Can You Measure the Supreme Court Like GDP?Spoiler: Probably Not.

For years, economists, political scientists, and journalists have tried to quantify the United States Supreme Court as if it were a quarterly earnings report. The idea is appealing; reduce complex judicial decision-making into a clean, digestible metric—something akin to GDP or election results.

The reality? Not so simple.

A recent study from the National Bureau of Economic Research “Ruling for the Rich: The Supreme Court Over Time” study —attempts to do exactly that. The authors argue that economic metrics may better predict Supreme Court decisions than traditional legal frameworks like originalism or textualism. The paper even garnered attention in The New York Times but beneath the headlines lies a methodology that feels less like rigorous SCOTUS analysis and more like a thought experiment that forgot to invite reality. Or, as lawyers might say: interesting theory—unsupported by the record.

When “Data” Depends on Who You Ask

The study relies on Yale students to classify decades of Supreme Court cases based on whether each decision shifted resources between “the wealthy” and “the poor.” That classification then drives conclusions about pro-business rulings and alleged judicial bias. To their credit, the authors acknowledge subjectivity. To their detriment, they proceed anyway.  Only 10% of the coding was double-checked—and even then, reviewers agreed just 64.5% of the time. Translation: in over one-third of cases, two evaluators could not agree on what the case meant economically. For a study making sweeping claims about Supreme Court bias, that’s less statistical rigor and more educated guesswork with citations.

When Every Case Becomes About Money (Even When It Isn’t)

Consider Massachusetts v. EPA (2007). The study treats it as redistributing resources from the wealthy to the poor.  But that case is fundamentally about administrative authority and environmental regulation—not a direct economic transfer between classes. It’s a dispute over federal power, not a line item in a balance sheet.  Meanwhile, cases like Bell Atlantic Corp. v. Twombly (2007) (which significantly impacted business litigation by tightening pleading standards—are not given the same analytical spotlight.

That disconnect did not go unnoticed. Professor Jonathan H. Adler criticized the study’s assumptions, while Professor Mitu Gulati noted that relying on undergraduate coders for such nuanced legal determinations gave him “the heebie-jeebies.” A rare moment of consensus across disciplines.

What the study Misses: How Litigation Actually Works

Here’s the part missing from the model: litigation strategy. By the time a case reaches the Supreme Court, it is not random. It is the result of years of strategic decision-making—especially in business litigation. Supreme Court appeals are expensive. Elite advocates command hourly rates approaching $3,000. Briefing alone costs tens of thousands of dollars. Total litigation costs can easily reach into the millions.

Businesses are not guessing when they take cases to the Supreme Court. They are investing. In practice, companies typically pursue Supreme Court review only when:

    •    The issue is existential;

    •    The likelihood of success is high; or

    •    The financial burden is offset.

In other words, businesses are highly selective litigants. They bring cases they expect to win.

Selection Bias ≠ Supreme Court Bias

This is where the study’s central thesis begins to unravel. If businesses disproportionately win before the Supreme Court, that may not reflect pro-business Supreme Court decisions. It may reflect the simple reality that businesses bring stronger, more strategically chosen cases.  Advocacy groups, governments, and criminal defendants operate under different incentives. They often pursue litigation for policy or necessity—not just return on investment. When one group selects only high-probability cases and another does not, the outcomes will look skewed. That’s not ideology. That’s selection bias.  Or, more bluntly: if you only take good bets, you tend to win more.

The Limits of Quantifying SCOTUS

There is nothing inherently wrong with applying economic analysis to the law. But any serious attempt to measure the Supreme Court must account for litigation strategy, case selection, and the limits of subjective classification.  Otherwise, we risk mistaking simplified models for meaningful insights—and confusing correlation with causation.  Or, in the spirit of the economists and an old joke: if your model requires assuming a can opener, it may be time to check the kitchen.