The Bureau of Labor Statistics tracks many of the most important economic numbers for the Federal Government. One of the most widely followed is the monthly Employment Situation Summary, which contains both the survey for the Unemployment Rate, which is calculated by a survey of workers, and the Payroll data, which is calculated from business. The first the "A" tables are often touted in the press, even though there is wide misunderstanding about what the "Unemployment Rate" means, and the second, the "B" tables, are followed more closely by the business press as a more accurate guage of the labor market.

This is why there is often a divergence in reports on the economy - the headline UR was 4.6%, which is very low. However the "establishment survey" reported that only 59,000 private payroll positions were created in September. To understand why these numbers are really saying the same things, a bit of explanation is in order.

The unemployment rate is a measure of how much wage inflationary pressure there is on the economy. This is called "wage push" inflation, because if people have more money to spend, they spend it, if the economy doesn't have enough extra capacity to produce more at the same or nearly the same prices, then this produces inflation, as "more dollars chase the same number of goods". The unemployment rate being low is not, intrinsicly, good news. Instead it says that this, is as good as it gets, and that the central bank, in our case the Federal Reserve, has to watch for the possibility of inflation.

To measure this, the Department of Labor figures out how many people could be working - that is the "non-institutional population" - figures out how many want to work or are working - that is "the labor force". The unemployment rate is the percentage of the labor force that is actively looking for work, but cannot find a job. There is a loophole here. People who are in jail or the military - the US has a high rate of incarceration and a large military - don't count for these purposes. People who want a job but have given up, don't count for these purposes. People who are working jobs that don't pay, do count as employed.

This is why many economists follow, instead the participation rate - the total percentage of the adult population working.

The establishment survey is a bit clearer, but there are caveats as to what we report when we report the number. The number - 59,000 private payroll positions - doesn't mean that exactly 59,000 times a new job was created. First it must subtract the number of job losses. In some months there are large job losses, but even larger job creation - the Clinton economic expansion featured this. The second thing you have to realize is that this is "seasonally adjusted". This means that the economy created 59,000 more jobs than it otherwise would have if normal seasonal hiring and laying off went on. This means that in times of year when business usually hire, say retail stores at Christmas, this number represents how many more people were hired than one would expect from ordinary business. This can be difficult to do, and it can also be a means by which numbers are manipulated.

Finally, since the payroll survey relies on data from established companies, and most hiring, and laying off, is done from small companies and new companies - a "new company", for example, could be a new McDonald's Restaurant, because a franchise isn't on the McDonald's corporation payroll, but the payroll of the company that owns the franchise rights - there is a "birth/death" model which figures from company creation how many jobs should be created by the new companies. This is because a new company shows up on records before the employees it hires show up on records. This is later corrected as the actual data comes in. This is one of the reasons why initial reports are "preliminary" and subject to revisions.

So the payroll report is, essentially, how many more jobs are being created than we would expect for the month.

This is why the two surveys can be saying the same things, even when the unemployment rate seems low, while hiring, well, just plain rots. The Unemployment Rate is saying "this economy is employing as many people as it can without generating inflationary pressures." Actually, to be more precise, it is employing more people than it can without generating inflationary pressures. The payroll survey agrees with this, it says "on the margins, employers don't think it is worth it hiring another worker." In fact, hiring is just keeping up with the growth in the civilian population since March. You can see this from the "Employment population" number.

There are a host of problems with the way we calculate these numbers, and these problems are having economic effect. One that the "A" table unemployment rate doesn't count imprisioned people, or people in the military. But people in both groups create inflationary pressure - they use resources that could otherwise be put to productive ends. By not counting them correctly, the unemployment rate understates the level of inflationary pressure. To be more accurate, both groups should be counted as "employed" - even if it sounds counter-intuitive that a person in a prison cell has a job.

Many people want the "Unemployment Rate" to more accurately reflect the state of the labor market. It doesn't do that, and only did that for the period where there was an inflationary bias to the economy and government policy. Right now there is a deflationary bias - the labor market and the unemployment rate are diverging, because in an inflationary biased economy, there is a tendency to employ everyone who can be employed. In a deflationary biased economy, there is not. The Unemployment Rate being a good reflection of the labor market was an effect of the Keynesian economy, and not a matter of definition.

People who do want the UR to be a more reflective of the labor market often do so because the UR is so widely reported, and used as a "factoid" that times are good. It isn't because of economic reasons, but political ones. The solution here is, first, report the Payroll number - which is closer to the reality of the labor market - and second to tell people what the UR really means - namely, the economy is basically topped out from monetary policy, and if we want growth, then there has to be increase in real productivity.

So what does this month's report mean?

The top level numbers tell a story of an economy that is "full up" on labor. That doesn't mean that everyone who wants or needs a job has one, or that people are employed in their most productive use. On the contrary, in a deflationary biased economy, labor is often under utilized and misutilized. On the other hand, in a deflationary biased economy, it says that the amount of useless labor is less than in an inflationary biased economy. Whether these matters are good depends on which side of the paycheck you are on.

However to really understand the employment report, it is necessary to look at the break down of who is gaining and who is losing jobs. For example, an economy which is losing service jobs and employing more higher paying manufacturing jobs is one that looks to grow for the long term, as over time higher paid workers will demand more services. Conversely, an economy that is laying off production and balancing it with lower paying service jobs, is one that is either installing new capital - so that it can lay off workers and have the same output - or one that is down shifting. Or both, often times the best moment to retool is when orders are down anyway.

By this measure we see some simple and interesting patterns:

1. Local government is retrenching. One of the biggest job losses was in education. Since this was back to school month, what this really meant is that schools brought back fewer teachers than expected. (Down 24,500)

2. Construction is moving from residential, to commercial. The "specialty contractors" for residential construction showed almost the same drop ( -17,500) and non-residential showed a gain (17,200). That's statistically insignificant.

3. Production workers showed big drops across the boards - 30,000 total, and that was roughly split between both durable and non-durable goods. Large drops are often caused by one sector being in trouble, such as the very cyclical automotive industry. This isn't that.

4. The residential real estate bust is starting to ripple through the job market. One area of strong growth over the last few years hasn't just been in building houses, but in selling them. Thus professional services related to mortgages, sales, architecture and legal aspects of real estate showed strong growth. Even as accounting and banking are showing modest increases, the sectors that have been hiring associated with residential real estate have been laying off.

5. People are eating out more, but the rest of the entertainment sector is down. This is a sign of a squeezed economy, as people cut back on amusements, but keep eating, and drinking.

Essentially what this report says is that economy has found, and is keeping, the workers that are productive for it. But it doesn't need much additional labor outside of particular specialties. It also shows the beginnings of a shift from residential and consumer driven growth, to business driven growth. However, the signs here are that business driven growth will produce much less employment, and since the current expansion has been the worst for employment since WWII - basically, since we started measuring these things comprehensively - the grumbling in the electorate is going to get louder.

My thesis since 2002 is that virtually all of the job growth generated has been inflationary. Many thought, or felt, that a drop in inflationary pressures would lead to the goldilocks scenario of growth without too much upward price pressure. That's what Bernanke thinks - that the slowing of growth would all come out of inflation. So while we would have somewhat lower GDP growth, we would also have much lower inflation. Since inflation, even the special multi-caveated, blinders on reactionary friendly kind that the Federal Reserve admits is happening - they have several layers of denial between themselves and the economy - is too high, way too high by their own measures - Bernanke was issuing the kind of inflation denial that sunk Ford and then Carter. Inflation, they thought, will take care of itself.

Instead what has happened is what is predicted by the "growth has been inflationary" thesis, namely, the drop in inflationary pressures has reduced the need for business to hire to chase the dollars that are flowing now. This has meant that employment has continued to slow, even as the Unemployment Rate stays high. Left to itself, the economy will either have to slow, as expectations for future inflation dampen both borrowing and chasing sales - or it will have to rev into one last lunge of debt chasing the last bit of lower rates. It seems likely we will get the second, rather than the first - that is, that people who missed out on the housing boom and historically low rates are going to take one last shot at buying a house, and then watch as the bottom falls out when the Federal Reserve gets serious about inflation fighting again.

Next year, in short, should be better than the recession watchers predict, but only at the cost of higher inflation, and a deeper recession later, when some neo-classical hard hat Fed Chairman takes over and pushes rates up to their Taylor rule level. While the popping of the recent oil pimple - not quite a bubble - will ease inflation pressure temporarily, all of the underlying pressures that will insure another round of oil inflation with the next major supply disruption - Katrina and Rita disrupted almost half of the American Gulf of Mexico output and refinery output - will set us on the next run to $80/bbl.

In short, the jobs report says the economy is coming to a "landing" of lower growth and somewhat lower nominal inflation - but that you, dear reader, are going to feel like one of the tires on a 747. And the real inflationary rate - that is the nominal inflation rate plus the deflation that globalization would otherwise be causing - is going to remain very high. This real inflation rate represents a transfer of wealth from those that have to pay higher prices - like people who buy houses and drive SUVs - to those who sell, like people who sell houses and don't have to buy new ones, and OPEC countries.

It's a tax, and it is going to people who have no intent, or ability, to do anything to make it come down.