It’s all quite clear as economists draw it up on their blackboards.  Growth in productivity – defined as the output produced per person-hour of labor – is what ultimately allows us all to enjoy a higher standard of living.  When we collectively produce more, we earn more.  Or to put it another way, we can afford to pay ourselves more without provoking inflation.

And since the mid-point of the last decade, the measures of economy-wide productivity produced by the U.S. Bureau of Labor Statistics have brought us mostly happy news.  Productivity growth in the last ten years, by historical standards, has been extremely strong, even in the labor-intensive services side of the economy where growth was once harder to find.  And even though this has fueled more profit growth than wage growth than some would like, it has helped the economy grow briskly and still keep the inflation monster at bay.

It’s a great trend, except for one small problem.  We really don’t understand it.  Every explanation that’s been offered for why the entire economy started getting better at producing more with less since 1996 seems to raise more questions than it answers.  And it’s a little scary heading into the future on a wave that could end as easily as it started.

It may be ending already.  According to BLS, the growth in output per hour in the business sector slowed to 2.3 percent in 2005, the slowest growth since 1997.  Growth had peaked at an astounding 4.3 percent in 2002, and has averaged 3.0 percent since 1996.  With the second quarter of 2006 showing a mere 1.5 percent growth rate in productivity growth economy-wide, policymakers and inflation hawks everywhere are taking notice.

But what led to the sharp growth in the first place?  Of course, computers and information technology are part of the puzzle.  It’s hard to find a workplace that the information revolution hasn’t touched in some significant way.  It’s natural to think that computers have made us more productive – except for the fact that we had PC’s and bar-code scanners and plenty of other gizmos around for at least a decade before the productivity spurt got started.

When you look at the productivity trends in individual industries, you often see little or no productivity impact even after computers arrive in force, only to find growth much later when  the investment matures.  That has led many to say that it is the integration of technology into the production cycle, rather than the presence of technology itself, that has put the economy on a faster track.  That’s easy to say, but hard to measure.

And when you peer underneath the hood even further, things get murkier yet.  Measuring productivity, especially in the knowledge-intensive, services producing economy, is anything but straightforward.  You can’t put what the office-dwelling, white collar world produces on a scale to be weighed or otherwise directly measured, can you?  Yet it has value, both to customers and to the owners of companies alike.

In fact, that last connection was the subject of a recent National Bureau of Economic Research study on the importance of proper accounting of so-called intangible capital.  Those are the things that we produce for the companies we work for – the knowledge, the procedures, even the reputation – that make owning the company valuable, yet cannot be packaged up and sold by themselves.

The importance of these intangibles is obvious from an examination of almost any company’s asset sheet, yet their importance is missed in most national economic accounting, including what is used to produce estimates of productivity.

But knowing something’s wrong with productivity estimates is one thing – correcting the problem is something else.  In fact, measuring the knowledge-based economy is hard enough to send some of us back to the safety of our blackboards.