The Amazon Effect on Jobs
Media coverage of the Amazon expansion keeps suggesting their growth comes at the expense of brick and mortar retail stores. The employment data suggests Amazon contributes to a decline in retail jobs that derives from more causes and long term trends.
Back in 1990 the Bureau of Labor Statistics (BLS) reported electronic shopping and electronic auctions employed a monthly average of 40.4 thousand people while 112.5 thousand worked at mail order houses. By 2016 the BLS reported 233.5 thousand employed in electronic shopping and electronic auctions, but mail order houses showed only a small increase to 127.5 thousand jobs.
As Internet access expanded to millions, the difference between the two categories faded as mail order houses doing catalog sales started using the Internet to compete. In response the BLS combined the two series after 2016. Combined employment since 1990 has an annual growth rate of 3.34 percent, more than triple the national average. The monthly average employment for 2018 came to 398.7 thousand jobs.
While the employment totals in retail trade continue to increase, jobs go up so slowly that the share of retail employment in national establishment employment declined year by year since 1990. In 1990 retail was 12.04 percent while in 2018 it was 10.62 percent. It may sound small but if retail employment retained its 1990 share in 2018, retail employment would have 2.114 million more jobs than it does.
Retail sub sectors in motor vehicle and parts dealers, furniture and home furnishing stores, electronic and appliance stores, food and beverage stores, health and personal care stores, gasoline stations, clothing and clothing accessories stores, sporting goods hobby book and music stores, general merchandise stores, office, supplies, stationery and gift stores all have a smaller share of jobs in 2018 than 1990, and smaller shares in 2018 than 2017.
In addition to electronic shopping a few other retail sub sectors have job growth that standout from the general decline. Home centers are a sub sector exception, which has a higher share of establishment employment with annual job growth of 2.89 percent since 1990. The Home Center job increase has come at the expense of job declines at paint-wallpaper stores, and hardware stores.
Employment at used car dealers has increased at nearly five times the annual growth rate for new cars since 1990: new cars .74 percent, used cars 3.59 percent. Jobs at cosmetic and beauty supply stores increased at a growth rate more than 7 times the rate for pharmacy and drug stores: .53 percent compared to 3.72 percent. Stores selling used goods such as thrift stores and consignment stores have the second highest annual rate of job growth in retail, 4.19 percent; only the electronic shopping that includes Amazon has higher job growth. Pet stores also have a high rate of job growth since 1990: 3.53 percent. These jobs combined have only 8.5 percent of 2018 retail employment, and .9 percent of national employment.
The following retail sub sectors all lost jobs from 2017 to 2018: household appliance stores, electronics stores, pharmacy and drug stores, clothing stores, shoe stores, sporting goods stores, hobby toy and game stores, sewing needle work and piece work stores, music instruments and supply stores, book periodical and music stores, department stores, warehouse clubs and super centers, office supply and stationery stores, gift, novelty and souvenir stores.
The electronic shopping total of 398.7 thousand jobs in 2018 represents 2.5 percent of retail trade employment compared to 2010 when 249.8 thousand jobs were 1.6 percent of retail jobs, a mere .9 percent gain for electronic shopping. Since the end of the last recession in early 2010, electronic shopping has added only 148.9 thousand jobs, not many jobs to replace the 2.114 million jobs there would be without such a low, average growth in retail trade. The Amazon effect is a part of the loss of retail jobs, but the expansion of electronic shopping will bring a net loss of jobs in retail and the economy.