China’s higher education system employs a unique postgraduate enrollment mechanism called “recommendation for postgraduate studies,” or “保研” (Bao-Yan) in Chinese. Due to its structure and timing, commitments between students and universities are often made and then broken, leading to a chaotic landscape of mutual defaults. This admission process has evolved into a disturbing game between students and admissions offices. Colloquially, students refer to the act of reneging on commitments as “鸽” (ge), meaning “to pigeon,” a playful nod to the widespread practice. Here’s a breakdown of the admission process, an analysis of its complications, and a proposed solution.
The Story
The entire process is lengthy, spanning from May to October.
Signaling Phase. From May through the summer, students participate in summer camps and fly-ins for on-site exams and interviews. This phase serves as both an assessment tool for universities and an opportunity for students to explore potential postgraduate programs.
Key Event: Recommendation Quota. In September, at the start of their final undergraduate year, students must secure a recommendation quota from their undergraduate institutions to participate in the later central admission process. Each institution has a relatively fixed annual quota, and students generally estimate their chances based on academic standing. However, the allocation of these quotas is inherently unpredictable. Sometimes, students who received strong offers during the signaling phase unexpectedly lose their quota, while others who did not anticipate receiving one might get lucky due to peers opting for industry positions or studying abroad.
Central Admission Phase. On September 28, the National Educational Bureau’s central admission system opens, allowing students (with a recommendation quota) and universities to officially “match.” A student can list up to three universities without priority ranking. Listing a university acts as a “request” to its admission office, which can then send back a time-restricted offer through the system—the duration set by the university. Although the admission system remain open for a month, an offer’s time window is typically brief—often lasting about an hour, or even as short as seven minutes. If a student accepts the offer within the allotted time, they are officially admitted. If not, the offer is lost, slot clears and the student may request another university.
Complications
The current admission process inevitably breeds chaos, as the incentives in such a mechanism lead to exploding offers and widespread defaults.
Here’s the logic: the Signaling Phase complements the Central Admission phase by enabling universities to assess candidates through summer camps and interviews. However, since whatever committment made in the signaling phase is non-binding, offers can be deferred by students or even, the university. When these non-obligatory offers from the previous signaling phase—coupled with inherent uncertainties—carry into the flawed central admission system, students tend to strategically delay their “request” action in the central admission system, opting to wait for offers from more preferred institutions. Similarly, universities, aiming to secure enough candidates, may extend more offers during the summer camps than they have quota for, anticipating that some students will decline. As a result, the process becomes a tangled mess with no equilibrium, resulting in often inefficient matches and regrets.
A Solution
Despite its flaws, the system has merits: most importantly it supports centralized admission while granting universities autonomy: centralized admission is well-suited to China’s context; whereas the signaling phase gives universities considerable freedom, enabling targeted interviews and assessments rather than relying solely on standardized exams. To retain a run in the centralized system as a final closing of the admission process, while allowing the signaling phase that breeds uncertainties and b-matching—implementing the Deferred Acceptance algorithm as the final matching mechanism could be a step in the right direction. It’s a proven solution in many similar matching markets, and seems to be the minimal change for decent improvement.