This is one of those days when I can’t decide whether to be an optimist or a pessimist. The ISM report was truly bullish and David Rosenberg convinces me that the fourth quarter economic numbers don’t make any sense at all.

The ISM’s manufacturing survey showed an increase from 54.9% in December to 58.4% in January. As you can see from the following chart, all of the components of the index are marching ahead:

It’s pretty hard to dismiss these numbers outright, so I flipped over to the WSJ Real Time Economic blog to get their take.

It’s the sixth month in a row above the 50 level that indicates manufacturing growth, and a level of 58.4 corresponds to annualized GDP growth of about 5.5%, suggesting the economy is carrying strong momentum into the first quarter after its best quarterly performance in six years.

The catch: Is the growth creating jobs? ISM’s manufacturing employment index topped the 50 level for the third time in four months, rising 3.1 points to 53.3 in January, its highest since April 2006.

But ISM’s service-sector employment gauge has been much slower to recover. In December, ISM’s nonmanufacturing employment index came in at an uninspiring 43.6; the sector employs nearly nine in ten U.S. workers. Economists expect it to remain below 50 when the January report is released Wednesday, though the headline index is likely to cross back into expansion.

So, they conclude that it’s a positive but caution that the jobs picture is still the big black cloud that’s yet to be dispelled. I guess, more to the point, that given that most of the economy is employed by the service sector, there isn’t all that much in the report that suggests we are about to see a turnaround in employment, unless that ISM report on Wednesday holds some surprises.

Now to put a bit of a damper on all of this, consider this analysis of fourth quarter GDP results from Rosenberg:

But consider that never before have we seen a 5.7% GDP growth rate in a quarter when the aggregate workweek was cut by 0.5%. There is no way that productivity is running as high as the data suggest. Either the hours worked and employment data are wonky (though the jobless claims data are still consistent with moderate job loss) or the GDP data are wonky, or maybe they both are.

Q4 real GDP in the U.S. came in at 5.7% QoQ annualized rate, but as we saw in the first take of the Q3 results, watch out for the revisions

Here is another way to assess the data: We saw history in the making — an eye-popping 5.7% GDP growth rate the exact same quarter that the unemployment rate rose 40 basis points, to 10%. It is like Houdini’s rabbit! This has never happened before. Normally, when we see a GDP number like this the unemployment rate declines 20 basis points during the quarter in question. The flip side is that in the past, when the unemployment rate rose as much as it did in the fourth quarter, believe it or not, in those quarters real GDP actually contracted fractionally (at a 0.5% annual rate).

We went all the way back to 1947 and so we can say with 100% confidence that at no time in the past 62 years has a 5.7% GDP advance coincided with such a rise in the jobless rate. It makes no sense. In fact, if you believe that GDP result, then you de facto are of the view that all of a sudden, with no capital deepening or major technological change in the past half decade to speak of, the potential growth rate in the United States has reached an epic scale of 7% (the growth rate in the economy that keeps the unemployment rate stable) using a classic Okun’s Law rule.

That is really tough to swallow. As we said: expect big-time revisions.

He has a lot more to say on the subject, particularly concerning the unlikely scenario of a robust rebound while consumer spending and housing remain in the dumpster, so take a look at the full article.

At this point you might be as conflicted as I. Are we seeing the beginning of real recovery or is it some sort of manufacturing blip that is going to do little to drive the economy over a longer term? More importantly, even if it is a sustainable trend, will it do that much for the overall economy given the slice of the pie that manufacturing represents? I honestly have no idea. Emotionally I’d like to wrap my arms around the ISM data and say we are finally making some real progress but my head’s telling me to wait and see what comes out not just in the next week or two but over the first quarter.

All of this is useless if it doesn’t begin to make a meaningful dent in unemployment. So far, that does not appear to be happening but the WSJ post usefully reminds that job growth does tend to lag other indicators. Let’s hope so, but I still don’t get the feeling that we’re really entering into a period that will mark significant recovery. Pessimism isn’t my natural state but it’s still the dominant mood I feel right now.