Are diamonds useful as investment portfolio diversifiers? In their draft papers entitled “An Examination of Diamonds as an Alternative Asset Class: Do They Have What it Takes to Make a Portfolio Sparkle?” of June 2012 and “The Return Characteristics of Diamonds” of July 2012, Kenneth Small, Jeff Smith and Erika Small investigate diamonds as an asset class (returns and correlations of returns with other asset classes). They consider price histories of an aggregate diamond index and of indexes for several classes of diamonds segregated by quality and size. They relate diamond index returns to those for U.S. equities, world equities, emerging markets equities, long-term U.S. corporate bonds and precious metals. Using weekly levels of diamond and other asset class indexes during 2001 through 2011, they find that:
- On a risk-adjusted basis over the sample period, diamond indexes outperform U.S. stock market and long-term U.S. corporate bond indexes, but not a world stock index. Specifically:
- The aggregate diamond index gains 38% (3.2% annualized), about three times the gain of the S&P 500 Index. Annual volatility of the aggregate diamond index is about half that of the S&P 500 Index.
- High-grade diamonds generally outperform low-grade diamonds, with the one-carat flawless (fine) diamond index generating an annualized return of 6.4% (5.2%). The annual volatility of the one-carat flawless (fine) diamond index is about twice (a little higher than) that of the S&P 500 Index.
- Diamond index weekly returns exhibit low correlations with U.S. inflation, the S&P 500 Index, the long-term U.S. corporate bond index and precious metals over the sample period. Specifically:
- The overall correlation between the aggregate diamond index and the S&P 500 Index is 0.06, with rolling 52-week correlations ranging from -0.18 to 0.27. Over the entire sample period, rolling correlations show little or no trend. Results are roughly similar for the world equity index, emerging markets equity index and long-term U.S. corporate bond index.
- The overall correlation between the aggregate diamond index and gold is -0.01, with rolling 52-week correlations ranging from -0.32 to 0.24. Results for silver and platinum are roughly similar.
- The Fama-French three-factor model has little explanatory power for diamond indexes.
In summary, limited evidence from index levels suggests that various classes of diamonds are potentially good diversifiers of stocks, bonds and precious metals.
Cautions regarding findings include:
- As noted in the papers, the sample period is very short for return comparisons and correlation analyses.
- The studies employ indexes rather than tradable assets, thereby ignoring the costs of constructing and maintaining index funds. These costs may be high for diamonds due to limited market depth and high trading frictions, such that funds would materially underperform indexes.
- Due to high trading frictions, investors would have to hold diamonds for a very long time to approach gross class return (tactical allocation adjustments would be very costly).
- Holding a diversified portfolio of physical diamonds is impractical for most investors. An exchange-traded fund such as PureFunds Diamond/Gemstone ETF (GEMS) offers access to retail investors, but this fund does not have enough history for inference.
