trend prediction approaches
    • Shifting will experience false positives that drag on performance during Bulls
    • Generates tax costs in taxable accounts that reduce return benefit
    • Shifting as risk levels change can avoid the largest part of Maximum Drawdowns
    • Missing the largest part of Maximum Drawdowns can increase returns
    • Requires active oversight and time commitment.

Think about these historical and projected returns, and how you are most comfortable with managing risk levels.






Comments are closed.