By Barry Ritholtz

The popular definition is a 20 per cent drop from peak to trough in multiple broad market indexes. I have seen similar round numbers in other definitions -- e.g., a correction is a 10 per cent decline, a dip is a 5 per cent drop and a crash is a fall of 30 per cent or more.

Other than the fact that these are all base 10 numerals -- a coincidence of primates having 10 fingers and 10 toes -- there is no rational basis for these percentile heuristics. There certainly doesn't seem to be any hard data supporting the significance of these percentages.

Now consider the opposite: The definition of a bull market is a 20 per cent rally from the lows. Why 20? Why not 25 per cent or 30 per cent or perhaps 21.759 per cent? The origins of these numbers have been lost to history, but the bigger question for investors is: Are they useful? Do these definitions assist in managing risk, deploying capital or even in thinking about market cycles?


My answer is a definitive no.

Just consider the depth of the 2015 correction peak to trough: The Standard & Poor's 500 Index fell 15.2 per cent, while the Russell 2000 Index lost 27.2 per cent. Was it helpful to know the S&P 500 wasn't in a bear market, but the small-cap Russell was? Recall the deepest correction since March 2009, the May-October 2011 slide: It was a 21.58 per cent peak-to-trough decline for the S&P 500, while the Russell 2000 fell 30.7 per cent. What should you have done knowing the S&P 500 was in a bear market and that the Russell 2000 had crashed?

What good did it do investors to know that a bear market had begun based on the traditional definition?

These definitions are pointless, unable to help investors in any meaningful way. They don't assist in managing risk; they don't inform as to when or how to deploy capital. At best, they may reveal what some other investors, similarly relying on meaningless numbers, may believe. It seems to be one of those trading myths that get passed along from generation to generation, with no one considering whether it has any actual validity.

These market definitions are deeply unsatisfying.

I have been toying with better ways to define markets for 20 years. In the early 2000s, I began to consider a different set of definitions for bull and bear markets. The idea was to create something useful that I could use as an investor, reflecting what was actually occurring during those longer market trends. I found it practical to start with the market gains or losses, then add equal parts long-term economic trends and investor psychology -- specifically regarding valuations -- to the equation.

Thus, my definitions of bull and bear markets are as follows:

Secular bull market: This is an extended period of time, typically 10 to 20 years, driven by broad economic shifts that create an environment conducive to rising corporate revenues and earnings. Market volatility tends to decrease. Its most dominant feature is the increasing willingness of investors to pay more and more for a dollar of earnings as the bull market progresses.


Secular bear market: This reflects the opposite: After an extended secular bull run, it is a period marked by increased volatility, frequent cyclical rallies and retreats in an economically challenging environment. The dominant feature is that investors become less and less willing to pay for that same dollar of earnings.

The two factors missing from the percentage-only definition are the broader secular underpinnings and the idea of earnings multiple expansion or contraction.

As I noted in November before about secular economic cycles:

"Waves of industrial, technological and economic progress make their way into employees' wages, consumers' pockets and corporate profits. Improving standards of living are reflected in the psychology of an era. Not surprisingly, markets do well, as investors become willing to pay more for a dollar of earnings as the cycle progresses. Multiple expansion, in the form of rising price-to-earnings ratios, drives returns even more than rising profits."

Hence, these are not short-lived and modest phenomena, and they instead represent significant society-wide changes. Think about the 20 years after World War II, or the tech era of the 1980s and 1990s. Both were 20-year-long booms with similar characteristics.

It is also why I have repeatedly argued that it isn't the valuation of markets that is so important, but rather, which direction earnings ratios are moving.


When we look at the sources of market gains, earnings improvements are often a much smaller factor than multiple expansion. By my estimates, three-quarters of the gains of the 1982-2000 bull market may be attributable to rising price-to-earnings ratios. At the start of that bull market, the S&P 500 earnings in inflation-adjusted dollars for the year were $31.72 as of Dec 31, 1982; 18 years later, at the end of the bull market, they had more than doubled to 70.39. But during the same period of time, the broad indexes gained 1,000 per cent. The P/E ratio for the index was about seven at the start of that cycle, and ended at about 34. Most of the market's gains were attributable to the psychology of paying more for the same dollar of earnings, not rising corporate earnings.

People who are perplexed by a market that keeps rising -- despite the usual wall of worry -- should look at the psychology underlying this expansion. Bear markets begin when this psychology eventually begins to shift. So pay attention to what is actually driving markets in this period, and not the pundits.

Barry Ritholtz is a Bloomberg View columnist. He founded Ritholtz Wealth Management and was chief executive and director of equity research at FusionIQ, a quantitative research firm. He blogs at the Big Picture and is the author of "Bailout Nation: How Greed and Easy Money Corrupted Wall Street and Shook the World Economy."