NEW YORK (CNNMoney) -- Unemployment has tumbled by nearly a full percentage point in the last two months, even though employers didn't hire many people.
That's a neat trick which puzzled economists and investors trying to make sense of the closely-watched government report Friday.
But the report was not quite as illogical as it seemed.
The disconnect is partly because the two closely watched numbers the Labor Department reports every month -- the number of jobs created and the unemployment rate -- come from two completely different surveys.
But a few other factors explain the disconnect between the numbers -- including severe weather, sampling issues unique to the month of January, and real changes in the economy --and show that the job market really is getting better.
1. The weather: A big part of that disappointment was due to the exceptionally bad weather. The Labor Department estimated there were 886,000 workers who had a job but couldn't get to work due to weather. That's the fifth-largest total on record, and more than twice as large as the normal number of workers trapped at home in January.
But while employees got a few snow days, all those closed businesses meant fewer counted as on the job.
Deutsche Bank's economists calculated that without the bad weather, job growth would have been close to 200,000 in the month, rather than the anemic 36,000.
So the job market could actually be improving more than the job growth number made it appear.
2. A statistical anomaly: The reason the unemployment rate fell to 9.0% from 9.4% in January is because of a large increase in people no longer counted as part of the workforce. And that's the part of the report that has economists scratching their heads.
Typically when there's a jump in workers leaving the labor force, it can be because people become discouraged and stop looking for work. But that wasn't the case this month -- the number of discouraged job seekers also fell in the month.
So what was the issue?
Part of the answer is a factor unique to January. That's when the Labor Department adjusts the population figures it uses each year. So because the overall population measure had changed, the Labor Department's take on how many more people said they had jobs compared to December was significantly lowered.
"January is always a difficult month for estimating seasonal factors, and it was much worse this month," said Lakshman Achuthan, managing director of Economic Cycle Research Institute.
Some experts saying the report foreshadows larger changes taking place in the economy, like unemployed people starting new businesses or finding some other way to make money.
"A lot of these folks, even if they're not in regular jobs, you and I know they're earning money," said Jeff Miller, CEO of money management firm NewArc Investment. "The [Labor Department] is always attempting to model the creation of new businesses. Usually that works pretty well, but sometime it doesn't."
3. Economic improvement: Another problem is that the models the Labor Department uses to calculate numbers don't work as well when the economy is taking a sharp turn for the better (or the worse), said Achuthan.
So the apparent disconnect between the sharp drop in unemployment and the anemic job growth could be a sign that the economy is actually getting better and the numbers haven't caught up with that reality.
"These numbers don't make sense on the surface, but that makes sense, since we're around an inflection point," he said.
The big drop in unemployment, even in the face of weak hiring is consistent with the other views of a labor market finally turning the corner. Recent economic readings -- such as steady declines in those filing for jobless benefits and strong consumer spending seem to back that up.
It's quite possible that the 9% unemployment rate will prove to be too optimistic of a snapshot, but at the same time, the 9.8% reading in November may also prove to have been too pessimistic.