When Donald Trump makes more sense on the minimum wage than you do, perhaps it’s time to reevaluate your stance.

Seattle entrepreneur Nick Hanauer found himself in this unenviable position last week. While Trump tried to articulate a nuanced wage approach–keep the federal wage requirement unchanged, and allow states to go further–Hanauer argued for a more aggressive stance, pointing to a new report from the labor-backed National Employment Law Project (NELP).

The analysis, which Hanauer describes as “first-of-its-kind,” looked at changes in total private sector employment following past increases in the minimum wage. Finding little relationship, Hanauer exuberantly concluded that it’s “nonsense” to suggest raising minimum wage could have a negative impact of jobs.

Hanauer may be right that this NELP analysis is first of its kind, if only because academic professionals with a PhD after their names wouldn’t identify themselves publicly with a product like this.

Economists have known for years that there aren’t enough people earning the minimum wage to sway total employment one way or the other. Consider: According to the most recent datafrom the Bureau of Labor Statistics, less than four percent of the hourly workforce earns at or below the federal minimum wage of $7.25. That’s just three million people in a combined hourly and salaried workforce of over 150 million people.

Even if one-sixth of those minimum wage-paying jobs disappeared as a consequence of a higher minimum wage, it’d be difficult if not impossible to detect this in the larger churn of the economy. That’s why NELP’s analysis doesn’t pass the laugh test: Economists have invested years–in some cases, a large portion of their careers–studying how to best measure the impact of a higher minimum wage, because it’s a futile task to try and find it in top-line employment numbers.

This isn’t even a controversial proposition. In a recent interview with an NPR affiliate in Boston, Dr. Arin Dube–a UMass-Amherst economist who’s frequently cited as an expert by NELP–explained:  “We don’t have [an ideal experiment] where we randomly assign different minimum wages to different labor markets, instead we have over two or three dozen states that have followed different minimum wages, and we have to track what’s going on to jobs and compare these states with other similar states.”

Dube supports a higher minimum wage, though his research is an outlier compared to (as a recent Federal Reserve Bank of San Francisco paper put it) the “overall body of recent evidence” on this subject. Still, he readily acknowledges that measuring the jobs impact of the minimum wage is “a hard question to answer,” and thus requires more than a superficial analysis of total private sector employment.

What’s most curious about the argument attributed to minimum wage opponents by NELP and Hanauer–that raising the minimum wage “always hurts the whole economy”–is that opponents aren’t actually making this case. If they were, Hanauer would have been able to fill his op-ed with dozens of recent quotes from opponents (like myself) warning that a high minimum wage would crater the “whole” economy.

Instead, a careful look at the quotes that NELP selected for its report demonstrates that opponents are concerned about the policy’s impact on young and less-skilled employees. Here’s their quote from New Jersey representative Jim Saxton in 1996: “[The minimum wage] hurts exactly those workers it intends to help—the poor, the unskilled, and the young.” Here’s their quote from Texas Senator Ted Cruz in 2016: “I think the minimum wage systematically hurts the most vulnerable.”

These sentiments are entirely consistent with the fact-based case against a higher minimum wage, which is that it destroys job opportunities for the least-skilled employees–not that it destroys entire economies. Even an oft-cited 1980 quote from Ronald Reagan about the “misery and unemployment” caused by the minimum wage ignores the fact that his primary concern was young workers, for whom he’d proposed a lower minimum wage.

NELP acknowledged the simplicity of its analysis: “It’s not a regression, academic-type study,” conceded one of the study’s authors in a Washington Post interview about the organization’s “ridiculously simple” approach (the Post’s words, not mine). I couldn’t have said it better myself: With hundreds of thousands of entry-level jobs at stake, this is too important a debate for policymakers to ignore the hard “academic-type” evidence in favor of a “ridiculously simple” study.