CNNMoney

Manufacturing: Not a Magic Pill for the Economy

In 1983, in the midst of a deep recession, Bruce Springsteen recorded the song "My Hometown" that reportedly was based on the closing of the Karagheusian Rug Mill in Springsteen's own home town of Freehold, N.J.

The song's theme of economic hardship from the displacement of manufacturing jobs resonated, propelling it to 6th place on the Billboard charts.

Today, in the midst of an even deeper and more prolonged recession, the 1980s debate about whether manufacturing matters is front and center.

In his State of the Union speech, President Obama said that the blueprint for economic recovery "begins with American manufacturing."

There are, of course, welcome developments. But anecdotes about some firms' successes are not representative of the performance of an entire industry. It is magical thinking to believe that manufacturing can reclaim the role it had in the mid-20th century, and be the main driver behind the resurgence of today's economy.

Let's start with some perspective: the 300,000 new manufacturing jobs created since the depths of the Great Recession represent only 8% of total job growth.

This is less than proportionate to the relative size of manufacturing. Its current share of employment is only about 9% of the nation's overall total.

In the 1980s, manufacturing employment commanded a 21% share of the overall total. Over the past three decades, employment in manufacturing has decreased about 40%.

So while manufacturing has been a bright spot lately, this is a story of productivity gains, not of employment growth.

Thanks to productivity gains, the employment drop occurred while the value added by manufacturing increased by 40%. Hourly compensation to workers has remained stagnant. So the question arises: who benefits from policies to support manufacturing, workers or owners?

It is easy for a casual observer and politician to attribute an outsized role to manufacturing in employment terms, given the attention afforded factory closings or hiring surges by a particular firm.

Manufacturing is characterized by "churning" -- simultaneous job creation and destruction. On average, about one in five manufacturing jobs are either destroyed or created each year, and that churn is not especially concentrated within some narrowly defined manufacturing sector.

And job destruction occurs in good times as well as bad; the Karagheusian Rug Mill closed in 1964 in the middle of a decade-long national economic expansion.

The U.S. experience is shared by other countries, formerly known as "Industrial Nations." Over the past three decades (up to just before the most recent recession), the share of manufacturing employment in the United Kingdom fell from 25 % to 9%.

Even Germany and Japan -- two countries seen as manufacturing powerhouses -- have had substantial declines in the share of employment in manufacturing, declining from 27% in Germany in 1991 to 19% in 2007, and in Japan from 32% to 20% over the same period.

The case has also been made that manufacturing matters because of exporting.

A bit more than half of all U.S. exports are manufactured goods, and two-thirds of these manufacturing exports come from four sectors: chemicals, transportation equipment, computers and electronic products, and machinery.

Policies to promote exporting would, therefore, disproportionately favor a relatively small set of firms in these sectors.

How small a set? Only about 4% of manufacturing firms in the U.S. exported in 2000 and 96% of all U.S. exports are sold by just 10% of this already small set of firms.

It may be possible to expand the set of firms that export, rather than just the export activity of those that already sell abroad, but the extreme concentration of exporting gives one some pause about export-promoting policies.

And exporting is not an end in itself. Is there some special feature of exporting that benefits workers as well as owners?

There is evidence of a wage premium paid to workers in exporting firms, but those workers also tend to have more education and skills than those in non-exporting firms. Part of the premium is due to this.

Since higher education and skills result in higher wages, it would be well worth considering policies promoting the skills and education of workers, regardless of the industry in which they are employed.

Safety nets for those suffering from economic dislocation, regardless of the industry in which they had worked, are also important policies in a modern, dynamic economy on both equity and efficiency grounds.

It is much less apparent that the same can be said of policies targeted to manufacturing industries.