In 2017, Wisconsin offered Foxconn a record-breaking subsidy to build an LCD factory in the state, only to see the promised factory fall behind schedule and grow progressively smaller. Now, the Wisconsin Department of Administration has requested a reassessment of the costs and benefits to the state regarding the far-tinier facility.
The report, which was conducted by Tim Bartik of the Upjohn Institute for Employment Research, finds that the smaller facility raises the already unusually high cost per job even further. If the subsidy levels in the current contract are kept, each Foxconn job would cost taxpayers about $290,000, Bartik found, compared to $172,000 if Foxconn built the original $10 billion, 13,000-job facility. For comparison, Bartik estimated the subsidies Virginia offered Amazon for its second headquarters amounted to between $10,000 and $13,000 per job.
“The most important conclusion of this analysis is that it is difficult to come up with plausible assumptions under which a revised Foxconn incentive contract, which offers similar credit rates to the original contract, has benefits exceeding costs,” Bartik wrote. “The incentives are so costly per job that it is hard to see how likely benefits will offset these costs.”
While Bartik produced the memo in response to a request from the Wisconsin Department of Administration, he notes that it was produced independently and that its conclusions do not necessarily reflect the views of the department or the Upjohn Institute. He doesn’t know whether Gov. Tony Evers has been briefed on his report, but he says he shared it with Mark Hogan, the head of the Wisconsin Economic Development Corporation, who objected to its conclusions.
Hogan, who served under Evers’ predecessor Scott Walker, said in a statement that the current contract protects Wisconsin taxpayers. “This study makes assumptions that could not occur under the existing ‘performance-based’ contract between WEDC and Foxconn. The plain fact is the company would not be able to retain any incentives if, by the year 2023, it had only created either the 1,500 or 1,800 jobs the study is based on,” Hogan wrote in a statement.
Bartik says he doesn’t find a persuasive defense in the idea that the project has a reasonable cost per job if it goes into default, all clawbacks are promptly paid, and Foxconn still keeps 1,500 jobs in the state.
The analysis comes five months after a Foxconn executive met with Evers and expressed interest in revising the company’s contract with the state. Foxconn hasn’t said what it would push for, but Evers administration documents obtained by The Verge summarize the meeting and the company’s broad requests: updating the contract to reflect the smaller factory, including additional Foxconn subsidiaries, and extending the period Foxconn can qualify for capital investment tax credits.
Such changes make sense for Foxconn. The company has radically scaled back its plans and likely wants assurances that it won’t be found in breach of contract. But Evers was critical of the deal during his campaign and would likely be reluctant to agree to these changes without getting concessions of his own. In this context, the new economic impact assessment can be seen as setting a new baseline for further negotiations.
The Foxconn deal, which was signed by Walker and touted by President Trump, was supposed to be transformational for Wisconsin. As the first large-panel LCD factory in the Western Hemisphere, it was supposed to span 20 million square feet and create 13,000 jobs. Proponents of the deal said the benefits more than justified the considerable costs: at least $4.5 billion in subsidies and other incentives that the Wisconsin Legislative Fiscal Bureau estimated wouldn’t be recouped until 2043.
The Fiscal Bureau analysis was based on a best-case scenario. It relied on economic impact estimates supplied by the consulting firm EY (formerly Ernst & Young), which Foxconn had hired to pitch its project. It also assumed Foxconn would actually build what it promised and hire at an extremely fast rate. Instead, the company has repeatedly scaled back its plans and fallen far behind on hiring. Rather than a 20 million-square-foot factory manufacturing large LCD screens, Foxconn says the factory it’s now building will be less than 1 million square feet and make smaller screens. While the company had initially planned to employ 5,200 people by next year, it now says the new factory will employ only 1,500 people. Even that seems like a stretch goal: at the end of 2018, Foxconn employed only 156 people in the state.
Bartik’s assessment looks at the economic impact of a facility requiring $2 billion in investment and employing either 1,500 or 1,800 people. His assessment is the latest in a series of moves that indicate that the Evers administration is trying to evaluate what Foxconn is building and what it will mean for the state. In June, the Evers administration also hired LCD industry analyst Bob O’Brien to help officials clarify what Foxconn is doing.
Assessing Foxconn’s impact has been difficult partly because, even as the company has repeatedly changed its plans, it has insisted it will still invest $10 billion in Wisconsin and create 13,000 jobs. As recently as June, Foxconn told The Verge that the factory it’s currently building is only the first step. What the subsequent steps will be or when they might happen was never explained.
Nevertheless, the further Foxconn falls behind the timeline that’s laid out in its contract, the less tenable these claims become and the more pressure there is for Foxconn and Wisconsin officials to come to the table and revisit what the company is building and what the state will pay for it.
Skeptics of the deal, like Gordon Hintz, the Democratic State House minority leader and member of the Wisconsin Economic Development Corporation board, support the reassessment.
“It’s important that the public whose tax dollars are at risk for this project have reasonable expectations about what the costs are going to be per job and what they can expect if what is currently being discussed comes to fruition,” Hintz says. “It shouldn’t be a surprise the project is dramatically smaller with fewer benefits than was originally proposed. Hopefully Foxconn will quit talking about numbers associated with a project that is no longer happening.”
Of course, there’s no guarantee that Foxconn will build what it is now saying it will, either. When The Verge spoke with O’Brien in June, his observations of LCD manufacturing machinery supply chains indicated that Foxconn was unlikely to meet the 2020 deadline it set for itself. If the last two years are any guide, any given Foxconn plan is only good until the next one.