Association for Public Policy Analysis & Management

Technological innovation is an essential factor that shapes long-term environmental impacts of social-economic development. It is often an empirical question whether environmental regulation, while cleaning up pollution, promotes innovation from regulated firms and their long-term competitiveness. On the one hand, a higher compliance cost directly reduces the budget available for other business activities, including research and development. On the other hand, the induced innovation hypothesis, initially introduced by Hicks (1932) and restated by Porter (1991) and Acemoglu (2002; 2012), suggests that regulation can affect the direction of innovation, and if well designed, improve competitiveness.

Taking China’s Cleaner Production Standards as an example, we investigate the effect of environmental regulation on firm innovation. We show that for this type of well-specified performance standard, stricter mandates lead to higher innovation, and the effect does not rely on individual incidence of policy enforcement.

The concept of cleaner production was introduced to China in the 1990s and institutionalized by the Cleaner Production Promotion Law enacted in 2002. Compared to traditional pollution control practices that focus on end-of-pipe emissions, cleaner production regulates the source in the production phase by setting up performance standards. Following the law, performance standards were designed and implemented for 48 four-digit industrial sectors during 2003-2010. These standards have a specific focus on water pollution, because it was the major policy focus during that time in China. Follow-up cleaner production audit may be conducted at firms in industries with those standards, but the list of firms in an industry to be audited is publicized in advance.

In order to measure innovation activities, we compiled a dataset including all the individual patent applications at the State Intellectual Property Office (SIPO) of China. The SIPO dataset was merged with the Annual Survey of Industrial Firms (also called the Chinese Industrial Enterprise Database) from the National Bureau of Statistics, to aquire firm-level observations, on the condition that a patent applicant and a firm shared the same name. The merged firm-level dataset includes 826,000 firms from 542 four-digit sectors during 1998-2013.

Our identification strategy stems from the variation in three dimensions: existence of cleaner production standards between sectors, presence of a specific requirement for water recycling between regulated sectors, and temporal regulatory status within a sector. In other words, difference-in-difference-in-difference (DDD) estimators and a lower dimensional difference-in-difference (DID) estimator have been adopted.

The DID estimation shows that cleaner production standards induced firms in the regulated industries to file more water technology patents. The DDD estimation that accounts for the presence of water recycling rates reveals that, however, this induced innovation effect was driven by the mandate for water recycling. Further estimation based on a subsample of industries with water recycling mandates shows that a higher rate led to more patenting. These results are robust over Poisson and OLS estimations and various specifications. Cleaner production audits that were directly enforced at the firm level, on the other hand, did not have any additional effect on innovation.