After keeping the markets on edge in the days and weeks leading up to today’s decision, the Fed has decided once again to leave interest rates unchanged.

As we’ve spent this week discussing “all things Fed,” it’s worth spilling a little ink on what the quasi-government-but-really-private organization is supposed to do.

Originally, the Fed was built as a lender of last resort. In times of panic when banks needed to sell good assets for cash to meet depositor demands, someone had to be there with the ability to make good.

This is the proper role of a central bank, and is exactly what the Fed did in March 2009 when it printed $1 trillion to buy government-insured mortgage-backed bonds. Fed actions thawed the frozen market, which resuscitated the buy side and helped things return to normal, albeit at a lower level.

But over the last 100 years Congress piled other responsibilities onto the Fed, most notably full employment.

The monetary institution is supposed to use interest rates and the printing press to prod businesses into hiring more people. This part of Fed policy has failed, but you wouldn’t know it by looking at employment figures from the Bureau of Labor Statistics (BLS).

Here’s how it works.

The Fed tries to create jobs through lower interest rates. If consumers and businesses can borrow at cheaper rates, then the thinking goes they will borrow more money. With that extra cash, they’ll spend more on goods, services, equipment, etc., thereby driving up demand for such things. Businesses then will meet increased demand by hiring more people to build, sell, and maintain all the new stuff.

Only, it hasn’t happened that way.

Some industries, such as autos, benefitted from higher sales based on lower rates. Most did not.

Take housing. It is logical that with very low interest rates home buyers would rush out to purchase homes, thereby driving up new home construction. But the pace of home sales since 2008, particularly new home sales, has been anemic at best.

At the height of the market, our economy was churning out more than 1.3 million new homes a year. After falling to less than 300,000, the number recently topped a mere 500,000.

If we’re not building a lot of new homes, we don’t need more framers, roofers, plumbers, etc. The same is true of other industries – without soaring demand, why hire more people?

A key to gauging employment is the rate of new business creation. It’s not just small businesses that hire the most people, but small new businesses.

This is why the BLS tries to figure out how many net new businesses opened up each month, and therefore how many people were hired at these emerging companies. The calculation is called the “birth/death adjustment,” and it is very important to the monthly jobs report.

In August, the U.S. economy created 173,000 jobs. Within that number, the BLS estimates that 111,000 jobs were created at new businesses. In other words, a majority of the jobs are based on a guess. They’re not based on verifiable facts.

And that’s just August. The BLS estimates 658,000 jobs from new businesses year-to-date. Compare that with the roughly 1.6 million total new jobs – not just from new businesses – estimated for the year. More than one-third of the new jobs are guesses, and not very good ones at that.

Looking to other sources, it’s clear that new businesses aren’t popping up all over the place. The polling company Gallup recently published a graph showing the number of businesses failing each year compared with the number of businesses created. From the beginning of the chart in the 1970s through 2007, more companies opened each year than shut their doors.

Since 2008, the numbers have been reversed. The data only goes through 2012, but it shows that our economy consistently sheds companies, and presumably jobs.

Now we have even better data.

Just this year the Kauffman Foundation created a Startup Activity Index, which contains monthly data starting in 1997. Currently the overall index is negative. It has been below zero since 2010, dragged lower by the rate of new business creation per 100,000 people. This number bounced between 165 and 200 from the late 1990s through 2008, then dropped to 133 in 2009.

This year it sits at 130, well below the average of the 2000s.

For all the money the Fed created in its attempt to drive interest rates lower, it has not, and cannot, force economic activity. I guess there are limits to what the Fed can do… but that’s OK. The BLS is there to guess at numbers to make us feel better.

Rodney