Donald Trump never tires of reminding us that he is a businessman, and in Betsy DeVos, he has nominated a secretary of Education who endorses a business model for improving elementary and secondary schooling. The problem is, it’s the wrong model.

DeVos’ prescriptions include for-profit school management, taxpayer-funded vouchers to cover private school tuition and parental choice as the primary vehicle for regulation. Yet where such free-market remedies have been tried, they have yielded disappointing results.

Only a few cities in the U.S. implemented voucher systems, but results in these cities — notably, Milwaukee, leading the way in 1990, followed by Cleveland and Washington — have not proved positive. A different way to offer school choice — publicly financed but independently managed charter schools — has proved much more popular. Introduced by Minnesota in 1992, charter schools now number nearly 7,000 across the country, yet they too have posted uneven results, led to greater student segregation and in large part depressed teacher pay.

In no state has this been more true than DeVos’ home state, Michigan, which thanks to her efforts is home to far more commercially managed charter schools than any state in the country. After controlling for demographics, Michigan, according to a recent Urban Institute study, ranks 47th out of all states in reading and math.

The fundamental problem with the free-market model for education is that schools are not groceries. Education is complex and the immediate consumer, after all, is a child or adolescent who can know only so much about how a subject should be taught. The parent, legislator and taxpayer are necessarily at a distance.

Although the free-market model isn’t a good fit for schools, there are five business concepts that should be embraced by education reformers and policymakers.

Much as early stage investment in promising companies can deliver outsized rewards for investors, early stage investment in schooling can deliver significant rewards for society. Chicago economist James Heckman analyzed data from Michigan and North Carolina going back several decades and found that no other infusion of public dollars comes close to matching the rate of return of high-quality early childhood education.

Since the days of Henry Ford, business has understood “efficiency wage theory.” It costs less to pay more, as employers attract and retain better workers and thus improve production and even reduce costs of both supervision and turnover. Studies show a similar tight relationship between teacher pay and educational outcomes.

Retaining good teachers and grooming administrators from within the ranks instead of handing over the reins to outsiders constitutes another significant lesson from the corporate playbook. As the business historian Alfred Chandler documented, great organizations develop talent internally. Education researchers have repeatedly shown, in particular, that teacher turnover impairs student achievement. In addition, superintendents without classroom experience tend to be out of step with pedagogical needs.

Pay for performance — another cardinal objective of business-minded reformers like DeVos — sounds logical but backfires. Instead, reformers should follow the lead of W. Edwards Deming, the father of the modern Japanese auto industry, who contended that merit ratings and pay generate fear and undermine teamwork.

Separate longitudinal studies of merit-based pay for teachers in Nashville and Chicago, completed in 2010 by researchers at Vanderbilt University and Mathematica, bear him out. They found no effect on student achievement.

“Cease dependence on inspection to achieve quality,” Deming wrote.

That recommendation should be applied to the annual testing of students in reading and math mandated by the No Child Left Behind Act in 2001 and reauthorized by the Every Student Succeeds Act in 2015.

Instead of “routine inspection,” Deming urged detailed analysis of small samples. Bucking widespread practice, the Finns do exactly that, with high-quality exams administered to small groups of students. Teachers consequently feel no pressure to “teach to the test,” students get a well-rounded education and administrators gain superior understanding of student progress. Finnish teens score at or near the top of international educational assessments.

Samuel E. Abrams is the director of the National Center for the Study of Privatization in Education at Teachers College, Columbia University. He wrote this for the Los Angeles Times.