Our shortage shortage

It was only a few years ago that many electronics-industry pundits and self-serving spokespeople were yelling "shortage!" By extrapolating from a few data points, people who should have known better—or perhaps did not want to—asserted that we were facing a shortage of fab capacity, passive components, engineers in the pipeline, and even skilled engineers. The sky was soon to fall.

You all know what has happened since then. We have plenty of capacity, plenty of components, and plenty of good engineers looking for work. Depending on how you measure it, unemployment among engineers is between 4 and 7%. However, many observers think that, because of data-collection methods, this number understates reality by a factor as high as two.

What happened? The simple answer is that lots of things have happened. In other words, there's no single or dominant answer. Overproduction, too many new products that consumers don't want or are perhaps not ready for (or vice versa), competitive sources for software and hardware design and manufacturing expertise outside the United States, Y2K-spawned overbuying, and many other factors all contribute to the situation. I am sure any EDN reader could make as good a list of reasons as any professional pundit (not that most of the pundits and professors would ever admit that they were wrong).

But one thing puzzles me. When the shortage-claiming crowd was in full roar, we heard from educators that the United States was producing too few new engineers. My translation: They needed more bodies to fill those classrooms. And they needed them at the master's- and doctoral-degree level, too, they insisted—not just at the undergraduate level.

The number of graduating engineering students has been dropping steadily over the past decades. Yet, engineering projects still get done. In many cases, engineers and programmers without formal college training do this work, and, because they lack formal training, they don't show up in the enrollment numbers; in other cases, engineers trained or working outside the United States perform these tasks.

But I think another factor plays a role. Just as computers and software have increased productivity in many aspects of business, and we now take many of these gains or capabilities for granted (ATMs and debit cards everywhere, routine online ordering, and much, much more), these tools have increased the productivity of engineers in design and manufacturing. And they don't affect only the upfront design, either. They affect all aspects of the product cycle, including testability, mechanical fitting, bill of materials, project scheduling and coordination, production planning, and test data.

In short, we have become victims of our own successes and accomplishments. We have done such a good job that the tools and resources we develop dramatically increase our productivity, enable multimillion-gate designs, and ensure smoother testing and manufacturing. Whether we'll need as many engineers in the future as we have in the past, despite what the pundits and academics claim, is an open question.

Contact me at bschweber@edn.com.