One month after South Korea abolished the Prosecutors’ Office and launched the Public Prosecution Office and the Serious Crimes Investigation Agency, the new division between investigation and prosecution is producing longer procedural routes for warrants, evidence checks and jurisdictional decisions. Prosecutors can no longer conduct supplementary investigations directly, leaving them to return cases to police when evidence or paperwork needs correction.
Police data show a marked change in warrant handling during the first weeks of the new system. From October 2 through October 24, police applied for 1,936 detention warrants, of which 701, or 36.2 percent, were not submitted to courts by prosecutors. The non-submission rate had been 22.2 percent from January through September 2026. Over the same comparison periods, the rate for arrest warrants rose from 8.4 percent to 21.0 percent, while the rate for search and seizure warrants increased from 2.8 percent to 11.7 percent.
Warrant review now requires more handoffs
The figures describe different warrant categories and cannot be combined into a single overall rate from the available material. They also do not establish that every additional non-submission resulted solely from the institutional overhaul. Field accounts, however, identify a recurring mechanism: even a simple wording error can be sent back to police because a prosecutor’s direct correction could be interpreted as intervention in an investigation. A correction that once might have been handled within the prosecution process can therefore require a formal supplementary request and another police step.
That pattern is visible in the volume of cases returned for more work. Between October 2 and October 29, Public Prosecution Office prosecutors requested supplementary police investigation in 12,465 cases, an average of 445 a day. Prosecutors may interview people involved or review materials, but evidence obtained through those activities faces restrictions on its use at trial. When additional evidence must be formally secured, police may need to conduct the supplementary work.
The same limitation can affect cases already in court. If a victim presents new evidence during a trial, prosecutors may have to seek additional police investigation or rely on a court production order rather than develop the evidence directly. The available reports do not quantify resulting delays, but the added procedural handoffs create a documented route by which evidence review, victim redress and trial progress can take longer.
The new serious-crimes agency is still building capacity
Operational gaps are also visible at the Busan regional office of the Serious Crimes Investigation Agency. As of October 29, it had received notices concerning 631 police cases, inherited 18 cases from the former prosecution organization and directly received 36 complaints, accusations and petitions from citizens. It had asked for some police cases to be transferred, but none of those requested transfers had been completed. No case involving Ulsan had yet been completed by the Busan office and sent to the Ulsan Public Prosecution Office.
The Busan office was operating with 73.3 percent of its authorized staffing. Construction of its premises was expected to finish around December 22, while access to the criminal justice information system was scheduled for early December. Until that system became available, officials had to request necessary record checks from other institutions through formal documents. These are separate constraints from the loss of direct supplementary investigation powers, but they add more handoffs to an agency still building its basic operating capacity.
Jurisdiction remains another unsettled area. Police, the Serious Crimes Investigation Agency and the Corruption Investigation Office for High-ranking Officials have differed over which institution should handle some matters, including cybercrime. In Ulsan, the Public Prosecution Office is using an internal review team rather than a separate unit for reviewing police decisions not to refer cases. The next measurable indicators will be whether warrant non-submission and supplementary-investigation requests remain elevated, whether requested case transfers begin to move, and whether staffing, premises and information-system access reach their planned operating levels.