Diversification is being blamed for a failure it doesn't own.
Investors generally assumed that by spreading their investments across many asset classes — by investing in a wide array of betas — they would avoid the risk of the sort of across-the-board decline in their investments experienced in the recent credit crisis. This has left many questioning the practical merits of diversification, with some deeming diversification to have failed and others throwing their hands up in frustration about the way correlations converge toward one during periods of market stress.
Common wisdom often fails us in the realm of investing, and in this case it is the deeply ingrained association of diversification with the adage — “don't put all your eggs in one basket” — that leads us to rely on false assumptions. The risks associated with assets and asset classes cannot be separated and placed in independent baskets as one might do with eggs. Diversification, in the context of investing, cannot serve to protect the whole by separating the parts. The fact that investors assume it does lulls them into a false comfort that leads to disappointment and deters them from looking for tools that do aim to protect the whole.
What is it that diversification does do then? Recall there are two categories of risk, idiosyncratic and systematic. Idiosyncratic risk is, by definition, risk specific to a particular asset or group of assets. Idiosyncratic risk can be diversified away by combining it with a large enough array of other idiosyncratic risks. That is diversification — a destructive rather than protective process aimed at dissolving idiosyncratic risks. We seek to dissolve or destroy idiosyncratic risk because we cannot expect compensation from a risk that can be diversified away.
Having rid the portfolio of undesirable, uncompensated idiosyncratic risk, what remains? We know the remaining systematic risks won't have natural offsets or hedges, i.e., we cannot diversify them away.