Are commodity markets characterized by herd behaviour?

Abstract

Twenty years ago, Pindyck and Rotemberg concluded that commodity prices exhibited excessive co-movements and that commodity markets were characterized by herd behaviour. The herding hypothesis has recently experienced a revival. A number of studies have concluded that commodities have become 'financialized' and contaminated by the stock market because of the large influx of hedge funds, index trackers and financial investors. Analysing monthly prices of 20 commodities for the period 1986-2010, we find that there has been a tendency toward increased co-movements across commodities and between commodities and the stock market after 2004. However, this result is mainly driven by the extreme price movements after 2008. There is no strong evidence of financialization or contamination from the market activities of financial investors prior to 2008.

Does the stock market drive herd behavior in commodity futures markets?

“…Similarly, Boyd, Buyuksahin, Haigh, and Harris (2009) examine trading data and find that herding among hedge funds does not destabilize the crude oil market. More recently, Steen and Gjolberg (2013) examine the correlation patterns and principal components describing commodity returns and find no significant support for herd behavior.…”

Commodity value-at-risk modeling: comparing RiskMetrics, historic simulation and quantile regression

“…In the latter paper, they found that skewness and kurtosis have not changed significantly either. In another paper, Steen and Gjøberg (2013) found no support for commodities having turned into "one" asset.…”

Commodity market risk from 1995 to 2013: an extreme value theory approach

“…The authors argue that owing to the passive and long-only nature of commodity index investments, these are unlikely culprits of inflated commodity prices. Steen and Gjølberg (2013) There is also a growing body of literature that addresses the issue of increasing commodity market volatility. McPhail et al (2012) study corn futures traded on the Chicago Board of Trade and use a structural vector autoregressive model and variance 310 decomposition to analyse corn price volatility.…”