The importance of sector trends has long been recognized as a key investment factor by market participants; "a rising tide lifts all boats". It is nevertheless only since the late 1990s that empirical research has emerged supporting this previously poorly evidenced investment tenet.
Sectors are obviously comprised of individual equities that will individually broadly under- or outperform respective benchmarks.
Conventional economic theory argues that stock market prices are unpredictable and follow a "Random Walk". This concept was extended to form a cornerstone of the "Efficient Market Hypothesis" (EMH) formulated by Eugene Fama in 1970.
However, there is now an overwhelming body of evidence documenting the failure of the Random Walk assumption to account for observed price momentum.
In particular, an important 1993 study by Jegadeesh and Titman demonstrated that stock returns exhibit momentum behavior at intermediate horizons (3-12 months). This effect is known as the "Momentum Anomaly", and has subsequently been backed up by a number of additional studies, including several compelling pieces of research from behavioral finance proposing models of systematic departures from rational behavior by traders to account for this.
Building on this body of work Tobias Moskowitz and Mark Grinblatt published their landmark study, "Do Industries Explain Momentum?" in the August 1999 Journal of Finance.
The answer to the question posed was a resounding 'yes'. These findings also partly allowed market rationalists to counter behavioral explanations for the Momentum Anomaly; sector/industry constituents operate under identical market dynamics and regulatory environments, and tend towards similar capital structures, etc. As such there are natural fundamental reasons why stocks in industry groups should move together.
Moskowitz and Grinblat found that:
Following the publication of the Moskowitz/Grinblat research a number of studies emerged tending towards a consensus estimate that approximately 50 percent of investment returns are attributable to sector factors.
In time strict rationalists fought back arguing mainly transaction costs and market impact made momentum strategies prohibitively expensive for institutional investors, albeit without disputing the existence of the Momentum Anomaly per se.
Although these objections appear flawed, we nevertheless consider estimates in the range of 50 percent of returns attributable to sector factors excessive in the recent market environment, where long term cross-sector correlations have been high in the period from mid 2003 to late 2007 after hitting a cyclical trough in 2001 in the wake of the TMT boom and bust.
As of late summer 2008 there is increasing evidence that long term cross-sector correlations are breaking down again, and they certainly exhibit cyclical behavior. In any case, regardless of the precise magnitude of sector factors they undoubtedly warrant close investor scrutiny.
To learn more about MarketScalpel's unique tools and visualization features for monitoring and anticipating sector rotation and trends please visit the Market Navigator online research platform section, which includes a full demonstration with stale data, or alternatively contact us directly.