The innovation benefits and limits of institutionalized science-industry collaboration: Evidence from China’s academician workstations
Using a staggered difference-in-differences approach on China's Academician Workstations, this study finds that while institutionalized science-industry collaboration initially boosts inventive output and collaboration, these innovation benefits diminish as the program expands and local density increases, suggesting that such organizational forms scale more easily than the underlying conditions required to sustain their effectiveness.
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
Governments around the world often try to speed up innovation by building bridges between scientists and businesses. The idea is simple: scientists have deep knowledge about how the world works, while companies have the factories, money, and market needs to turn that knowledge into new products. When these two groups work together, the theory goes, everyone wins. But there is a hidden complexity in how these partnerships are formed. Sometimes, a partnership is a genuine, deep collaboration where a scientist spends years helping a company solve a specific engineering problem. Other times, a partnership is a formal agreement, a handshake on paper that looks good for a report but involves very little actual work. This distinction matters because a government can easily create thousands of formal agreements, but it cannot easily force the deep, sustained effort required to actually invent something new. The question researchers have long asked is whether simply setting up these official bridges is enough to generate new ideas, or if the bridge needs to be walked across every day to be useful.
A team of researchers from Hefei University of Technology set out to answer this question by looking at a specific program in China called Academician Workstations. In China, an "academician" is a top-tier scientist, someone who has reached the highest level of recognition in their field, similar to a distinguished fellow in other countries. The government created these workstations to connect these elite scientists with private companies. The program was designed to help firms access the very best minds in the country. The researchers wanted to see if this arrangement actually led to more inventions and better technology, or if the program eventually became too big to be effective. They did not just ask the companies if they were happy; they looked at the actual results. They gathered a massive amount of data, tracking thousands of companies and their patent applications over many years, comparing those that set up a workstation with similar companies that did not.
The story of the program is one of rapid growth followed by a quiet slowdown. It began in the early 2000s as a small experiment. In those early years, the program was selective. When a company wanted to set up a workstation, it was usually because they had a specific, difficult problem they needed to solve, and they found a scientist who was genuinely interested in helping. During this early phase, the researchers found that the program worked very well. Companies that established a workstation started inventing more. They filed more patent applications, and they did so with a wider range of technologies. They also started working with more partners and involving more of their own employees in the creative process. It seemed that the access to a top scientist was giving these companies a real boost, helping them find new solutions and expand their capabilities.
However, the program grew very quickly. By around 2016, the number of new workstations being created began to surge. The government was encouraging more companies to join, and the program became a standard part of the business landscape. As the program expanded, the researchers noticed that the magic began to fade. When they looked at the companies that set up workstations after 2016, they found no significant increase in invention compared to similar companies that did not have a workstation. The surge in new partnerships did not lead to a surge in new ideas. In fact, the quality of the inventions, measured by how often other scientists cited them, did not improve at all. The researchers concluded that while the government could easily scale up the number of formal agreements, it could not scale up the deep, personal engagement and the perfect match between a scientist's skills and a company's needs that were required to make those agreements productive.
The study also looked at why this happened. It turned out that the success of the early partnerships depended heavily on the specific conditions of the match. In the beginning, the workstations that worked best were led by scientists who were still actively working and had only a few other commitments. They had the time and energy to focus on the company's problems. They also worked better when the company was large and had its own team of researchers ready to learn from the scientist and apply the new knowledge. But as the program grew, many scientists were asked to lead too many workstations at once, stretching their attention too thin. Many companies set up workstations just to look good or to get government support, without having the internal capacity to use the scientist's help. The researchers found that as the number of workstations in a city increased, the benefit of adding one more workstation actually decreased. It was as if the program had become too crowded, and the special value of having a direct line to a top scientist was diluted by the sheer number of connections.
This does not mean that the later workstations were useless or that the scientists stopped working. It simply means that the average result changed. The researchers suggest that the program became a case where the form of the partnership was easy to copy, but the substance was hard to maintain. A government can issue a certificate and set up a meeting, but it cannot force a scientist to spend hours troubleshooting a factory machine, nor can it force a company to hire the right people to learn from that scientist. The study shows that institutional support is powerful when it helps a few specific, well-matched pairs work together deeply. But when that same support is spread too widely, trying to connect everyone to everyone, the deep, productive work that leads to real innovation becomes harder to sustain. The lesson is that creating a platform for collaboration is only the first step; the real value comes from the quality of the work that happens on that platform, and that quality is much harder to guarantee when the program gets too big.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.