Where are all the jobs? Even in the midst of a multi-year recovery, labor-market indicators make for grim reading. The hires rate sits well below pre-recession levels, the unemployment level has declined less than the number of people leaving the labor force, and the quit rate is barely off its 2009 bottom.

Conventional wisdom holds that small businesses are the engine of employment growth, the solution to a lack of labor-market dynamism.

And small businesses are critical to a healthy US economy—97 percent of all firms have less than one hundred employees. But it is start-ups, which just happen to be among the smallest firms, that drive employment growth. Between 1994 and 2000, these newly born businesses contributed an average of 190 percent of newly created jobs. Put another way, start-ups created jobs and older firms destroyed them. Even in the middle of the Great Recession, start-ups were creating jobs. Granted, the job creation slowed, but it never dried up.

The question economists should be asking is whether entrepreneurs are creating a sufficient number of new companies. And, more to the point, whether these companies are creating the needed employment opportunities.

According to the Bureau of Labor Statistics, start-ups are being born at about the same rate as before the recession. The entrepreneurship rate—the number of birthed business per 1000 people in the labor force—sat at 1.45 in June 2013, already back to healthy pre-recession levels. The number of business births, after rising throughout the ‘90s and the early ‘00s, declined as the Great Recession took hold. But the bounce back was rapid. Excluding the large jump in the first quarter of 2013 (the numbers were inflated due to an administrative change), business births have averaged 196,000 since the beginning of 2010—nearly the same as the 2002 to 2007 average of 208,000.

The problem is that the number of jobs an average start-up brings into this world (i.e. birth weight) has been in steady decline. During the ‘90s, newly birthed businesses came with six new jobs. By 2007, they came with four. And this “small baby” phenomenon will continue.

There remains a particular entrepreneurial vibrance in the US economy. People are starting companies at the same rate as before the recession, but steadily fewer people are tied to each new business—a function of the increase in US labor productivity. In the third quarter of 1998, the US economy brought two hundred thousand new businesses and 1.2 million new jobs into the world. When the same number of establishment births occurred in the fourth quarter of 2010, there were only 787,000 jobs associated with them—a 37 percent decline.

Births, and in general young firms (open for less than one year), are by far the most critical component of employment growth. Young firms produce more than 100 percent of the new jobs in the US economy, but the past decade has seen a slow and steady decline in the number of jobs they contribute. Between March 1999 and March 2000, young firms created 4.6 million net jobs. During the same period in 2012 through 2013, there were 2.8 million created by young firms. This was still well above the total 2.55 million jobs created—but not enough. Old businesses were a drag on employment growth in both periods, accentuating the importance of young firms in generating jobs.

Less than 27 percent of the establishments in the US were born before 1993. And young businesses fail at an extremely high rate. For example, of the 634,000 businesses born between March 1999 and March 2000, only 212,000 were still around by March 2010—a 66 percent fail rate. But businesses born more than fifteen years ago employ 73 percent of the workforce, while those less than five years old employ just 14 percent. Most jobs are created in the US economy by start-ups, but most people work for older firms. Older firms don’t create jobs, they preserve—and eventually destroy—them.

The entrepreneurial spirit of the US economy is intact. There are still enticements to start businesses, create new products, and innovate. It is the dynamism of the labor market that has declined. Start-ups are launching with fewer employees, and not hiring quickly even if they survive against the odds. The average number of employees for a middle-aged firm (five years old) is eight.

The labor market is struggling to get back to full employment because it takes fewer people to build a start-up. These young companies delivered more than four million new jobs to the US economy a decade ago, but advances in technology have increased productivity and slashed this figure by 37 percent. This trend is not going to reverse itself—productivity is unlikely to decline—more entrepreneurs are needed to make up the lost jobs. To fix the unemployment problem, the US needs to encourage the start-up culture. The spirit is there, but the jobs are lacking. It is America’s start-up problem.

Samuel Rines is an economist with Chilton Capital Management in Houston, TX. Follow him on Twitter @samuelrines.

Image: Flickr/401(k) 2013. CC BY-SA 2.0