1. The Insurance Information Institute’s (III) Opinion Research Corporation International poll estimates that 95% of homeowners in the United States have homeowners’ insurance. More information on how the III arrived at the 95% figure is available here.
2. We attribute this metaphor to Irwin (2018).
3. Arnott (2012) provides more information about the potential flaws of this standard de-risking approach.
4. According to the Morningstar report by Holt (2017), these fund companies and their TDF assets as of year-end 2016 include Vanguard ($280 billion), Fidelity Investments ($193 billion), T. Rowe Price ($148 billion), American Funds ($54 billion), JPMorgan ($45 billion), TIAA-CREF Asset Management ($31 billion), Principal Funds ($26 billion), American Century Investments ($17 billion), John Hancock ($16 billion), and BlackRock ($12 billion).
5. Based on a holdings analysis in FactSet as of May 31, 2018, 91% of the iShares MSCI ACWI Index is in developed equities (54% in US equities and 37% in developed ex US equities).
6. This observation is not unique to the 2030 vintage. Across all vintages, TDFs are heavily allocated to developed-market stock and bond markets, exhibiting a large home-country bias.
7. The pass-through of rents is not 100% as some slippage from maintenance costs and so forth occurs.
8. From March 1997 to March 2018, the quarterly correlations of returns to US inflation were 50% for bank loans, 31% for EM currency, 28% for high-yield bonds, and 21% for EM equities. These levels exceed or are comparable to the correlation of US TIPS return to inflation, which is measured at 22% over the same span.
9. Inflation is measured as the 12-month percentage change in the US CPI City Average Seasonally-Adjusted Index, as provided by the US Bureau of Labor Statistics (BLS).
10. West and Ko (2017), the first of eight articles in a series by Research Affiliates focused on the needs of financial advisors, provides more details about the impact of using historical returns to forecast the future.
11. Pollock (2018) provides more information on some TDFs’ boosts in stock exposure.
12. Consistent with the analysis by Johnson (2012), inflation surprises reflect the difference between actual inflation at the end of the quarter and expectations for inflation at the start of the quarter, so that a “high (low) inflation surprise” is a quarter that falls in the top (bottom) third of all historical inflation surprises from June 1973 through March 2018.
13. This basket is an equally weighted mix of traditional and stealth inflation-fighting asset classes, which are added into the mix in the first month the returns are available. The asset classes, representative indices, and dates of available data are as follows: 1) High-yield bonds, IA Barclays US HY Corporate Bond Index (Jan 1973–Dec 1992) and BofA ML US High Yield, BB-B Rated, Constrained Index (Jan 1993 onward); 2) Commodities, S&P GSCI TR Index (Jan 1973–Dec 1990) and DJ UBS Commodity TR USD Index (Jan 1991 onward); 3) REITs, FTSE REIT All REITs TR Index (Jan 1973); 4) Bank loans, CSFB Leveraged Loan Index (Jan 1992); 5) EM equities, MSCI EM GR USD TR Index (Jan 1988); 6) EM currencies/local bonds, JPM ELMI Index (Jan 1994–Dec 1993) and JPM GBI EM Global Diversified TR USD (Jan 1994 onward); and 7) US TIPS, Barclays US Treasury US TIPS TR USD Index (Mar 1997–Jun 1998) and Barclays Capital US Treasury Inflation Notes: 10+ Year Index (Jul 1998 onward).
14. We acknowledge that rolling 12-month volatility exceeding 30% is extreme. Since 1973, such bouts of severe turbulence occurred slightly less than 5% of the time, notably in April–May 1975, October 1987–September 1988, and March–September 2009. When we relax the threshold and use 12-month volatility of 20%, we find that highly volatile markets occurred 15% of the time from January 1973 to March 2018. Over this period, the average 12-month rolling return is 1.5% for the S&P 500, 3.2% for a developed 60/40 portfolio, and 4.3% for the inflation-fighting basket. Therefore, at the 12% level, inflation-fighters typically outperformed conventional assets, although the latter did deliver positive, albeit average returns. In the section “When to Insure?” we use the 12% threshold to characterize calm versus volatile markets.
15. This calculation is based on rolling 12-month returns of the US stock market from 1802 through March 2018. The duration of a market rally is the total number of consecutive positive rolling return periods without a gap exceeding two rolling periods. For example, the length of the latest rally is 9.25 years or 99 12-month rolling outcomes (99+12)/12. Within this span, the longest time during which the rolling 12-month return was consecutively negative was two periods (the 12-month periods ending January 2016 and February 2016); that is, the largest gap is two, and therefore this period is captured within the latest rally episode. The sources of data used to compute the US stock market are Ibbotson/Morningstar, Robert Schiller’s Online Data, and William Schwert’s Indexes of US Stock Prices.
16. Aside from these excessively stretched conditions, we believe reversal risks are skewed to the upside for other reasons. First, the United States is emerging from a period of highly irregular monetary policy, which was designed to bring about market and macroeconomic stability and keep interest rates low (i.e., bond prices high). With the removal of such accommodative policy, market and macroeconomic volatility should naturally rise as the Fed “put” is stricken further out of the money. Second, the US Federal Reserve’s Federal Open Market Committee’s dual mandate of full employment and price stability skews inflation risks to the upside. From 2012 to mid-2018, the Fed’s preferred measure of inflation—the Personal Consumption Expenditures (PCE) inflation rate—has fallen below its long-term 2% target nearly 95% of the time (73 of 77 months). In light of this persistent undershooting and the recent emphasis on this target’s symmetrical nature, the Fed will likely tolerate modest overshooting of the PCE inflation rate even after having met the 2.0% PCE objective in June 2018. Third, the Fed isn’t the only entity seeking higher inflation. Because fiscal austerity measures are unpopular, governments are also incentivized to create higher inflation in an effort to reduce the real value of their debt burdens. In the United States, late-cycle fiscal stimulus and trade tariffs only increase inflationary upside along with growing debt and deficits. Finally, decomposing the components of the core inflation basket also points to a continued upside for inflation. US dollar depreciation in recent years may result in flattening (instead of falling) import prices, causing the core goods component of inflation to be less of a drag on realized inflation moving forward. On the core services front, a negative unemployment gap indicates higher upside potential to services prices as wage pressures loom amid an economy with an unemployment rate that, since early 2017, has fallen below NAIRU (the non-accelerating inflation rate of unemployment).
17. Shepherd (2017) provides more information on the volatility characteristics of commodities and REITs.