Delegation Markers, Residual Vagueness, and State Win Rates in Korean Regulatory Adjudication
This paper empirically tests whether express delegation clauses in Korean statutes effectively transfer interpretive authority to agencies by analyzing 8,268 adjudication cases, finding that courts actually rule against the state more frequently when citing delegation-dominant provisions compared to indeterminacy-dominant ones, thereby suggesting that formal delegation markers do not necessarily result in the expected deference.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
In the world of law and government, there is a constant tension between writing rules that are clear and writing rules that are flexible. When a legislature creates a law to manage a complex industry, like food safety or finance, it cannot possibly predict every future situation. To handle this, lawmakers leave gaps in the text. They might explicitly say, "The Minister will decide the details," handing the power to fill those gaps to a specific agency. This is a clear transfer of authority. Alternatively, they might use vague language like "reasonable measures" or "where necessary," leaving the meaning open to interpretation without pointing to a single decision-maker. For decades, legal scholars have debated whether these different ways of writing laws lead to different outcomes. Does giving an agency clear permission to make rules make the government stronger in court, or does it make the government more vulnerable to challenges? The question matters because it touches on how much power the government should have versus how much control the courts should keep, and whether industries can unfairly influence the rules that govern them.
A team of researchers set out to answer this question by looking at thousands of real court cases in South Korea. Instead of trying to guess what judges were thinking, they focused on who actually won. They gathered a massive collection of published court decisions involving government regulations and taxes, covering more than 8,000 specific instances where a law was cited. They then sorted these citations into two groups based on the language used in the law itself. One group contained laws that explicitly delegated power to an agency, while the other group contained laws that relied on vague, indeterminate concepts. The researchers then calculated how often the government side won when the court cited a law from the first group compared to the second.
The results were surprising and ran counter to a common intuition. If courts simply respected the government's authority when a law explicitly gave it power, the government should have won more often in those cases. Instead, the data showed the opposite. When the court cited a law with an explicit delegation of power, the government lost more often than when the law was vague. Specifically, the government's win rate dropped by about 7.4 percentage points in cases involving explicit delegation compared to cases involving vague language. This suggests that when a law hands power to an agency in a clear way, it actually invites a specific type of legal attack that is successful. Opponents of the government are not just arguing that the agency made a mistake; they are arguing that the agency stepped outside the boundaries of its authority entirely.
The researchers dug deeper to understand why this happened. They looked at the reasons judges gave for their decisions to see if they were explicitly striking down the government's rules because the law was too broad. They found that judges rarely used the formal language of "delegation limits" to overturn the government. Instead, the government was losing because judges were reading the specific rules the agency created very narrowly. When a law says "the Minister will decide," the agency writes a rule. If that rule is too broad or tries to cover too much ground, a judge might say, "This specific rule does not fit within the narrow scope of the law," and throw it out. The government loses not because the judge is ignoring the law, but because the judge is holding the agency to a strict reading of the authority it was given.
This pattern was not consistent across all time periods. The effect was strongest in cases decided before the year 2000. In more recent decades, the difference in win rates between the two types of laws disappeared, becoming statistically indistinguishable from zero. The researchers noted that this shift might be due to changes in how laws are written or how courts interpret them over time, but they could not say for certain because they were looking at laws that still exist today, which might not perfectly match the laws that were in force decades ago. They also found that this dynamic was specific to administrative and tax cases; in criminal cases, the pattern did not hold, likely because the types of laws used in criminal prosecutions are different and less likely to contain these specific delegation clauses.
The study also revealed a significant challenge in how legal research is often conducted. The researchers initially tried to have legal experts read the court opinions and decide if the judges were being "strict" or "flexible" in their interpretation. They found that even trained lawyers could not agree on what the judges were doing; their opinions varied wildly depending on the specific case. This lack of agreement forced the researchers to abandon the idea of measuring judicial attitude and instead rely on the hard fact of who won the case. This shift from guessing intent to counting outcomes proved to be the key to finding a clear signal in the data.
Ultimately, the paper shows that the way a law is written has a measurable impact on who wins in court, but not in the way many might expect. Giving an agency clear power does not guarantee that the agency will win; in fact, it can make the government more susceptible to losing if the agency oversteps the narrow boundaries of that power. The study does not prove that the government is being captured by special interests or that the courts are biased against the state. Rather, it provides a clear, data-driven map of how different legal texts play out in the real world, showing that the specific words lawmakers choose create different risks for the government when those laws are challenged in court.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.