In July 2017, then–Wisconsin governor Scott Walker announced that Taiwan-based tech manufacturer Foxconn would build a $10 billion display panel plant in the Badger State, creating 13,000 jobs. The project would, of course, be boosted by state and local government inducements, including subsidies, tax breaks, infrastructure, and public seizure of private land for the facility. The project quickly became a symbol of then-president Donald Trump’s “Make America Great Again” agenda, and Trump went so far as to scout possible sites for the plant from a helicopter alongside Foxconn founder and CEO Terry Gou. Trump, Walker, and Gou did the ceremonial shoveling of dirt at the groundbreaking in June 2018 in the Racine County town of Mount Pleasant in the state’s southeast corner, with Trump proclaiming the plant would be “the Eighth Wonder of the World.”

Yet, as the project proceeded, skepticism grew that it would resemble anything like what Walker, Trump, and Gou had touted. The skeptics have subsequently been proven right, as Foxconn has repeatedly scaled back its ambitions and changed the plant's output. Currently, the facility is manufacturing servers and 5G networking gear, and Foxconn now expects to invest just $672 million and employ fewer than 1,500 people at the site.

This tale is chronicled in the book Foxconned by corporate communications executive and independent writer Lawrence Tabak. The book does such a good job of weaving together economics, history, and politics that it is being published by the University of Chicago Press. It effectively illustrates what can go wrong when government officials try to orchestrate economic development.

Corporate welfare / Tabak describes how "Wisconsin [promised it] would provide up to $3 billion in incentives based on a 17 percent rebate of Foxconn's payroll and a 10 percent rebate of its capital investment," while Mount Pleasant handled the acquisition of the real estate and the building of the infrastructure.

Mount Pleasant officials paid for the property they took, but many of those payments were shady. The officials offered some owners of large tracts $50,000 per acre, but others got "predevelopment market prices, which could be as low as $5,000 per acre." Officials notified owners of small tracts that their properties were in the path of highway construction and offered them "market value plus 40 percent." But they also misled homeowners who balked at selling into believing that they would lose the ability to drive to and from their homes.

Mount Pleasant officials also voted to declare the properties blighted. Tabak points out that "one of the qualifying criteria is a crime rate three times the rate of the surrounding community, which is nonsensical for the Foxconn property." All but one official ignored that requirement.

Officials pulled off this "land grab" without voter approval. "In fact," Tabak writes, "all the decisions had been made behind closed doors, with the help of economic development staffers in Madison and Milwaukee, lawyers and contractors, including [Foxconn advising firm] Ernst & Young's people in Chicago."

The Wisconsin Economic Development Corporation (WEDC) played a key role in recruiting Foxconn. Tabak reports that the WEDC's budget was $43 million in 2017, about nine-tenths of which came from taxpayers. About $29 million of that went to "salary and operating expenses," with most of the balance going to "grants." "In other words," he interprets, "in Wisconsin it cost taxpayers some $30 million to find companies that deserve $10 million in grants." In making this barb, Tabak is being kind; if, as he reports, taxpayers footed around 90% of the WEDC's budget, the cost to taxpayers was more like $39 million.

Sham studies / Proponents of taxpayer-funded economic development tout "economic impact studies" to justify using tax dollars for these projects. Ernst & Young produced one for the Foxconn deal. According to Tabak, "E&Y used established benchmarks to predict that for every ten jobs Foxconn creates, another seventeen jobs would be created in Wisconsin." They further estimated that a typical Foxconn job would pay a salary of $53,875. But some critics dismissed those numbers. Noting Mount Pleasant's nearness to the Illinois border, they asked how many Illinoisans would get the new jobs and whether Wisconsin businesses would receive all the Foxconn capital expenditures.

Other critics, such as John Locke Foundation economist Roy Cordato, were dismissive of such studies altogether. Cordato told Tabak:

The main flaw in economic impact studies is that they do not look at alternative uses of resources, do not provide a true cost/benefit analysis. In fact, the most common models, such as the IMPLAN one used by E&Y for their Foxconn study, are incapable of producing a negative number.

Tabak cites a 2019 report by George Mason University's Mercatus Center on such economic analyses, writing:

They used Wisconsin's Foxconn project as their core case study and concluded through an analysis of likely scenarios and alternative uses of funds expended that the net effect for Wisconsin would not be the economic windfall promised by the consultants but a long-term burden on the state and its taxpayers.

Yet, despite the warnings of economists, industry observers, and other skeptics of cronyism, politicians pushed the Foxconn deal forward.

TIF abuse / Government officials who imagine that they foster economic development rely on the "tool" of tax-increment financing (TIF). Tabak explains how TIF works:

An area that is in need of improvement is typically designated as "blighted," and the property tax revenue — which normally goes to various recipients such as schools — is frozen at the current level, most commonly for twenty-three years. Public funds are then used to make the area more attractive — razing slums, improving roads, adding infrastructure. In a successful TIF district, the subsequent improvements spark development that increases the tax base; the increase, or "increment," over the frozen amount is funneled into the TIF authority that oversees the district. The authority then pays back any debt or costs incurred by the TIF development, until the district either matures after twenty-three years or sooner if the debts are retired.

Manufacturers need land and infrastructure. They can pay for it themselves or lobby government officials for subsidies. Officials use TIF because "it allows municipalities to borrow money to invest in land and infrastructure that in turn can attract corporate and industrial development." "But," Tabak warns, "TIF also comes with risk and the potential for abuse." The risk is whether a manufacturer will actually make investments and produce what consumers want. The abuse refers to misspending TIF funds.

In the case of Foxconn, Tabak suggests that officials spent too much compensating businesses to move off seized land, authorized unnecessary purchases for the fire department, and paid too much for "professional services." The main problem may be that Foxconn never really planned to produce flat-screen TVs at the site; as noted, its capital expenditures have ultimately amounted to less than $1 billion. The economy of southeastern Wisconsin did not boom as a result of the deal.

To prevent future economic development fiascoes, some state legislatures have imposed limits on the use of TIF. One legal limit is the maximum area of a city that may be designated for TIF. Another is the maximum amount government officials may agree to spend on a deal that uses TIF. A third restraint is legislation that prohibits government officials from transferring property from one private party to another. But officials manage to circumvent these legislative fixes. According to Tabak, for instance, the Wisconsin legislature made an exception that allowed Mount Pleasant to spend more than the legal maximum on the Foxconn deal. And although Tabak characterizes the real estate around the town as "pristine and well maintained," government officials declared it blighted, took it, and gave it to Foxconn.

Part of the problem with industrial policy, as Tabak sees it, is that government officials don't have the necessary expertise. He reckons that Wisconsin officials lacked knowledge of "Asian business culture, the psychology of auctions, and the competitive realities of flat-panel LCD manufacturing." The general problem is that no one can foretell the industries and firms of the future. He quotes Vivek Wadhwa, a Silicon Valley entrepreneur with academic appointments at Harvard Law School and Carnegie Mellon: "Most of the top-down cluster-development projects in the United States and around the world have died a slow death in relative obscurity." He also quotes economic research concluding that "direct, top-down policies are most likely to fail." Perhaps government officials and the citizenry will see the futility of picking winners and leave economic growth to the spontaneous order of the market.

Tabak favors a multilateral agreement among state governors to refrain from offering tax breaks and subsidies to businesses seeking locations. He believes that if Congress set a legal maximum on tax breaks and subsidies, the number of deals would decrease. Deal-making would not disappear, however, if state economic development agencies remain in place with their budgets and vested interests. He correctly identifies these deals as "corporate welfare" and expects both the political left and right to see their common interest in resisting it.

Some errors? / Although Tabak demonstrates a good understanding of economics, he appears to make a few errors.

In a passage where he laments job losses in manufacturing and the harm associated with it, he states, "During the middle of the 'Glorious Thirty,' the unemployment rate for men between twenty-five and fifty-five across the US was 5 percent." By the "Glorious Thirty," he means the 1950s–1970s, which was generally a good time for U.S. employment — indeed, even better than Tabak describes. Taking "the middle" to be the 1960s, the unemployment rate for men between 25 and 54 years old rose from about 4% in 1960 to 5.6% in 1961, and then gradually fell to below 2% in 1969. The median unemployment rate for prime-aged men was 3.0% over the entire decade of the 1960s.

He then turns to unemployment in the wake of the Great Recession of early this century. "In 2010, in the wake of the global financial crisis and subsequent recession," he writes, "the number ballooned to 20 percent and the subsequent recovery largely bypassed this demographic." If, by "the number," he means the unemployment rate for prime-aged men, that number peaked at 10% in September and October of 2009. I corresponded with Tabak and, concerning this passage, he wrote that he may have been referring to the percentage of prime-aged men not in the labor force, which is a larger group than those deemed "unemployed" in government statistics. Labor force participation by prime-aged men has been in long-term decline in the United States, but it still was 89% in 2010, which means that only 11% of prime-age men were not in the labor force.

In another passage where the author discusses wealth inequality, he presents a graph of the percent change in inflation-adjusted wealth beginning in 1989 for various groups. Underneath, he writes, "Since 1989, the wealth of the top 1 percent has skyrocketed by nearly 300 percent." One line in the graph shows just that. Another line shows that the wealth of the bottom 50% rose from 1989 to 1999, fell over the next 10 years, and then recovered by 2018. The upshot is that the bottom 50% had no more wealth in 2018 than in 1989. The author uses data from the Federal Reserve's Distributional Financial Accounts. Using the same data, I calculated that the inflation-adjusted wealth of the top 1% grew 243% from 1989 to the fourth quarter of 2019 — a large gain, but well short of 300%. My calculations also indicated that the inflation-adjusted wealth of the bottom 50% grew by 100%. In my correspondence with Tabak, he stood by his view that there is "striking inequality" and it certainly is true that the wealth gap between the top and bottom of the distribution widened over this period. Still, as I read the data, the wealth of the bottom half doubled over this time, and they are not as asset-poor as Tabak's graph indicates.

Conclusion / There is now a "new contract" between Foxconn and Wisconsin. Foxconn will pay taxes on its property, which will cover interest on the bonds for the land and infrastructure. As for subsidies, the state will now only pay Foxconn up to $80 million. But taxpayers still face the day when the principal on the bonds comes due.

Tabak's work is a persuasive argument against government officials who imagine that they can plan economic development. He recognizes "how inept the public sector has proved itself to be in picking winners." Ironically, however, the reforms he recommends — such as limits or prohibitions on tax subsidies — rely on government officials to do the right thing. Even he must be pessimistic about the prospect of that, given that he believes the political class is beholden to its patrons in the business community.