Jason Kelly on Market Timing
December 13, 2005 • Posted in Individual Gurus
Guru Grades ranks a group of 29 stock market experts according to our assessments of the accuracy of their stock market forecasts. Since Jason Kelly has been at or near the top of the list, we asked him to encapsulate his thinking on stock market timing in a short piece for this blog. What makes him get in, and what makes him get out? Jason graciously agreed. Here is Jason Kelly on market timing:
Stock Market Timing
by Jason Kelly
When Steve LeCompte at the CXO Advisory Group LLC pointed me to my #1 position on the list of tracked stock market gurus, I was pleased. When he then asked me to explain how I successfully forecast the direction of the market, I was mortified.
Why? Because anybody who’s been an investor for more than an hour knows the futility of market timing. It leads to short-term trading, about which Jack Bogle, the founder of Vanguard and long-time table-pounder for index funds, said, “After nearly 50 years in this business I do not know of anybody who has done it successfully and consistently. I don’t even know anybody who knows anybody who has done it successfully and consistently.”
Probably the best observation on the future of the market was provided by J.P. Morgan. When asked which way the market would go, he replied, “It will fluctuate.” That’s about all there is to say and it’s never wrong.
With those caveats duly stated, let me share techniques that have kept me on a steady upward path, even though failing at several points along the way. Keep firmly in mind that nobody knows where the market goes. Neither I nor anyone else in Guru Grades can tell you precisely where the market will be at any point in the future or what will happen to it between now and then. All we can hope to do is put our finger on a general trend, thereby raising the odds a little and putting a range on risk. Sound like I’m hedging? I am.
Having some idea of the future direction of the market is important. While it’s possible to cherry pick your way to positive returns in a falling market, it’s a lot easier to win in a rising market. Investors must therefore think about the future of the market. I describe it in The Neatest Little Guide to Stock Market Investing as a moving target on a shifting field on a cloudy day.
How can we stalk this most elusive quarry, the future of the market?
My first step was to understand the news cycle. It’s helpful to see how the market responded to news in the past. I know, past performance is no guarantee of future results. The past is an imperfect guide, but it’s our only guide.
Over the years, I have assembled a collection of newspaper clippings and magazine articles and kept tabs on how the stock market behaved at the time. I find reading current news stories next to old news stories on similar topics to be revealing because they’re often very similar. The “very similar” part is key. Mark Twain said that history doesn’t repeat; it rhymes. It’s never exactly as it was before, but it’s pretty close.
For instance, we have hurricanes every year, and the press always reports with roughly the same urgency seen for Katrina and Rita a few months ago. Think back to Hurricanes Andrew, Charley, Frances and Mitch. They even hit the same places year after year, but those who don’t understand the news cycle continue to panic year after year.
It’s a continuing source of amusement to me that seasonal weather patterns make news. To say that a massive winter storm looms is almost as un-newsworthy as announcing that “Winter Arrives — Again.” Yet, it happens every year, with the press dutifully mentioning the rising cost of heating oil and dolefully questioning whether the consumer will freeze spending for other things. The stock market reacts with a spill for one or two days before moving on to other issues. Miraculously, warmer weather arrives in spring. Then summer surprises with higher gas prices, and the stock market takes another breather…until autumn descends with earnings reports not nearly as bad as everybody expected after the “awful” summer, and we get a rally into winter.
Skeptical about the cycle of news? Look back at, say, 1998. What was the news just before the Lewinsky scandal broke in January? A massive ice storm strikes New England, resulting in a widespread loss of power and several deaths. Let’s review the rest of that year to see what’s familiar:
In February, the Senate passes resolution 71 encouraging President Clinton to take all necessary action to respond to the threat posed by Iraq’s weapons of mass destruction programs. Heard anything along those lines since? Also in February, an earthquake in Afghanistan kills 5,000.
In April, Citicorp and Travelers merge to create the largest financial services conglomerate in the world, Citigroup. Ever notice that the latest merger is often the largest in some way? Also in April, a tornado strikes Nashville, the first in 11 years to make a direct hit on a major city. Weather again.
In May, India tests underground nuclear weapons and Pakistan responds with its own explosions. The United States, Japan and other nations impose economic sanctions. War again. Also in May, another earthquake in Afghanistan kills another 5,000.
In June, an intestinal viral outbreak in Taiwan kills 31 children and infects 10,000 more. Health officials say there is no vaccine for the virus. Hundreds of panicked parents jam emergency health hot lines. Bird flu, anyone?
In July, a tsunami destroys 10 villages in Papua New Guinea.
In August, U.S. embassies in Kenya are bombed by a group linked to Osama Bin Laden. The U.S. responds by pounding targets in Afghanistan and Sudan with cruise missiles. Notice anything familiar?
In September, MCI and WorldCom merge to create MCI WorldCom.
In October, Iraq announces that it will no longer cooperate with U.N. weapons inspectors. At the end of the month, Hurricane Mitch slams Honduras, Nicaragua, Guatemala and El Salvador, killing more than 10,000.
In November, Daimler-Benz completes a merger with Chrysler to form Daimler-Chrysler. Also in November, President Clinton orders air strikes on Iraq, calling them off at the last minute when Iraq promises to “unconditionally” cooperate with inspectors. Inspectors return to Iraq but at the end of the month report that Iraq has once again stopped cooperating. Finally in November, America Online announces its acquisition of Netscape Communications.
In December, Exxon announces a $73.7 billion deal to buy Mobil, thus creating Exxon-Mobil, the largest company on the planet. Also in December, Iraq announces that U.N. weapons inspections will no longer take place on Friday, the Muslim day of rest. Iraq refuses to provide test data from the production of missiles and engines. President Clinton again orders air strikes on Iraq. At the end of the month, the U.S. House of Representatives approves two articles of impeachment accusing President Clinton of perjury and obstruction of justice.
During 1998, the S&P 500 index rose from 970 to 1,229 (27%). That’s just one year, chosen at random.
News rhymes. The media hypes the latest weather, earthquake, tsunami, merger, war or risk to the consumer’s ability to buy things. The U.S. stock market has seen it all before and generally gone up.
As a final example, consider the reporting of Hurricane Katrina. The U.S. economy would be greatly affected by the devastation in New Orleans. That was a head-scratcher to me. How could a hurricane in New Orleans affect shopping habits in, say, New York, New Hampshire, New Jersey and Nevada? I looked back at the August 1992 assault of Hurricane Andrew, only the third Category 5 hurricane to strike the continental United States. It caused $26.5 billion in damage but had no lasting impact on the stock market. I therefore recommended to my readers that they ignore Katrina as a market event. (Only as a market event. I would never say that news events affecting people are unimportant just because they have no bearing on the stock market.) Wal-Mart confirmed my view in early October by reporting that its September (post-Katrina) same store sales rose 3.8%, proof that the American consumer was alive and well. We are now in the middle of a robust holiday shopping season, with Katrina long forgotten as an economic issue.
Most of what reaches us is noise. Forecasting the stock market requires tuning out noise. Most people fail at market prediction because they focus on issues that don’t matter to the market.
What should they focus on?
After understanding the news cycle, we are much better prepared than most people to tune out what doesn’t matter. We are free to move on to hard numbers that do matter.
Above all, we have to assess the earning power of companies. When companies make money, stocks go up. When companies make money faster than expected, stocks go up faster. Companies report sales and earnings. Relying (again) on history, we can use this data to assess how expensive the market is now, and the odds of it going up or down. The concept of mean reversion is helpful. When an asset or an asset class reaches 2x, 3x, 10x, 50x its historical valuation, odds are it will go back down. The odds of it continuing upward decrease as it rises. Take housing, which has been on a tear the past few years. The odds of it continuing on this trajectory are low. Not nil, but low.
Regarding the economy, we can measure inflation and monitor the Federal Reserve’s response. Interest rates have a direct impact on stocks. If those rates rise (fall), stocks go down (up). Nobody knows what the Federal Reserve will do with rates, so we just guesstimate the odds and accept some unavoidable risk. Surprises are a part of the business.
Let’s apply these factors to develop a snap forecast for 2006.
We already know roughly what to expect in the news. Outlier events such as a large-scale terrorist attack or a Federal Reserve policy change are unquantifiable risks, therefore ignored in our forecast.
We know that the Fed is targeting inflation with interest rates as its main tool. Rising interest rates should slow America’s gross domestic product (GDP). GDP grew 4.4% in 2004 and is on track for 3.6% this year. Next year, it will probably grow about 3%, its long-term rate.
With a slowing GDP, we should expect slowing corporate earnings growth. Companies have been doing very well the past two years, growing earnings 20% in each. That trend will break to the downside. Earnings growth next year will be in the 4%-8% range, also in line with its long-term rate.
We’re not winning any Nobel prizes here, just relying on the old faithful concept of reverting to the mean.
The S&P 500 index gained 9% in 2004 and is up less than 4% so far this year. Those gains came with strong corporate earnings growth, and that growth is about to slow. If earnings grow at 6%, the mid-point of the 4%-8% range noted above, then we should expect the market to gain about 6% next year if the price/earnings ratio (P/E or “multiple”) remains constant.
Our final step is to look at what might change the P/E, what investors are willing to pay for earnings. When investors get excited, they pay more. When they get scared, they pay less. Will the multiple grow or shrink next year? The most likely factor to affect the mood of investors is the Federal Reserve’s action regarding interest rates. We’re now back in the realm of the unknowable. If the Fed stops raising rates in early 2006, the mood will improve, the multiple will expand and the outlook for stocks will brighten. If the Fed keeps raising rates, the opposite will occur.
Working off our base expectation of a 6% rise in the stock market, let’s assign a 2% increment up or down based on Federal Reserve action. That leaves us with a projected 4%-8% gain in 2006, in line with corporate earnings growth. Because this is a rough estimate, we should just say that the market will rise modestly in 2006, or that we expect single-digit gains next year. Maybe that we’re mildly bullish.
So how will I decide when to be in and when to be out of the stock market in 2006?
Watching, Waiting and Trading Against Noise-Events
During 2006, I will watch and wait for investor overreactions to news that doesn’t matter, keeping in mind as a benchmark the above forecast for a modestly up year. If something makes the price of oil spike and stocks plummet in reaction, I will look to buy the dip. If the Federal Reserve stops raising rates and investors celebrate by bidding stocks up 20%, I will look to sell the jump. Of course, I will also revise my benchmark expectations if corporate earnings surprise.
Similarly, I will watch for overreactions in individual stocks. For instance, Decker’s Outdoor (DECK) fell from a 52-week high of $49 at the end of 2004 to $23 on October 5, driving the P/E under 13. I saw that drop as an overreaction to transient news about a basically sound company in product transition mode, and told my readers to buy. Then the company issued an earnings projection below Wall Street’s consensus. The stock fell further, and I told readers to double down at $17 on October 28 for an average cost of $20. It simply became a better bargain because investors had again overreacted to bad news. DECK closed at $30.35 on December 9. We are up 52% so far, and still own it.
That’s how it’s done, at least at my desk.
Thanks to Jason Kelly for providing this piece. Perhaps one can better discern what is a noise-event for the U.S. markets by living somewhere else. (Jason lives in Sano, Japan.)