Just as a trader needs to watch for shifts in market sentiment, he also should watch for shifts in inter-market correlations. Though there are logical fundamental reasons for these correlations, they do shift from time to time. For example, crude oil/Canadian dollar and copper/Aussie dollar often move together and gold/S&P 500 often are correlated negatively. These pairings make sense because the currency-commodity connection is a major driver of market movements. The global economy drives expectations and capital flows.
For example, Canada is an exporter of oil. As a result, crude oil price changes will impact demand for Canadian oil. Gold is a commodity/currency that acts as a safe haven basket and when the equity market tanks or surges, gold should react inversely. The yen also behaves as a safe haven basket to global chaos. As a result it is expected to be correlated inversely to the S&P and directly correlated with gold. However, the problem with these enduring assumptions is that they can be misleading and are too general. Correlations are dynamic and endure many deviations during the year.
Let’s start with the Canadian dollar and oil. In this past year, it has experienced a high correlation of 95% and then proceeded to weaken to a level just above 40% before rebounding. It even went negative for a short period in January and February this year before returning to more normal levels (see “Loonie correlations”).
Gold and the SPX Index provides another example of the potential roller coaster ride for correlations. In May 2011, gold and the SPX had a correlation of 58%. This peak was followed by a descent into extreme negative correlation of –88% on Sept. 21. Once this extreme was reached, the negative relationship weakened and it moved positive to levels near 35% correlation before it came back down to zero (see “Golden fade”).
When we examine the JPY/USD and gold relationship, we can see how gold movements become critical in understanding market sentiment. When the yen increases in strength, so does gold. Both are acting as risk havens to market fear. The JPY/USD-gold correlation reached its highest level of 88% on Sept. 21, 2011 and then proceeded to go to zero in February (see “Yen for gold”).
The deviations are not random; they reflect a discontinuity in economic conditions. When there is deviation, it means something else is happening in the world economy, of which the trader needs to take notice. So when the SPX sells off but gold doesn’t go up, we have a clue that the traditional relationships are being trumped. When the Canadian currency strengthens and so does the price of crude, the trader can go on with traditional strategies. But when it doesn’t, it means there is an opportunity somewhere.
Traders always are looking for an edge to signal a trade and the breakdown in traditional correlations is a powerful signal.Once you can define what is causing this breakdown, you can measure its significance and that sets up a reversion to the mean opportunity.
When the Canadian dollar failed to rally on higher crude oil prices at the end of 2011, you would know it had some built in strength once the impediment — perhaps the Keystone pipeline decision — was removed.
Traders scan the markets for price patterns, examine fundamentals and perform technical analysis on price data. But correlation analysis often is an afterthought; it should be a key part of the process of deciding what to trade. When correlations diverge from expectations, trading decisions need to be reevaluated.
Abe Cofnas is author of “Sentiment Indicators” and “Trading Binary Options: Strategies and Tactics” (Bloomberg Press). He is editor of “Fear and Greed Trader” at Agora Financial and can be reached at firstname.lastname@example.org.