For the past few months, it seems like macroeconomic forces such as terror, inflation and interest rates have dominated investment decisions. Hopefully, now that Wednesday's Federal Open Market Committee meeting has settled the interest rate question for at least the next six weeks, investors can focus on upcoming second-quarter earnings reports.

Unlike trying to handicap what the Fed might do and then divine its impact on the stock market, earnings provide a more immediate influence. And they're ultimately the most important criteria in determining a stock's value.

While there is no way to know for certain how a company's stock will react following its earnings release, there is often a sell-the-news reaction. But according to research by Keith Miller, a quantitative analyst at Smith Barney, there are also discernable price performance patterns prior to the reporting date.

Specifically, he concludes that "the rewards from accurately forecasting earnings surprises are much greater in the two- to five-week period before the earnings report than they are in the weeks after." Miller's study shows that the top quintile of stocks in the Russell 2000 with the largest earnings surprises outperform the broad index by some 4 percentage points in the four weeks before they report.

Getting Surprised

The key, of course, to capitalizing on this price run-up ahead of earnings lies in the phrase "accurately forecasting earnings surprises." Thankfully, Miller has provided a glimpse into his (and Smith Barney's) quantitative methodology for finding candidates. His recent report describes the calculation for standardized unexpected earnings, or SUE, as the difference between the company's actual reported earnings and the mean of analysts' estimated earnings, divided by the standard deviation of the estimates. The higher a stock's SUE, the greater the expected surprise. This historic or realized SUE is based on the most recent 16 quarterly earnings reports.

Here are five of the independent variables -- and the justification for applying them -- that the forecast model uses in predicting earnings surprises for the upcoming quarter as described in Miller's research report:

-- Realized SUE of the two previous quarters: Earnings surprises are serially correlated; i.e., they tend to repeat in the same direction.

-- Earnings estimate momentum: Revisions in earnings estimates tend to precede earnings surprises in the same direction.

-- Price momentum: Price movements tend to precede earnings surprises in the same direction.

-- Sales growth: A positive surprise often follows an increase in sales over the most recent four quarters.

-- Profit margin momentum: Large increases in profit margin tend to precede large positive earnings surprises.

Note that SUE normalizes earnings surprises by considering the level of analyst uncertainty. That leads to the assumption that the narrower the range of analysts estimates (meaning that everyone is pretty much in agreement about what the company will report), the greater the impact of any surprise or deviation from that mean will have on the stock price.

These may all seem like fairly obvious criteria, but you should never overlook the obvious. Since the model's inception in January 1996, 78% of the companies predicted to post a positive earnings surprise did in fact come through. This compares with about 57% for the Russell 1000 as a whole. The model's predictive ability was in great evidence in Miller's May 1 report, which identified 20 candidates, 19 of which went on to report positive earnings surprises. The cumulative outperformance of those stocks vs. the Russell 2000 was 6 percentage points during the four-week period preceding the report dates.

The table below lists companies scheduled to report within the next five weeks that have positive SUE or are expected to deliver an upside earnings surprise.

Companies with Forecasted Positive Earnings Suprises
These stocks could be interesting plays
Company Name Report Date First Call Estimate Closing Price on 6/29
Juniper (JNPR) 7/6/04 $0.04 $24.10
Motorola (MOT) 7/20/04 0.17 18
United Healthcare (UNH) 7/16/04 0.91 61.40
Wellpoint Health (WLP) 7/19/04 1.82 110.00
Lexmark (LXK) 7/21//04 0.96 95.60
Tellabs (TLAB) 7/22/04 0.06 8.69
Starwood (HOT) 7/26/04 0.37 44.33
Cigna (CI) 8/2/04 1.25 68.65
Source: Smith Barney

Buying Call Options

I think this is a situation where it's best to keep the strategy as simple as possible; therefore, I'd suggest buying short-dated calls on the names listed in the table.

This will help reduce the cost of the options and thus the risk (remember that the maximum loss of a long call is limited to its purchase price), while offering unlimited profit potential.

I usually shy away from buying options with less than 30 days remaining because that's typically not enough time to allow a thesis to play out. But in this case, we have a specific date/time window in which the expected move will occur. So, I'd focus on buying calls with the nearest expiration date after the reporting date. For example, since Juniper ( JNPR) and Motorola ( MOT) report on July 6 and July 12, respectively, I'd look to purchase the calls that expire on July 16. But for names like Tellabs ( TLAB) and Starwood ( HOT), you'd need to buy their related options with an August expiration.

Many people make the mistake of buying out-of-the-money calls because they're cheaper and provide more leverage. But even though they are less expensive in terms of absolute cost, the fact that they have no intrinsic value and are more subject to accelerating time decay as expiration approaches can offset a favorable move in the price of the underlying stock.

I would suggest buying calls that at are least partially or one full strike in the money. For example, on Wednesday with Motorola trading at $18.16, you could buy the July $17 call for $1.25 a contract. This means the option has $1.16 of intrinsic value and just 9 cents of time value; its break-even point is $18.25. Compare this with the July $19 call, which is trading at 25 cents (all of which is time value) and has a break-even point of $19.25 -- a 5.4% move in the underlying stock. If Motorola's price remains the same from now until the day after its earnings announcement, the $17 call will actually decline by less and incur a smaller loss than the $19 call.

Paying for calls that contain some intrinsic value still provides plenty of leverage and limited downside, and it requires a significantly less capital outlay than buying a comparable number of shares of the underlying security.

For example, if you wanted to buy 100 shares each of all seven of the stocks listed in the table, you'd have to pony up about $21,400, assuming a 50% margin requirement. But a purchase of one in-the-money call (one call represents 100 shares) would only cost around $2,135, or about one-tenth of the price and one-twentieth of the risk.

Don't be tempted by the relative low price of out-of-the-money options; buying near or slightly in-the-money calls may cost more on an absolute basis, but they still offer great leverage and provide better odds for realizing a profit.
Steven Smith writes regularly for TheStreet.com. In keeping with TSC's editorial policy, he doesn't own or short individual stocks. He also doesn't invest in hedge funds or other private investment partnerships. He was a seatholding member of the Chicago Board of Trade (CBOT) and the Chicago Board Options Exchange (CBOE) from May 1989 to August 1995. During that six-year period, he traded multiple markets for his own personal account and acted as an executing broker for third-party accounts. He invites you to send your feedback to steve.smith@thestreet.com.