Internet Explorer 11 is not supported

For optimal browsing, we recommend Chrome, Firefox or Safari browsers.

Analysis: The Winners and Losers of Retail's Transformation

Retail employment data for local areas suggests some places are benefiting significantly from e-commerce, while much of the rest of the country is lagging far behind.

By Mike Maciag | July 2018
Many types of brick-and-mortar retailers are struggling to remain viable, upended by the proliferation of e-commerce. The retail industry isn’t shedding any jobs as a whole. But different segments of the sector are headed in opposite directions, with jobs tied to online sales and distribution far outpacing the rest of the industry. 

One troubling potential consequence of retail’s shift to e-commerce is that many of the newer jobs created typically aren’t spread out very evenly geographically. While brick-and-mortar stores are everywhere, the same can’t be said of distribution facilities and call centers, which are often fairly concentrated. For this reason, how retail's transformation unfolds carries important long-term implications for local economies.

To approximate the effects of the sector’s shift across the country so far, we reviewed county-level job estimates for several types of industries reported in the Labor Department’s Quarterly Census of Employment and Wages. The two industries tracked in federal data considered to best reflect e-commerce-related jobs are "electronic shopping" and warehousing-storage. Over the 10-year period ending in the third quarter of 2017, the two industries created a combined 528,000 jobs, an increase of 59 percent.

Much of this expansion of e-commerce is taking place in relatively few places. Just 31 of the nation’s more than 3,200 counties accounted for half the national growth in the two associated industries over the decade. Only about 14 percent of the U.S. population resides in these 31 counties.

While many of the areas adding the most e-commerce jobs are large urban jurisdictions, several others are exurban and well outside any major population centers. All tend to be in locations favorable for logistics. The following 25 jurisdictions added the highest tallies of jobs in electronic shopping and warehousing over the decade ending in the third quarter of 2017 by our calculations: 

Jurisdiction Jobs Change 2007 Q3 Average 2017 Q3 Average
King County, Wash. 36,549 10,967 47,516
San Bernardino County, Calif. 26,937 8,451 35,388
Riverside County, Calif. 20,482 7,589 28,071
Middlesex County, N.J. 12,207 8,914 21,121
Maricopa County, Ariz. 11,707 14,470 26,177
Dallas County, Texas 11,482 11,179 22,661
San Joaquin County, Calif. 9,937 4,729 14,666
Los Angeles County, Calif. 8,736 23,511 32,247
Hendricks County, Ind. 7,650 3,489 11,074
Tarrant County, Texas 7,529 6,572 14,101
Luzerne County, Pa. 7,343 4,651 11,994
Baltimore City, Md. 6,922 784 7,706
Salt Lake County, Utah 6,676 4,866 11,542
Franklin County, Ohio 6,499 18,568 25,067
Kenosha County, Wis. 6,372 195 6,663
Fulton County, Ga. 6,401 4,829 11,230
Mercer County, N.J. 6,122 596 6,718
Will County, Ill. 6,048 1,276 7,325
Bullitt County, Ky. 5,846 845 6,691
Lehigh County, Pa. 5,649 3,639 9,288
Northampton County, Pa. 5,116 2,040 7,356
Denton County, Texas 4,896 1,214 6,109
Lexington County, S.C. 4,467 243 4,710
Johnson County, Kansas 4,446 2,025 6,471
San Francisco County, Calif. 4,308 3,098 7,405
NOTE: Electronic shopping figures were not published for both quarters for Hendricks, Kenosha and Northampton counties, so changes instead reflect total nonstore retail (of which electronic shopping is a subset.) As a result, totals for these counties include no more than a few hundred jobs not related to e-commerce.
Source: Governing calculations of BLS Quarterly Census of Employment and Wages for NAICS 493 and NAICS 4541.
It's worth noting that the two industries alone don't capture all e-commerce growth in other parts of the economy, such as delivery services.

Meanwhile, the trajectory of traditional brick-and-mortar jobs is headed in another direction, with the employment base shrinking in most localities. Over the decade, approximately 61 percent of counties reporting data experienced declines in total retail jobs when segments of the industry less affected by e-commerce -- food stores, gas stations, vehicle and parts dealers and nonstore retailers -- are excluded. The majority of counties similarly sustained net losses in overall retail employment across the board.

Industries Counties Reporting Data Counties Losing Jobs Share of Counties Losing Jobs Median Jobs Change
Electronic Shopping & Warehousing (1) 608 195 32.1% 33.3%
Retail Affected by E-Commerce (2) 1,194 731 61.2% -3.8%
Retail Affected by E-Commerce (3) 3,114 1,877 60.3% -4.5%
All Retail Sector 3,114 1,745 56.0% -1.9%
SOURCE: Governing calculations of BLS Quarterly Census of Employment and Wages data for Q3 2007, Q3 2017. (See methodology)
  1. Includes counties with reported data for at least one of the two industries for both years.
  2. Represents changes in total retail employment, excluding food stores, gas stations, vehicle and parts dealers and nonstore retailers. Only counties with reported data for all five industry classifications are included.
  3. Represents changes in total retail employment, excluding food stores, gas stations, vehicle and parts dealers and nonstore retailers when reported. Includes all counties, except for those without reported retail job totals.
Large urban and suburban counties with particularly steep losses in traditional brick-and-mortar jobs over the decade included DuPage County, Ill., (-25 percent); Franklin County, Ohio, (-18 percent); and Hamilton County, Ohio, (-18 percent). 

Still, there are many places where brick-and-mortar retail remains strong. Fast-growing regions where Americans are migrating to or those with booming economies experienced particularly sizable gains. Orange County, Fla., added nearly 12,400 such jobs (+28 percent), more than any other county nationally over the decade. 

Retail and E-commerce Jobs Data for U.S. Counties

Make a selection to view retail employment by county and changes in jobs over the 10-year period ending in the third quarter of 2017.

As a whole, retail employment climbed slightly over the decade. Grocery stores, gas stations and other segments of retail with less of an online presence rebounded from losses sustained during the recession. But other types of employers recorded significant declines. Clothing stores, for instance, cut 126,000 jobs over the ten-year period while electronics and appliance stores lost about 79,000 jobs nationally.


Methodology and Data Notes

The Labor Department’s Quarterly Census of Employment and Wages (QCEW) reports industry-level employment data for U.S. counties. Counties’ quarterly private sector job totals were computed by calculating three-month QCEW averages for both Q3 2007 and Q3 2017. The two quarterly averages were then compared over the decade for the following NAICS industries: Retail trade (44-45), electronic shopping and mail-order houses (4541), warehousing and storage (493), nonstore retailers (454), gasoline stations (447), food and beverage stores (445) and motor vehicle and parts dealers (441). E-commerce jobs were considered to be those classified in the electronic shopping or warehousing industries. Companies often classify their establishments inconsistently, with some fulfillment centers classified as electronic shopping and similar facilities classified as warehousing.

Some counties lack quarterly employment data for one or more industries as BLS does not publish local data if no jobs are reported or if necessary to protect employers’ confidentiality when there are only a few establishments for an industry. Industries without reported job totals for both quarters were excluded from our calculations. Warehousing and storage, not classified as a segment of the retail industry by BLS, includes companies not associated with online shopping and e-commerce. It’s unlikely, however, that these other employers account for a large portion of the job growth measured over the decade. Other smaller industry classifications not included are also associated with e-commerce, but data aren’t reported in a way that can be parsed out.

From Our Partners