But in the media you get these "analysts" who talk a good game, acting as though a 50% to 70% probability is something meaningful. "The market has turned. The recession's over," they say. They claim we have the first balance-sheet recession in 70 years -- but still want to compare garden-variety recessions to what we have now.

Again, we can only know which of the moves (above and below the 200-day moving average) will be the real "indicator" in 6 months. To use these as indicators today makes as much sense as driving forward while looking in the rearview mirror.

The New Normal Is Still in Our Future

Now let's take that principle a little further. Two weeks ago, I detailed how air, trucking, and rail shipping are down 20% year over year. Global trade's down about 30% in the major exporting countries (see below).


Click to enlarge.

Is it the end of the world? Do we just keep falling? No. At some point, 6 months or a year from now, the year-over-year comparisons become easier. If you are at 100 and fall to 80, then a year later you're at 88 and -- Voila! -- you have a 10% increase! And the perma-bulls will be talking it up, ignoring the fact that you're still down 12% from the peak.

The point is that we have fallen quite a bit in a lot of major categories. There's really only so far you can fall. And then, once you reach that new lower level of the new normal, you begin to rise. At some point, we'll be on the path to "recovery." That doesn't mean that we'll be back to the halcyon days of mid-2007 within a year. It just means that we'll have stopped falling and have  adjusted to the levels of the new normal.

The Hidden Problem Within Unemployment Data

This is going to be most evident and painful in the unemployment numbers. Last month saw the number of unemployed rise by 345,000. What wasn't in the headline data was that 217,000 of those jobs were estimated from the "birth-death" ratio. The US economy creates new businesses that don't get counted in the data, so the Bureau of Labor Statistics estimates what that number is by using previous data patterns. When the economy turns, it overestimates new jobs in recessions and underestimates them in recoveries. No conspiracy, it's just the best methodology we currently have.

But does anyone really think 200,000 jobs were created last month? The real number of lost jobs is worse than the headline. And next month the birth-death number will likely be over 200,000 again. Add another 100,000 or so to the headline number to get closer to reality.