Amazon jobs may not be worth all the tax breaks cities give, study says

Take a peek inside an Amazon fulfillment center

Amazon plans to open a 855,000-square-foot warehouse in Fresno expecting to employ about 1,500 workers when it opens.
Up Next
Amazon plans to open a 855,000-square-foot warehouse in Fresno expecting to employ about 1,500 workers when it opens.

Amazon's new fulfillment center under construction in southwest Fresno will employ 1,500 workers when it opens this year. The company's decision last summer to locate the center in Fresno came only after the city affirmed a package of economic incentives that will be worth up to $30 million over the next 30 years.

The rebates that the company will get on property and sales taxes are among the types of job-growing strategies used by communities across the country to entice Amazon and other companies. But the effectiveness of such incentives is being questioned by a Washington, D.C., think tank's economic analysis of the employment effects of Amazon's fulfillment centers.

In the Economic Policy Institute's report entitled "Unfulfilled Promises," EPI economist Ben Zipperer and economic analyst Janelle Jones concluded that on average, "Amazon's fulfillment centers are ineffective at providing net job growth."

"When Amazon opens a new fulfillment center, the host county gains roughly 30 percent more warehousing and storage jobs but no new net jobs overall, as the jobs created in warehousing and storage are likely offset by job losses in other industries," the report states.

Zipperer told The Bee that while Amazon certainly generates job growth within the warehousing industry sector when it opens one of its massive order-filling centers in a community, it does not appear to translate into employment gains across the entire local economy. "It seems to be shifting the composition of employment in some places," Zipperer said. "You do see more warehousing workers. They are definitely hiring."

In Fresno County, where total employment amounted to about 393,000 workers across all industries last fall, warehouse employment is a relatively small component, according to figures from the state Employment Development Department and the federal Bureau of Labor Statistics. A total of 34 companies operated general warehouse/storage businesses – not including agricultural cold storage – and collectively employed fewer than 1,200 workers in the county.

That's well below peak warehouse employment of almost 1,600 workers in 2006, before the economic recession, but it's still a rebound from fewer than 830 workers in the recession's aftermath in 2011.

How many Amazon jobs?

When Amazon's warehouse opens, probably this summer, the company expects to have about 1,500 employees. Nearby to the west, a new Ulta Beauty distribution center is also expected to open this summer with more than 540 employees. Together, they will more than double the ranks of warehouse employees in Fresno County. The number of workers at both centers is expected to expand seasonally to handle holiday rush periods.

But the EPI researchers suggest that in a community's overall economy, the effects of an Amazon warehouse coming to town may not be all they're cracked up to be. The overall number of people employed in a county after Amazon arrives is relatively flat when compared to before the company opened a center.

In the report, the EPI economists offer two possible explanations: "That the jobs created in the warehouse and storage sector are offset by job losses in other industries, or that the employment growth generated by Amazon is too small to meaningfully detect in the data."

The report asserts that instead of providing millions in tax rebates and other economic incentives to lure Amazon in hopes of stimulating job growth across a local economy, "state and local governments should invest in public services (particularly in early-childhood education and infrastructure) that are proven to spur long-term economic development."

"These (economic incentive) strategies sacrifice future tax revenue and hope there is some kind of payoff," Zipperer said. "They're taking revenue that we would have had to spend on something else and instead spend it on Amazon in the hope that it will result in more jobs for everybody.

"What we're finding is, that sacrifice is a very risky bet," he added. "A lot of the things we know communities want their governments to spend that money on that generate employment – things like education and infrastructure – tend to have much higher job multipliers."

In Fresno, however, where the incentives are the result of a job-growth initiative championed by then-Fresno City Councilman – and now mayor – Lee Brand, the EPI study is a subject of scorn and skepticism.

"It's hard to take any study seriously that only does enough research to get the answers they want, and clearly EPI wanted to try to disprove Amazon's economic benefit to the communities where they locate," Brand said. "Any Fresno high school student can tell you that the addition of 2,500 jobs starting at $15 an hour with benefits at the full operation of the Amazon fulfillment center will increase employment and reduce poverty in the area."

"We doubly benefit because all the tax incentives for Amazon and Ulta are conditioned on those job promises and we expect nothing but a positive impact from the presence of these new e-commerce centers in our city," Brand said.

What is Fresno putting up?

Brand's Economic Expansion Act, approved by the City Council two years ago, calls for Fresno to rebate 90 percent of the city's share of property taxes that Amazon's subsidiary, Golden State FC, would pay on the increased value of the site for the next 30 years, as well as a rebate of the city's entire share of sales and use taxes paid by the company on purchases it makes in Fresno.

To qualify for those incentives, however, Golden State FC is required to create at least 750 new full-time jobs at the center. The incentives are hard-capped at $30 million, depending on the number of jobs created by the company.

An economic analysis prepared for the city's Economic Development Department estimated that the actual property tax rebates would amount to about $15.3 million over 30 years, plus about $750,000 in sales tax rebates. That same analysis, conducted before annual increases to California's minimum wage kicked in, anticipated that the potential payroll at the Amazon fulfillment center would add up to more than $2.2 billion over 30 years – money that would circulate within the community as employees spend at local businesses, buy homes, and pay sales and property taxes.

Amazon weighed in with its own rebuttal of the EPI study, insisting that its warehouses do stimulate additional jobs in communities beyond the company's direct hiring. "In addition to the 200,000 Amazon employees in the U.S., we know … Amazon's investments led to the creation of 200,000 additional non-Amazon jobs, ranging from construction jobs to healthcare industry positions," the company said in a written statement. "Over the last five years, counties that have received Amazon investment have seen the unemployment rate drop by 4.8 percentage points on average, and in some areas, the rate has been lower than the state average."

An Amazon spokeswoman said the company's estimates are based on modeling methods used by the U.S. Bureau of Economic Analysis.

In many areas of the country, however, the five-year period cited by Amazon has been characterized by economic recovery and corresponding employment growth, seemingly making it difficult for Amazon – or anyone else – to know definitively whether all of the additional job creation claimed by Amazon are directly attributable to the company, or due to organic growth that would have happened in the local economy regardless of Amazon.

In Fresno County, overall private-sector employment grew by more than 31,000 jobs from 2012 through 2016 – an increase of about 11.2 percent before Amazon even formally had Fresno in its crosshairs. The warehouse sector, in the meantime, gained just over 200 jobs.

Amazon map 2.JPG
Online retail giant Amazon expects to employ 1,500 people at a new order-filling center that will open in the summer of 2018. Tim Sheehan tsheehan@fresnobee.com

Amazon has fulfillment centers that it either operates or uses as outsourced warehouses scattered across the state. It has six in San Bernardino County alone. Prior to Amazon opening its first center there in 2012, fewer than 10,000 people worked in warehouse jobs. As of last summer, those jobs had nearly tripled to almost 30,000 – the effects of not only Amazon, but other companies opening warehouses there as the Inland Empire becomes a major logistics center serving Southern California.

Even at that pace of industry growth, warehouse employment was less than 5 percent of San Bernardino County's overall job pool.

What about the wages?

According to state and federal estimates, the average weekly wages for warehouse workers in Fresno County's pre-Amazon era climbed by 44 percent since 2000 – from $583 a week in 2000 to $842 a week last fall. Wages actually peaked at an average of $884 a week in 2015. Across all industries in Fresno County, the average wage last fall was about $804 per week.

By comparison, the city's economic analysis of Amazon estimated an average wage of $26,000 per year for employees – or about $500 a week.

In San Bernardino County, the average weekly wage of warehouse workers has actually fallen slightly in the Amazon era, from $728 a week in 2011 to $713 a week last summer – and the loss is greater when inflation is factored in.

In the meantime, wages for workers across all industry sectors in San Bernardino County rose from an average of $730 a week to $811. In four of the six California counties in which Amazon has established fulfillment centers, the average weekly wage for employees in the warehouse sector has gone down, not up. The two exceptions were in Los Angeles County, where the company has a center in Long Beach, and Sacramento County.

With its proximity to the Bay Area over the Altamont Pass, San Joaquin County in the northern San Joaquin Valley has two Amazon centers in Tracy and one in Stockton. Before the first one opened in 2013, warehouse employment was fewer than 5,000 jobs and the average wage was about $915 a week. Last summer, employment had grown to more than 12,600 workers. But the average wage had declined by almost $90 a week, to about $826.

Shelley Burcham, economic development manager with the city of Tracy, said no economic incentives were offered to Amazon to build there, but the city did expedite its planning and permitting processes for the company.

Burcham said that when the first center opened, it started with about 1,500 workers, and now Amazon has about 3,500 in one of the centers. "It's not just entry-level jobs," she said. "They work with folks to train them and encourage them to move up the ladder in the company."

During peak seasons, the number of employees ramps up to about 7,000. "People in the industry talk about this 'Amazon effect,' that in the peak seasons when they need more workers, sometimes workers in those industries may jump from one job or another for a dollar or two more an hour," she said. "But we've not heard that from employees here."

But while the EPI study suggested that Amazon generally doesn't provide a net boost to jobs across a county's overall economy, Burcham said the fulfillment centers in Tracy have created a net gain in jobs.

Economist Jeffrey Michael, director of the Center for Business and Policy Research at Stockton's University of the Pacific, said he generally agrees with EPI's policy conclusion about communities throwing tax incentives at Amazon. "There isn't good justification for public subsidies for Amazon fulfillment centers," he said. "They are going to locate in places that allow them to best serve California population centers regardless of these subsidies."

In San Joaquin and Stanislaus counties, "I have no doubt that Amazon fulfillment centers are net creators of jobs," Michael added. "Amazon fulfillment centers are displacing some traditional retail jobs, but it is also true that they are creating new economic activity – basically as acting as an errand-running/shopping service for households."

Michael noted that increases in warehousing employment in a community also create more transportation jobs and non-payroll jobs for independent contractors..

Zipperer acknowledged that the EPI study's conclusions are averaged based on counties nationwide, and that some communities might have experienced overall job growth in the wake of Amazon.

"What we did is not only look at before and after an Amazon warehouse opens, but we try to compare it to before and after in similar places where Amazon didn't open," Zipperer said. "We try to form a good comparison between a county that looks (economically) like that county, but doesn't have Amazon."

"I'm pretty agnostic on how much of an effect that economic incentive packages have" in job creation, he added. "Research shows that they're pretty much a mixed bag. … I think I would have been surprised had we found a result that showed either a very strong (positive) or negative effect."

Related stories from Fresno Bee