Les spéculateurs ne créent pas nécessairement des bulles spéculatives…

Les travaux menés par Christopher L. Gilbert permettent d’identifier :

• trois bulles sur le marché du cuivre en 2004, 2006 et 2008 : « The first group consists just of copper which stands out with three clear bubbles identified in 2004, 2006 and 2008 respectively. The third of these bubbles, which is associated with rapidly falling prices, spans the end of the sample. The 2006 bubble, which correspond with that identified in the monthly data, is the only completely unproblematic case in that the estimated bubble period does not contain any holes ».
• une bulle sur le marché du soja au début de 2008, mais aucune sur celui du pétrole brut ou du nickel : « The second group consists of crude oil, nickel and soybeans. A bubble is identified in the soybean market, in this case at the start of 2008, while bubbles are not found for crude oil or nickel. Nevertheless, all three outcomes are marginal and would be altered by small changes in the test critical values. »
• ni sur les marchés de l’aluminium, du blé ou du maïs : « T he remaining group consists of aluminium, corn and wheat. Here the test outcomes show clearly that there was no bubble ».