BANGALORE/WASHINGTON U.S. economists are having an unusually tough time predicting the path of a recovery that doesn't seem to fit neatly into a bullish or bearish box.
Forecasts for economic data are all over the map, causing confusion about whether the actual readings are better or worse than expected and what that might mean for growth prospects.
Reuters examined its polling data from more than 80 economists and found the range of estimates for some of the most closely watched indicators -- employment, housing and gross domestic product -- widened dramatically in 2010.
The growing gap is a reflection of what Federal Reserve Chairman Ben Bernanke calls "unusual uncertainty" about the economic outlook. It suggests the bulls and bears are pulling even further apart as debate rages over whether the economy is destined for a double-dip recession.
For investors who rely on these estimates to gauge each day's batch of data, the wide span means the consensus forecast does not necessarily represent what most economists think.
Instead, it may mark the midpoint between two widely divergent schools of thought, which can lead to heightened market volatility even when the figures match expectations.
For the White House, which is eager for evidence its policies are helping repair the economy, each disappointing number that tanks the stock market provides fodder for critics who say the approach isn't working.
It is not uncommon to see a wider gap when the economy is pivoting from recession to recovery. The magnitude, however, is striking, and there is no sign the uncertainty is going away.
"At every turning point, half the people in the room think the recession will never end and the other half are optimistic," said Chris Rupkey, chief financial economist at Bank of Tokyo-Mitsubishi in New York. "This particular turning point seems to be, as the Fed chairman said, one of unusual uncertainty."
In the Reuters data, the average forecast range for the monthly employment report stands at 250,000 so far this year, far wider than the average of 170,000 since the start of 2005.
For existing home sales, the average range ballooned to 800,000 this year, nearly double the 2005-through-2010 range. And even with that unusually wide range, the actual housing data has come in below the lowest forecast twice so far this year, triggering heavy stock market declines.
On GDP, the range has averaged 1.9 percentage points, well above the 2005-2010 average of 1.3 points.
Other surveys show a similar pattern. Randall Moore, executive editor of Blue Chip Economic Indicators, said he noticed an abnormally large gap between the top and bottom 10 forecasts in his monthly GDP polling.
"You've still got a strong grouping in the middle. The outliers are just further apart," he said.
What divides the groups is disagreement over how well the economy will hold up once government stimulus wears off.
The bulls point out that corporations are cash-rich and household savings have grown, providing the building blocks for the recovery to pick up steam next year.
The bears think consumers have permanently changed their borrow-and-spend ways, which means the consumption-dependent economy will falter. They worry that political opposition will prevent the government from providing more assistance.
DISAPPOINTMENTS AND HALLELUJAHS
Tuesday's retail sales report for August illustrates how these varying views can affect investor perceptions of the economy. Depending on which forecaster you follow, retail sales might be the strongest since March or the weakest since June.
The consensus view calls for a rise of 0.3 percent, which would be a tad bit weaker than July's. But the forecasts range from a decline of 0.3 percent to an increase of 0.6 percent, and only 22 of the 77 economists matched the consensus call.
Reuters removes the highest and lowest forecasts before calculating the median, which helps reduce the impact of the extremes, but sharp differences in opinion remain.
Goldman Sachs, which had the highest retail sales forecast in the Reuters survey, said it based its bullish view on reports from retailers showing sales picked up in August.
Landesbank Berlin, which had the lowest August retail sales forecast, thinks the weak housing market will drag down sales of building materials, economist Uwe Duerkop said.
If the actual number comes close to either of those two forecasts, the market reaction could be volatile. Bank of Tokyo's Rupkey said stock investors were unusually keyed into the consensus, and even marginal misses triggered big moves.
"It's become quite important in setting the tone of daily trading. The disappointments or hallelujahs after some of these numbers... can be really extreme," he said.
Wells Fargo chief economist John Silvia said "turning point issues" make it particularly difficult for forecasters to get their calls right when the economy is shifting gears.
In the middle of a normal business cycle, it's easier to predict how many homes or cars consumers might buy in a given month based on population and income growth. Coming out of a deep recession, especially one triggered by a financial crisis, unpredictable factors such as volatile consumer confidence or reduced access to credit can skew buying patterns.
The housing market has been even harder to estimate because of a government tax incentive that drove a brief but powerful buying frenzy. That distortion can account for half of the 10 largest forecast misses so far this year.
Blue Chip's Moore suspects the rest of the discrepancies may be linked to a deeper problem: economists can't accurately tell which way the economy is headed because they are looking at it through outdated lenses.
"The template they're using is a poor one because they're treating the recession like the typical post-war recession and it just isn't," Moore said.
(Additional reporting by Chandra Ramarathnam in Bangalore; Editing by Andrew Hay)