A Computer Might Come After Your Job Next

The automated teller machine wiped out jobs in banking. IBM's Watson might soon do the same for retailers.

First it won at "Jeopardy." Now it might threaten millions of low-wage jobs. At least, that seems to be the implication of a Bloomberg Newsarticleon International Business Machines Corp.'s Watsonsupercomputer.

Developers are now figuring out how to use Watson's processing power to replicate the experience of working with an "experienced in-store salesperson" when shopping for clothes. The software would combine databases provided by retailers with customer preferences for style and fit to help people find what they're looking for.

If it works, this technology would be a boon for everyone who prefers to buy things from the comfort of home. Right now, only a small percentage of shopping occurs online. Shipping costs could be one reason. Another is that many people are hesitant to buy things over the Internet when they can't try them out first, especially clothing. That reticence could be overcome by these new technologies. If a computer knew your body shape and knew the dimensions of each piece of clothing, it could show you exactly how items would fit.

Instead of going to stores where you would have to struggle to find sales help, compete with other customers, browse from a limited selection, wait to try on clothes, and then stand in line before paying, you could just talk to Watson about what you wanted and have it trawl through its list of what's available. Then the clothes could be shipped from a warehouse. Retailers could save a lot of money by reducing or getting rid of their stores and the workers employed there. At least some of those savings would get passed along to consumers in the form of lower prices.

People who work in brick-and-mortar stores may not be as sanguine as the rest of us about these developments. According to the Bureau of Labor Statistics, there are more than 1.4 million Americans working in clothing and clothing accessories stores and another 1.5 million work in department stores.

Economists who study these issues think that the rise of the machines has contributed to the gradualpolarizationof the workforce into elites and relatively low-wage laborers. Until now, however, those poorly-paid workers -- including more than 15 million in retail -- were relatively safe compared with their middle-class brethren.

How many of those jobs will be obviated by the introduction of sophisticated computer software? Will the people displaced by machines be able to find gainful employment in the future? The track record of employment growth in sectors with low or declining productivity isn't encouraging.

This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.

    To contact the author on this story:
    Matthew C Klein

    Before it's here, it's on the Bloomberg Terminal.