Planners’ forecast for Ford plant was comically wrong; will anybody be held to account?
Michigan seems to be ending 2023 the way it began, with surprise news about Ford Motor Company’s $1.7 billion, taxpayer-subsidized deal to build an electric vehicle battery plant in Marshall. In February, the head of a local development authority predicted the plant would directly employ 2,500 people. In late November, Ford announced a scaled-down version that would employ just 1,700. That’s a 32% reduction – just in the number of promised jobs, not even jobs that may ultimately exist.
The Ford case is more evidence that Lansing politicians and state officials should not be in the corporate handout business. They have no special ability to predict which of their corporate supplicants will flourish or fail. Lawmakers buy positive headlines for themselves at great cost to Michigan taxpayers, even selling the alleged future benefits under the façade of science. The track record of state and local subsidies to corporations is abysmal, and it should outrage voters and taxpayers alike.
The Ford deal was trumpeted in a clichéd fashion. Politicians and corporate leaders held a press event to announce the project, and a statement from the Michigan Economic Development Corporation bragged that the plant would create 2,500 new jobs. Gov. Whitmer called it a “generational investment,” whatever that means. The agency declared that the project had a jobs “multiplier” of precisely 4.38, meaning that every job added directly at the Marshall plant would lead to 3.38 jobs being created elsewhere in Michigan.
Development corporation officials were very confident about this — that the multiplier was not, for instance, 4.37 or 4.39 — because their computers told them so. State officials use a software model called REMI (from the consultancy Regional Economic Modeling, Inc.) to make forecasts like these. While it has its legitimate uses, REMI does not have magical powers. The reliability of its output is only as good as the quality of the data being fed into it. In this case, the old programmers’ adage was precisely true: If garbage goes in, garbage comes out.
The MEDC’s precise multiplier is based on a bunch of assumptions. Can state officials produce accurate assumptions? Ford bigwigs couldn’t accurately foretell in February what the company would be doing nine months later. At the beginning of the year, Ford said it would create 2,500 new jobs to manufacture batteries, then it put the whole project on pause for a while, before announcing that the plant probably would only create 1,700 jobs after all. The state bureaucrats’ REMI model output was built in part on a jobs number that was off by more than 30%.
The irony is rich. The Michigan Economic Development Corporation’s REMI model can’t forecast the future because state officials can’t forecast the future. This Ford project is hardly the only example.
Another is from a $15.9 million incentive deal for Kmart in 2000. Officials claimed that by using REMI they could predict precisely how many jobs would be created in Michigan, directly and indirectly, in retail and manufacturing sectors among others, from 2000 to 2013. What they didn’t see or predict was that Kmart would be bankrupt in 17 months.
The predictions made by economic forecasts, even sophisticated ones like REMI, should be viewed with caution even when responsible modelers use accurate assumptions. But there is ample evidence to suggest that the development corporation and some of its contractors have not. Previous program research has shown that REMI assumptions falsely presume the state’s incentive is 100% responsible for the project and related jobs. That is just not true.
Maybe all of these erroneous forecasts would be tolerable if they were at least in the ballpark, or if the jobs programs delivered outstanding successes. They are not and they have not. In 2014 my colleagues found that just 2.3% of past projects delivered the predicted results. Many studies before and after that one show that such corporate subsidy programs fail to generate promised benefits.
Evidence shows that members of the political class themselves benefit greatly from such programs, though their constituents don’t get much in the way of advantages. This suggests that these corporate subsidy programs are just masquerading as economic tools. Their true function is to give elected officials a means for handing out goodies to politically connected CEOs and corporations.
That’s how incentives work to politicians’ advantage. State bureaucrats and lawmakers sell their claims with the help of models like REMI. In his Cato Journal article, “The Misuse of Regional Economic Models,” Edwin Mills explained that models like REMI are employed to sell the public on spending:
This is where REMI is so valuable. It is a complex computer model that lay people cannot understand or evaluate, and it has important scientific merits. Thus, the frequent government claim that the best scientific model available shows that x thousand jobs will be created by the project helps to carry the day.
Michigan taxpayers and voters deserve to be treated better by crystal ball-gazing politicians and state officials. Instead of offering billions in ineffective subsidies and bamboozling us with promises of a jobs nirvana, they could just stick to government’s knitting. Keep taxes reasonable and uncomplicated, lighten regulations on job providers, and provide key, quality public services. The economy doesn’t need politicians to pick winners and losers; it can function fine all on its own.
Permission to reprint this blog post in whole or in part is hereby granted, provided that the author (or authors) and the Mackinac Center for Public Policy are properly cited.