The good news: Retail sales were up 1.4% in July, a full half-point above the expectation. Hot weather and higher gas prices look like the motors of this upsurge.

The bad news: It's the first number since the Fed decided to pause on raising the Fed Funds rate. That might be a mistake, says one analyst:

``We clearly see inflation flying higher,'' said Joseph Brusuelas, chief U.S. economist at IDEAglobal in New York. ``If consumers don't back off, it means that growth's not moderating along the lines exactly that the Fed has forecast, and maybe the third quarter isn't going to be as weak as we all think.''

Current rate hike odds for the Sept. 20 meeting of the Fed are at 36% right now, a fairly significant upswing. I tend to think that economic forecasters do two things to lead to what I call "herd revisions" -- they over-weight the most recent economic data, and they try hard not to get too far outside what other people are saying (though that second effect has lessened in recent years as people take more extreme views to get on Cavuto or Dobbs or whatever...) At the beginning of the year I thought GDP growth was going to be 3.5-3.75% Q4 to Q4. So far we have about 2% of the growth in the books (2.04%, if you want to be precise.) An average of 3.1% for the last two quarters puts us in the midrange of that estimate. Worrying about whether the next quarter is 2.8% or 3.4% isn't something I can do -- the errors I make in forecasting make either of those numbers quite plausible while still permitting me to think my initial forecast is valid. And while I don't forecast oil, I did have a gold forecast around $650. (The model is one I use for my grad forecasting course, which I haven't taught for a few years, but once I year I update the model anyway to keep it ready.)

Moral of the story? We're trying to forecast a difficult environment with very short-term methods. Maybe we should take a giant step back for a broader view. And maybe so should the Fed.