Inter-basin water transfer (IBWT) policies alter the spatial distribution of water endowments and trigger changes in environmental regulation policies, which may unintentionally impact the research and development (R&D) activities in IBWT water-receiving areas. However, the existing studies failed to examine the relationship between IBWT policies and corporate R&D activities, and lacked the exploration of the micro-mechanism of IBWT's unintended impact on corporate R&D activities. Through the water delivery of China's South-North Water Transfer Project as a quasi-natural experiment, this study adopts a difference-in-differences approach to scrutinise the unintended impact of IBWT policies on corporate R&D activities. The findings show that IBWT policies can make the water a ‘resource blessing’ by directly improving the water endowment in water-receiving areas, thereby promoting corporate R&D activities. In addition, IBWT policies can also indirectly encourage local governments in water-receiving areas to strengthen the intensity of environmental regulations, ultimately promoting companies to improve R&D activities. Finally, the impacts of IBWT policies on corporate R&D activities in water-receiving areas are heterogeneous. Overall, this study contributes to understanding the complicated relationship between IBWT policies and corporate R&D activities, and provides insights into how IBWT policies affect corporate R&D activities.

  • The causal relationship between IBWT and R&D activities is examined.

  • The micro-mechanisms that IBWT affects corporate R&D activities are explored.

  • IBWT can improve water endowment, thereby promoting corporate R&D activities.

  • IBWT strengthens environmental regulation, which improves corporate R&D activities.

  • The impacts of IBWT on corporate R&D activities are heterogeneous.

Graphical Abstract

Graphical Abstract
Graphical Abstract

Natural resource endowments, including water, are an essential factor in determining the competitive advantage of a country and its companies (Gaur et al., 2014). Water is one of the critical factors of production, and its endowment can directly or indirectly affect the economic activities of all sectors and regions of the world. Therefore, water scarcity may not only have an essential impact on humans, society, and ecosystems, but may also pose a threat to economic growth (Distefano & Kelly, 2017). In addition, over the past two decades, the business community has come to recognise that increasing water scarcity can also pose challenges to business operations and production, including disruptions to production caused by water scarcity and environmental regulations that restrict access to water for production or supply chains (Larson et al., 2012). Consequently, research and development (R & R&D) activities, an essential part of corporate operation and production, are also inevitably plagued by water scarcity (Gasbarro et al., 2016).

However, the impact of water scarcity on R&D activities is complicated and uncertain. Water scarcity may inhibit population and investment growth and cause population and capital to flow to water-rich areas (McDonald et al., 2011). This makes companies located in water-scarce areas lack labour and capital for R&D activities, thereby falling into low-tech activities (McGuirk et al., 2015). In addition, although environmental regulations conducive to the sustainable development of mineral resources contribute to corporate R&D activities, the water environmental regulations caused by water scarcity may instead inhibit the corporate R&D activities (Rennings & Rammer, 2011). Due to the persistent water environment regulations, the company that still conducts R&D activities becomes ‘late followers,’ which means that their new products were often introduced by other companies many years ago (Rennings & Rammer, 2011). Ironically, however, water scarcity may also have tempting side effects for the company: water scarcity creates much demand for new products, services, and other R&D activities that help address water scarcity (Hecker & Huber, 2017). For instance, Jiang et al. (2021) revealed that high water prices triggered by water scarcity might encourage innovative behaviour by private companies in water-saving or alternative water supply technologies.

Since the uneven distribution of water in time and space is common in many countries, many solutions have been proposed to address water scarcity. As water can be redistributed across national, regional, and local boundaries, the inter-basin water transfer (IBWT) policies were proposed as a solution (Matete & Hassan, 2006). IBWTs typically span two or more basins to transfer water from water-rich areas to water-scarce areas to meet the water needs of water-scarce areas. Up to more than 160 major IBWT projects have been built or under construction around the world by 2015 (Zhuang, 2016), such as California State Water Project in the United States, James Bay Project in Canada (Desbiens, 2004), Snowy Mountains Scheme in Australia (Griffin, 2003), National River-Linking Project in India (Verma et al., 2009), and South-North Water Transfer Project (SNWTP) in China (Sheng et al., 2018). These IBWT projects have altered the water endowments and environmental regulation policies of these countries or regions while meeting the needs of various departments for water.

As the impact of water scarcity on corporate R&D activities is uncertain, it is also an open question as to whether the IBWT policies, which aim to address water scarcity, can unintentionally transform corporate R&D activities. Will the changes in water endowment caused by IBWT policies unintentionally become a kind of ‘water blessing’ and promote corporate R&D activities? Or will IBWT policies unintentionally trigger a kind of ‘water curse’ and inhibit corporate R&D activities? How does IBWT unintentionally affect corporate R&D activities? Is the unintended impact of IBWT policies on corporate R&D activities heterogeneous across different companies? These complicated entanglement issues between IBWT policies and corporate R&D activities have received little attention. Most existing studies also fail to examine the micro-mechanism of IBWT's unintended impact on corporate R&D activities. To respond to this research gap, we scrutinise the unintended impact of IBWT policies on corporate R&D activities and clarify the micro-mechanism in this study. China's SNWTP provides an ideal case for studying the complicated relationship between IBWT policies and corporate R&D activities. The identification strategy is based on the quasi-natural experiment of SNWTP's water delivery; the difference-in-differences (DID) approach is adopted to examine the unintended impact of IBWT policies on corporate R&D activities, which has found new evidence for us. This evidence contradicts the previous findings that natural resource endowments inhibit corporate R&D activities. However, it helps theoretically and empirically to contribute to our understanding of the complicated relationship between IBWT policies and corporate R&D activities.

By analysing Chinese and English academic literature, media, and government documents, we put forward some political and economic arguments according to the empirical findings. First, IBWT policies can make the water a ‘resource blessing’ by directly improving the water endowment in water-receiving areas, thereby promoting corporate R&D activities. Second, IBWT policies can also indirectly encourage local governments in water-receiving areas to strengthen the intensity of environmental regulations, ultimately promoting companies to improve R&D activities. Finally, the unintended impacts of IBWT policies on corporate R&D activities in water-receiving areas are heterogeneous.

The rest of this study is organised as follows. Section 2 reviews SNWTP's background and proposes a theoretical framework in which SNWTP affects corporate R&D activities. Section 3 introduces the data and proposes the empirical strategy to explore the unintended impact of IBWT policies on corporate R&D activities. Section 4 presents the empirical results and checks their robustness. Sections 5 and 6 examine the mediation and heterogeneity, respectively. Finally, we discuss the results and conclude with a conclusion, in which we show how the analysis of IBWT policies and corporate R&D activities could help expand the current understanding of the complicated relationship and micro-mechanism between the two.

Study area

China has attempted to transfer water from the Yangtze River Basin to North and Northwest China by constructing SNWTP to address water scarcity in North China. SNWTP includes eastern, middle, and western routes, with a total investment of more than 240 billion yuan (People's Daily, 2014), involving the resettlement of more than 300,000 people (Rogers et al., 2016). This project has become one of the largest and most ambitious mega IBWT projects globally (Pohlner, 2016). The entire SNWTP connects the Yangtze, Huai, Yellow and Hai rivers from south to north and eventually forms a vast water network with ‘four horizontal and three vertical’ in China (MWR, 2002).

The eastern route uses the Yangtze River Water Transfer Project in Jiangsu Province, the Beijing-Hangzhou Grand Canal, and the existing rivers in Huai and Hai River Basins for water delivery (Sheng & Webber, 2017). The middle route draws water from the Danjiangkou Reservoir and delivers water to Beijing through a newly constructed canal (see Figure 1). The western route plans to transfer water from the upper tributary of the Yangtze River, but is still in the planning stage due to complex terrain and ecological issues (Ma et al., 2016). The first phases of the eastern and middle routes started construction in December 2002 and December 2003, respectively, and were completed in December 2013 and December 2014, respectively. The second phase of the two routes is still under preparation at present.
Fig. 1

The water-receiving and non-water-receiving areas of the first phase of China's SNWTP.

Fig. 1

The water-receiving and non-water-receiving areas of the first phase of China's SNWTP.

Close modal

SNWTP completely changed the water endowment of the water-receiving cities and brought abundant water to arid North China. China promulgated special regulations for SNWTP's water delivery and management in early 2014 (State Council, 2014a). The eastern and middle routes have provided 29.95 billion cubic meters of freshwater to arid North China from the beginning of the water delivery to 2019 (Economic Daily, 2019).

SNWTP also triggered environmental regulation policies in water-receiving and water-source cities. The east route passes through China's most developed Yangtze River Delta and the Bohai Rim region, and the booming manufacturing industry also makes the water pollution in the region particularly prominent. The Danjiangkou Reservoir, which is the water source of the middle route, is also affected by industrial and domestic sewage. China issued several pollution control plans for the eastern and middle routes (State Council, 2003, 2006) to address water pollution in SNWTP. However, water pollution has not been resolved. For instance, due to severe water pollution in Shandong, the eastern route, which was scheduled to transfer water in 2007, was delayed until 2013 (China Water, 2010). Although the water quality of the middle route is generally better than that of the eastern route, a large amount of sewage is still being discharged directly into the Danjiangkou Reservoir, which is the source of the middle route, before its completion (China Central Television, 2013). At the same time that SNWTP was completed, China issued a series of regulations to promote improved water quality in water-receiving and water-source cities (State Council, 2014a, 2014b). These environmental regulations have significantly improved water quality. The water quality of the Danjiangkou Reservoir is stable at Grade II; more than 80% of the water of the middle route has reached Class I, and the water quality of the eastern route is stable at Class III (Economic Daily, 2019).

Potential mechanisms for SNWTP to influence corporate R&D activities

As a typical human activity, SNWTP not only changes the original spatial distribution of China's water resources, but also unintentionally affects the corporate R&D activities in water-receiving cities.

First, SNWTP improved the water endowment of water-receiving cities, thus providing more resources for corporate R&D activities. SNWTP has enabled arid North China to obtain more water, providing a resource guarantee and support for corporate R&D activities. Moreover, the redistribution of water rents caused by SNWTP also contributes to accumulating human capital (Stijns, 2006), encouraging corporate R&D activities (Kurtz & Brooks, 2011). Concurrently, the improvement in water endowment can trigger population growth, which will cause capital to flow to areas with abundant water endowments (McDonald et al., 2011). The redistribution of water can promote the flow of production factors (Shao & Yang, 2014) so that corporate R&D activities in regions with improved water endowment can gain more resources. Therefore, we expect that SNWTP will significantly improve the water endowment of water-receiving cities, thereby promoting corporate R&D activities in these cities.

Second, SNWTP also changed the intensity of environmental regulation of the local government in water-receiving cities, which indirectly affected corporate R&D activities. To improve SNWTP's water quality, China has incorporated pollution control performance into local officials' performance evaluation and links their career promotions with water pollution control performance (GOSC, 2013; GOCPCCC, 2015). This series of measures makes it necessary for local officials in the water-receiving cities to improve SNWTP's water quality by increasing the intensity of environmental regulations to achieve their career promotions (Sheng et al., 2018). Porter & Van der Linde (1995) argue that environmental regulation can encourage R&D activities by improving production processes. Jaffe & Palmer (1997) further proposed the weak and strong Porter hypotheses: the weak one emphasises that environmental regulation can promote R&D activities, while the strong one emphasises that environmental regulation positively impacts productivity. Therefore, we expect that SNWTP can trigger an increase in the intensity of environmental regulation in water-receiving cities, thereby improving corporate R&D activities.

In summary, we expect SNWTP to unintentionally improve corporate R&D activities in water-receiving cities in two micro-mechanisms. On the one hand, SNWTP may improve the water endowment of water-receiving cities, thereby directly improving corporate R&D activities. On the other hand, SNWTP may indirectly encourage companies to improve R&D activities by enhancing the intensity of environmental regulations in water-receiving cities. Figure 2 illustrates the above two micro-mechanisms.
Fig. 2

Direct and indirect effects of the SNWTP on corporate R&D activities.

Fig. 2

Direct and indirect effects of the SNWTP on corporate R&D activities.

Close modal

Empirical strategy and data

In order to examine the unintended impact of SNWTP on corporate R&D activities, we use 338 companies in the water-receiving cities and 161 companies in the non-water-receiving cities to establish treatment and control groups, respectively. Then, we adopt a difference-in-differences (DID) strategy to estimate SNWTP's unintended impact by following Bertrand & Mullainathan (2003) and Bertrand et al. (2004). This strategy compares the changes in R&D expenditures of the treatment group relative to the control group before and after SNWTP's water delivery. The basic DID regression is as follows:
formula
(1)
where i represents the company, j represents the city, and t represents the year. The independent variable (transferjt) is a dummy variable that equals one if the j-th city receives the SNWTP water before the t-th year; otherwise, it equals zero. Xijt represents a vector of the control variables. εijt represents the error term; μi, γt, and vj represent the corporate fixed effect, year fixed effect, and city fixed effect, respectively; α represents the intercept, and β represents the SNWTP effect of being estimated.

According to Veugelers (1997) and Aw et al. (2011), we choose R&D expenditure (expen) as the dependent variable to measure corporate R&D activities. Since R&D expenditure runs through corporate R&D activities, it can accurately reflect the company's total input in R&D activities.

We select the following variables to control the company characteristics to ensure company heterogeneity does not affect the results. (i) Company size (size), measured by total assets. Large companies are more inclined to increase R&D expenditure than small companies (Shefer & Frenkel, 2005). (ii) Company age (age). Mature companies are more willing to make R&D expenditures than young companies since they can get higher profits (Coad et al., 2016). (iii) Since financial leverage can affect the company investments, including R&D investments (Aivazian et al., 2005), we choose liabilities to assets ratio (lar), return on assets (roa), and cash ratio (cash) to control the influences of financial leverage. (iv) Government subsidies (subsidy). Government subsidies can reduce the risk of company innovation, which will affect corporate R&D expenditure (Arqué-Castells & Mohnen, 2015; Caloffi et al., 2018). (v) The number of employees with higher education qualifications (edu). The reason for choosing this control variable is that companies with many employees with higher education qualifications tend to have a strong tendency to innovate (Sun et al., 2020). (vi) Stock yield (yield). Stock yield can reflect the market prospects of companies, and an excellent market prospect can help companies increase R&D expenditures (Hirshleifer et al., 2013). (vii) Equity concentration (concen) is measured by the proportion of shares held by the company's top ten shareholders. The main reason for choosing this control variable is that equity concentration can affect corporate R&D activities (Minetti et al., 2015). (viii) Company ownership (owner) equals one for state-owned enterprises and zero for non-state-owned enterprises. State-owned enterprises have more advantages than non-state-owned enterprises in acquiring talents, technology, and capital, which are essential for R&D activities (Cao et al., 2020).

Furthermore, we also select the following control variables to control the city characteristics to ensure the results are not affected by city heterogeneity: (i) urban economic level (econ), measured by per capita GDP; (ii) urbanisation ratio (urban), and (iii) fixed-asset investment (fix).

Finally, we logarithmically process some variables, including R&D expenditure (expen), company size (size), the number of employees with higher education qualifications (edu), urban economic level (econ), and fixed-asset investment (fix).

The eastern and middle routes were completed in December 2013 and 2014, respectively. Therefore, we set the initial years of water delivery to the water-receiving cities of the eastern and middle routes as 2014 and 2015, respectively. SNWTP involves seven provinces or municipalities, including Beijing, Tianjin, Hebei, Henan, Hubei, Shandong, and Jiangsu. Hubei Province is the water-source area for the middle route rather than the water-receiving area. There are 60 cities in the remaining six water-receiving provinces or municipalities, of which 40 cities are water-receiving, and 20 cities are non-water-receiving (see Figure 1).

All Chinese A-share listed companies from 2007 to 2020 in these 60 cities are selected as initial samples. We choose 2007 as the initial year of the sample mainly because many of the data of listed companies before 2007 are missing. Moreover, we also need an appropriate period to conduct the DID estimation before SNWTP's water delivery.

All data come from the China Stock Market and Accounting Research Database (CSMAR), China City Statistical Yearbook, and China City Construction Statistical Yearbook. We follow the steps below to process the initial data:

  • Step 1: We exclude all companies in the sample that have special treatments (such as ST and *ST) and delisting;

  • Step 2: We exclude all financial companies in the sample. The reason is that the accounting standards of the financial industry are quite different from those of other industries, and the relevant indicators are not comparable between the financial industry and the non-financial industry (Fields et al., 2004);

  • Step 3: We exclude all companies whose addresses change during the sample period to avoid self-selected samples;

  • Step 4: We exclude all companies with severe data loss.

Finally, we obtained 5,221 observations from 499 listed companies in 60 cities. Table 1 summarises all variables.

Table 1

Summary of the variables.

VariablesDescriptionsObs.MeanStd. Dev.MinMax
expen The natural logarithm of R&D expenditure 5,221 18.05 1.615 9.839 23.86 
size The natural logarithm of total assets 5,221 22.38 1.456 18.39 28.64 
age Company age 5,221 16.54 5.803 37 
lar Asset-to-liability ratio 5,221 0.414 0.210 0.0111 3.513 
roa Return on assets 5,221 0.0398 0.0760 −1.919 0.469 
subsidy The natural logarithm of government subsidies 5,221 16.46 1.843 6.908 23.55 
edu The natural logarithm of the number of employees with advanced degrees 5,221 6.488 1.490 1.792 12.55 
yield Stock yield 5,221 0.405 0.604 −6.840 5.696 
cash Cash ratio 5,221 0.178 0.147 0.000555 0.925 
concen Ownership concentration 5,221 0.582 0.151 0.104 0.986 
owner Company ownership 5,221 0.377 0.485 
econ The natural logarithm of per capita GDP 5,221 9.888 3.494 0.976 12.49 
urban Urbanization rate 5,221 0.294 0.193 0.0637 0.941 
fix The natural logarithm of fixed asset investment 5,221 14.47 1.505 8.761 16.52 
VariablesDescriptionsObs.MeanStd. Dev.MinMax
expen The natural logarithm of R&D expenditure 5,221 18.05 1.615 9.839 23.86 
size The natural logarithm of total assets 5,221 22.38 1.456 18.39 28.64 
age Company age 5,221 16.54 5.803 37 
lar Asset-to-liability ratio 5,221 0.414 0.210 0.0111 3.513 
roa Return on assets 5,221 0.0398 0.0760 −1.919 0.469 
subsidy The natural logarithm of government subsidies 5,221 16.46 1.843 6.908 23.55 
edu The natural logarithm of the number of employees with advanced degrees 5,221 6.488 1.490 1.792 12.55 
yield Stock yield 5,221 0.405 0.604 −6.840 5.696 
cash Cash ratio 5,221 0.178 0.147 0.000555 0.925 
concen Ownership concentration 5,221 0.582 0.151 0.104 0.986 
owner Company ownership 5,221 0.377 0.485 
econ The natural logarithm of per capita GDP 5,221 9.888 3.494 0.976 12.49 
urban Urbanization rate 5,221 0.294 0.193 0.0637 0.941 
fix The natural logarithm of fixed asset investment 5,221 14.47 1.505 8.761 16.52 

Overall treatment effect

According to Equation (1), we use a two-way fixed-effect model to examine the unintended impact of SNWTP on corporate R&D activities. The results are shown in Table 2. Column 1 controls the company-level variables, and column 2 controls the company-level and city-level variables.

Table 2

Impact of the SNWTP on R&D expenditure.

Variables(1)(2)
transfer 0.1031** (0.0493) 0.0986** (0.0444) 
size 0.5490*** (0.0610) 0.5461*** (0.0607) 
age 0.0885*** (0.0199) 0.0169 (0.0389) 
lar −0.6908*** (0.1396) −0.6922*** (0.1416) 
roa 0.1078 (0.2346) 0.0932 (0.2374) 
subsidy 0.0251* (0.0138) 0.0246* (0.0138) 
edu 0.2208*** (0.0521) 0.2257*** (0.0519) 
yield 0.0817** (0.0399) 0.0860** (0.0407) 
cash −0.1786 (0.1207) −0.1768 (0.1211) 
concen 0.4216* (0.2270) 0.4255* (0.2202) 
owner 0.1542 (0.1886) 0.1383 (0.1858) 
econ  −0.1034** (0.0487) 
urban  −0.1736 (0.1300) 
fix  0.0681* (0.0350) 
Constant 2.1974** (0.9855) 3.0952** (1.3893) 
Company fixed effect Yes Yes 
Year fixed effect Yes Yes 
City fixed effect Yes Yes 
Observations 5221 5221 
R2 0.5531 0.5542 
Variables(1)(2)
transfer 0.1031** (0.0493) 0.0986** (0.0444) 
size 0.5490*** (0.0610) 0.5461*** (0.0607) 
age 0.0885*** (0.0199) 0.0169 (0.0389) 
lar −0.6908*** (0.1396) −0.6922*** (0.1416) 
roa 0.1078 (0.2346) 0.0932 (0.2374) 
subsidy 0.0251* (0.0138) 0.0246* (0.0138) 
edu 0.2208*** (0.0521) 0.2257*** (0.0519) 
yield 0.0817** (0.0399) 0.0860** (0.0407) 
cash −0.1786 (0.1207) −0.1768 (0.1211) 
concen 0.4216* (0.2270) 0.4255* (0.2202) 
owner 0.1542 (0.1886) 0.1383 (0.1858) 
econ  −0.1034** (0.0487) 
urban  −0.1736 (0.1300) 
fix  0.0681* (0.0350) 
Constant 2.1974** (0.9855) 3.0952** (1.3893) 
Company fixed effect Yes Yes 
Year fixed effect Yes Yes 
City fixed effect Yes Yes 
Observations 5221 5221 
R2 0.5531 0.5542 

Note: standard error in parentheses; *, **, and *** represent significant at 10, 5 and 1% levels, respectively.

According to Table 2, the coefficients of transfer are both significantly positive, which demonstrates that SNWTP's water delivery promotes the increase of corporate R&D expenditure in the water-receiving cities. Moreover, the coefficient of transfer decreases significantly after adding the control variables at the city level. It suggests that the impact of city-level factors on corporate R&D activities cannot be ignored; otherwise, it will cause the SNWTP's effect to be overestimated.

Robustness checks

Although the above analysis provides evidence for the SNWTP's unintended impact on corporate R&D activities in water-receiving cities, a series of robustness checks can help ensure the reliability of these claims and alleviate estimation bias caused by endogeneity. Therefore, we will present robustness checks from the following three aspects, including verifying the parallel trend assumption, placebo test, and propensity matching score (PSM) estimation.

Although the DID approach can alleviate the endogeneity when evaluating the SNWTP's effect, the effectiveness of this approach is based on satisfying a fundamental identification assumption, that is, the parallel trend assumption (Li et al., 2012). This assumption requires that the treatment and control groups have the same trend in corporate R&D expenditure before SNWTP's water delivery. Therefore, we conduct a parallel trend test to verify whether this assumption is satisfied (see Appendix A). The results show no significant difference between the treatment and control groups before SNWTP's water delivery. In addition, the effect of SNWTP on corporate R&D activities in water-receiving cities has continuity. It means that SNWTP can sustainably increase corporate R&D expenditure, and the growth rate will increase over time.

In order to exclude the potential impact of random factors on the results, we conduct a bootstrapping placebo test by constructing a pseudo-treatment group by following Bradley et al. (2016) (see Appendix B). The results indicate that the estimations in Table 2 do not cause severe omitted variable bias; therefore, the DID estimations in basic regression are not affected by random factors.

Finally, although the DID model can alleviate endogeneity through the difference approach, it cannot address the sample selection bias. To alleviate the endogeneity caused by the sample selection bias, we adopt the PSM approach to identify the SNWTP's unintended effect (see Appendix C). The results demonstrate that the PSM approach has addressed sample selection bias. Moreover, the results from the PSM estimation are similar to those in Table 2, confirming that SNWTP's water delivery has promoted corporate R&D expenditures in water-receiving cities.

Mediation analysis

In the previous section, we expected that SNWTP could promote corporate R&D activities through two micro-mechanisms, including improving the water endowment and enhancing the intensity of environmental regulations in water-receiving cities. Therefore, it is necessary to examine these two micro-mechanisms through mediation analysis by following Baron & Kenny (1986) and Biesanz et al. (2010). Since the significance of the relationship between the independent and dependent variables is tested separately before and after controlling for the mediator, the mediation analysis can verify the theoretical validity of the pre-specified mediator (Rucker et al., 2011). The equations for mediation analysis are as follows:
formula
(2)
formula
(3)
formula
(4)
where mediation represents the mediator. In order to test the mediating effect of water endowment, we use per capita available water (available) to measure the water endowment by following OhIsson (2000). The mediation analysis results on water endowment are shown in Table 3, where columns 1–3 represent the estimation results of Equations (2)–(4), respectively.
Table 3

The mediation analysis results on water endowment.

(1)(2)(3)
Dep. Var.expenavailableexpen
transfer 0.0986** (0.0444) 5.6777 (5.7464) 0.0971** (0.0478) 
available   0.0003 (0.0010) 
size 0.5461*** (0.0607) 0.3497 (0.7784) 0.5461*** (0.0606) 
age 0.0169 (0.0389) 3.4271 (2.3904) 0.0160 (0.0400) 
lar −0.6922*** (0.1416) −2.9372 (3.2948) −0.6915*** (0.1396) 
roa 0.0932 (0.2374) 0.6476 (5.7762) 0.0930 (0.2367) 
subsidy 0.0246* (0.0138) 0.3404* (0.1874) 0.0245* (0.0136) 
edu 0.2257*** (0.0519) −0.4215 (0.9950) 0.2258*** (0.0519) 
yield 0.0860** (0.0407) 0.5070 (0.9054) 0.0859** (0.0407) 
cash −0.1768 (0.1211) −0.0181 (2.5569) −0.1768 (0.1212) 
concen 0.4255* (0.2202) 4.3985 (6.1187) 0.4243* (0.2206) 
owner 0.1383 (0.1858) −1.2345 (3.3552) 0.1387 (0.1855) 
econ −0.1034** (0.0487) 9.2614* (5.0346) −0.1058** (0.0500) 
urban −0.1736 (0.1300) −1.6834 (4.8017) −0.1732 (0.1297) 
fix 0.0681* (0.0350) 0.0729 (2.1330) 0.0681* (0.0350) 
Constant 3.0952** (1.3893) 1.9256 (78.1207) 3.0947** (1.3888) 
Bootstrap  Z=4.51 P=0.000 
Company fixed effect Yes Yes Yes 
Year fixed effect Yes Yes Yes 
City fixed effect Yes Yes Yes 
Observations 5,221 5,221 5,221 
R2 0.5542 0.3660 0.5541 
(1)(2)(3)
Dep. Var.expenavailableexpen
transfer 0.0986** (0.0444) 5.6777 (5.7464) 0.0971** (0.0478) 
available   0.0003 (0.0010) 
size 0.5461*** (0.0607) 0.3497 (0.7784) 0.5461*** (0.0606) 
age 0.0169 (0.0389) 3.4271 (2.3904) 0.0160 (0.0400) 
lar −0.6922*** (0.1416) −2.9372 (3.2948) −0.6915*** (0.1396) 
roa 0.0932 (0.2374) 0.6476 (5.7762) 0.0930 (0.2367) 
subsidy 0.0246* (0.0138) 0.3404* (0.1874) 0.0245* (0.0136) 
edu 0.2257*** (0.0519) −0.4215 (0.9950) 0.2258*** (0.0519) 
yield 0.0860** (0.0407) 0.5070 (0.9054) 0.0859** (0.0407) 
cash −0.1768 (0.1211) −0.0181 (2.5569) −0.1768 (0.1212) 
concen 0.4255* (0.2202) 4.3985 (6.1187) 0.4243* (0.2206) 
owner 0.1383 (0.1858) −1.2345 (3.3552) 0.1387 (0.1855) 
econ −0.1034** (0.0487) 9.2614* (5.0346) −0.1058** (0.0500) 
urban −0.1736 (0.1300) −1.6834 (4.8017) −0.1732 (0.1297) 
fix 0.0681* (0.0350) 0.0729 (2.1330) 0.0681* (0.0350) 
Constant 3.0952** (1.3893) 1.9256 (78.1207) 3.0947** (1.3888) 
Bootstrap  Z=4.51 P=0.000 
Company fixed effect Yes Yes Yes 
Year fixed effect Yes Yes Yes 
City fixed effect Yes Yes Yes 
Observations 5,221 5,221 5,221 
R2 0.5542 0.3660 0.5541 

Note: standard error in parentheses; *, **, and *** represent significant at 10, 5 and 1% levels, respectively.

According to Table 3, the coefficient of transfer in column 1 is significantly positive. In contrast, the coefficient of transfer in column 2 and the coefficient of available in column 3 are both positive and insignificant. Furthermore, the results of the bootstrap test reject the null hypothesis, which indicates that SNWTP significantly improves the water endowment in water-receiving cities after water delivery, thereby triggering improvements in R&D expenditures. According to Baron & Kenny (1986), water endowment is a mediator between SNWTP's water delivery and R&D expenditure. Therefore, the above results suggest that SNWTP's water delivery improves the water endowment in water-receiving cities, thereby increasing corporate R&D expenditures.

In order to test the mediating effect of the intensity of environmental regulations, we adopt the urban sewage treatment rate (sewage) to measure the intensity of environmental regulations by following Song et al. (2022). As the urban sewage treatment rate can reflect the effect of environmental regulation at the output level, this indicator is often widely used as a proxy variable for the intensity of environmental regulation in the water sector (Peng, 2020; Song et al., 2020). A high urban sewage treatment rate suggests a high intensity of environmental regulations, and vice versa. The mediation analysis results on the intensity of environmental regulations are shown in Table 4.

Table 4

The mediation analysis results on the intensity of environmental regulations.

Dep. Var.(1)(2)(3)
expensewageexpen
transfer 0.0986** (0.0444) 0.0343** (0.0145) 0.0808* (0.0454) 
sewage   0.5187* (0.2905) 
size 0.5461*** (0.0607) 0.0025 (0.0021) 0.5448*** (0.0611) 
age 0.0169 (0.0389) 0.0356*** (0.0082) −0.0016 (0.0366) 
lar −0.6922*** (0.1416) −0.0124** (0.0058) −0.6858*** (0.1418) 
roa 0.0932 (0.2374) −0.0219** (0.0091) 0.1045 (0.2371) 
subsidy 0.0246* (0.0138) −0.0007 (0.0005) 0.0249* (0.0139) 
edu 0.2257*** (0.0519) −0.0006 (0.0015) 0.2260*** (0.0524) 
yield 0.0860** (0.0407) 0.0010 (0.0015) 0.0855** (0.0408) 
cash −0.1768 (0.1211) −0.0163** (0.0063) −0.1684 (0.1214) 
concen 0.4255* (0.2202) −0.0125 (0.0117) 0.4320* (0.2218) 
owner 0.1383 (0.1858) −0.0120* (0.0060) 0.1445 (0.1857) 
econ −0.1034** (0.0487) 0.0351** (0.0143) −0.1216** (0.0457) 
urban −0.1736 (0.1300) −0.0267** (0.0103) −0.1597 (0.1275) 
fix 0.0681* (0.0350) 0.0005 (0.0058) 0.0679* (0.0352) 
Constant 3.0952** (1.3893) 0.0713 (0.2602) 3.0583** (1.3912) 
Bootstrap  Z=4.23 P=0.000 
Company fixed effect Yes Yes Yes 
Year fixed effect Yes Yes Yes 
City fixed effect Yes Yes Yes 
Observations 5,221 5,221 5,221 
R2 0.5542 0.5790 0.5545 
Dep. Var.(1)(2)(3)
expensewageexpen
transfer 0.0986** (0.0444) 0.0343** (0.0145) 0.0808* (0.0454) 
sewage   0.5187* (0.2905) 
size 0.5461*** (0.0607) 0.0025 (0.0021) 0.5448*** (0.0611) 
age 0.0169 (0.0389) 0.0356*** (0.0082) −0.0016 (0.0366) 
lar −0.6922*** (0.1416) −0.0124** (0.0058) −0.6858*** (0.1418) 
roa 0.0932 (0.2374) −0.0219** (0.0091) 0.1045 (0.2371) 
subsidy 0.0246* (0.0138) −0.0007 (0.0005) 0.0249* (0.0139) 
edu 0.2257*** (0.0519) −0.0006 (0.0015) 0.2260*** (0.0524) 
yield 0.0860** (0.0407) 0.0010 (0.0015) 0.0855** (0.0408) 
cash −0.1768 (0.1211) −0.0163** (0.0063) −0.1684 (0.1214) 
concen 0.4255* (0.2202) −0.0125 (0.0117) 0.4320* (0.2218) 
owner 0.1383 (0.1858) −0.0120* (0.0060) 0.1445 (0.1857) 
econ −0.1034** (0.0487) 0.0351** (0.0143) −0.1216** (0.0457) 
urban −0.1736 (0.1300) −0.0267** (0.0103) −0.1597 (0.1275) 
fix 0.0681* (0.0350) 0.0005 (0.0058) 0.0679* (0.0352) 
Constant 3.0952** (1.3893) 0.0713 (0.2602) 3.0583** (1.3912) 
Bootstrap  Z=4.23 P=0.000 
Company fixed effect Yes Yes Yes 
Year fixed effect Yes Yes Yes 
City fixed effect Yes Yes Yes 
Observations 5,221 5,221 5,221 
R2 0.5542 0.5790 0.5545 

Note: standard error in parentheses; *, **, and *** represent significant at 10, 5 and 1% levels, respectively.

According to Table 4, the significantly positive coefficient of transfer in column 2 demonstrates that SNWTP's water delivery has prompted the local governments to strengthen the intensity of environmental regulations in water-receiving cities. The coefficient of sewage in column 3 is also significantly positive, which demonstrates that the increase in the intensity of environmental regulation stimulates an increase in corporate R&D expenditures. Similarly, the intensity of environmental regulation is also a mediator between SNWTP's water delivery and R&D expenditure, according to Baron & Kenny (1986). In addition, the results of the bootstrap test also reject the null hypothesis. Therefore, the above results suggest that SNWTP's water delivery encourages local governments to strengthen the intensity of environmental regulations, leading to an increase in corporate R&D expenditures.

In summary, the micro-mechanism of SNWTP's unintended impact on corporate R&D activities is reflected in two aspects. First, SNWTP increases the water endowment in water-receiving cities, so that companies can have more water resources for R&D activities. Second, SNWTP promotes local governments to strengthen the intensity of environmental regulations in water-receiving cities, ultimately promoting companies to improve R&D activities.

Heterogeneity analysis

Since the R&D activities of manufacturing companies are significantly different from those of non-manufacturing companies (Yew Kee et al., 2005), it is necessary to examine the difference in the SNWTP's unintended impact on the two types of companies. We divide the sample companies into manufacturing and non-manufacturing companies, and the results are shown in columns 1–2 of Table 5.

Table 5

The results of heterogeneity analysis.

Variables(1)(2)(3)(4)(5)(6)
Manufacturing companiesNon-manufacturing companiesYoung companiesMature companiesCompanies with low government subsidiesCompanies with high government subsidies
transfer 0.1032* (0.0520) −0.1409* (0.0755) 0.0313 (0.0732) 0.1301* (0.0730) 0.1967*** (0.0643) −0.0079 (0.0577) 
size 0.5726*** (0.0702) 0.3863*** (0.0719) 0.6685*** (0.0835) 0.5055*** (0.0727) 0.4655*** (0.0828) 0.5492*** (0.0848) 
age −0.0006 (0.0348) 0.0917 (0.1156) −0.0913 (0.0745) 0.0870* (0.0457) 0.0603 (0.0502) −0.0106 (0.0562) 
lar −0.2875* (0.1447) −1.4300** (0.5535) −0.3616* (0.1802) −0.9278*** (0.2975) −0.4890*** (0.1599) −0.9040*** (0.2354) 
roa 0.9215** (0.3703) −1.0550 (0.6473) 0.8888*** (0.2985) −0.5520 (0.4201) 0.1232 (0.4010) −0.0986 (0.3457) 
subsidy 0.0259* (0.0135) 0.0049 (0.0144) 0.0365** (0.0169) 0.0208 (0.0175) −0.0011 (0.0132) 0.0492 (0.0326) 
edu 0.2174*** (0.0517) 0.3814*** (0.0648) 0.2300*** (0.0510) 0.2059*** (0.0727) 0.1979*** (0.0591) 0.2493*** (0.0700) 
yield −0.0614 (0.1618) −0.3899 (0.2720) −0.2733 (0.2252) −0.2649* (0.1459) 0.0973 (0.0631) 0.0714 (0.0508) 
cash 0.0710 (0.0442) 0.1180 (0.0767) 0.0370 (0.0622) 0.1644*** (0.0553) −0.2513 (0.1774) −0.1259 (0.1585) 
concen 0.3427 (0.2510) 0.1247 (0.4812) 0.2543 (0.2773) 0.1406 (0.3470) 0.3430 (0.2281) 0.2924 (0.2028) 
owner 0.1872 (0.2132) 0.0782 (0.0683) 0.1486 (0.1235) 0.0755 (0.3118) −0.0799 (0.3206) 0.1591* (0.0881) 
econ −0.1458*** (0.0403) 0.1077 (0.1721) −0.2291** (0.0879) −0.0605 (0.0552) −0.0511 (0.0572) −0.0955 (0.0817) 
urban −0.0185 (0.1332) −0.4738** (0.2301) −0.6165 (0.3807) −0.1258 (0.1369) 0.1396 (0.1759) −0.2279** (0.1112) 
fix 0.0291 (0.0403) 0.3293*** (0.1113) 0.0587 (0.0434) 0.0636 (0.0496) 0.0694 (0.0502) 0.0481 (0.0572) 
Constant 3.4469** (1.4903) 0.2179 (3.6848) 2.6916** (1.2543) 2.5737 (2.4283) 3.9539* (2.0662) 3.6383 (2.3172) 
Company-fixed effect Yes Yes Yes Yes Yes Yes 
Year-fixed effect Yes Yes Yes Yes Yes Yes 
City-fixed effect Yes Yes Yes Yes Yes Yes 
Observations 3,938 1,283 2,268 2,953 2,610 2,611 
R2 0.6088 0.4583 0.5078 0.4631 0.4917 0.5125 
Variables(1)(2)(3)(4)(5)(6)
Manufacturing companiesNon-manufacturing companiesYoung companiesMature companiesCompanies with low government subsidiesCompanies with high government subsidies
transfer 0.1032* (0.0520) −0.1409* (0.0755) 0.0313 (0.0732) 0.1301* (0.0730) 0.1967*** (0.0643) −0.0079 (0.0577) 
size 0.5726*** (0.0702) 0.3863*** (0.0719) 0.6685*** (0.0835) 0.5055*** (0.0727) 0.4655*** (0.0828) 0.5492*** (0.0848) 
age −0.0006 (0.0348) 0.0917 (0.1156) −0.0913 (0.0745) 0.0870* (0.0457) 0.0603 (0.0502) −0.0106 (0.0562) 
lar −0.2875* (0.1447) −1.4300** (0.5535) −0.3616* (0.1802) −0.9278*** (0.2975) −0.4890*** (0.1599) −0.9040*** (0.2354) 
roa 0.9215** (0.3703) −1.0550 (0.6473) 0.8888*** (0.2985) −0.5520 (0.4201) 0.1232 (0.4010) −0.0986 (0.3457) 
subsidy 0.0259* (0.0135) 0.0049 (0.0144) 0.0365** (0.0169) 0.0208 (0.0175) −0.0011 (0.0132) 0.0492 (0.0326) 
edu 0.2174*** (0.0517) 0.3814*** (0.0648) 0.2300*** (0.0510) 0.2059*** (0.0727) 0.1979*** (0.0591) 0.2493*** (0.0700) 
yield −0.0614 (0.1618) −0.3899 (0.2720) −0.2733 (0.2252) −0.2649* (0.1459) 0.0973 (0.0631) 0.0714 (0.0508) 
cash 0.0710 (0.0442) 0.1180 (0.0767) 0.0370 (0.0622) 0.1644*** (0.0553) −0.2513 (0.1774) −0.1259 (0.1585) 
concen 0.3427 (0.2510) 0.1247 (0.4812) 0.2543 (0.2773) 0.1406 (0.3470) 0.3430 (0.2281) 0.2924 (0.2028) 
owner 0.1872 (0.2132) 0.0782 (0.0683) 0.1486 (0.1235) 0.0755 (0.3118) −0.0799 (0.3206) 0.1591* (0.0881) 
econ −0.1458*** (0.0403) 0.1077 (0.1721) −0.2291** (0.0879) −0.0605 (0.0552) −0.0511 (0.0572) −0.0955 (0.0817) 
urban −0.0185 (0.1332) −0.4738** (0.2301) −0.6165 (0.3807) −0.1258 (0.1369) 0.1396 (0.1759) −0.2279** (0.1112) 
fix 0.0291 (0.0403) 0.3293*** (0.1113) 0.0587 (0.0434) 0.0636 (0.0496) 0.0694 (0.0502) 0.0481 (0.0572) 
Constant 3.4469** (1.4903) 0.2179 (3.6848) 2.6916** (1.2543) 2.5737 (2.4283) 3.9539* (2.0662) 3.6383 (2.3172) 
Company-fixed effect Yes Yes Yes Yes Yes Yes 
Year-fixed effect Yes Yes Yes Yes Yes Yes 
City-fixed effect Yes Yes Yes Yes Yes Yes 
Observations 3,938 1,283 2,268 2,953 2,610 2,611 
R2 0.6088 0.4583 0.5078 0.4631 0.4917 0.5125 

Note: standard error in parentheses; *, **, and *** represent significant at 10, 5 and 1% levels, respectively.

In addition, companies in different stages of their life cycles often exhibit different characteristics in their R&D activities (Huergo & Jaumandreu, 2004). Therefore, it is necessary to examine the unintended impact of SNWTP on the R&D activities of young and mature companies. We use the median age of the sample companies to divide them into two types, young companies and mature companies, and the results are shown in columns 3–4 of Table 5.

Government subsidies often have a positive incentive effect on corporate R&D activities, mainly because they can not only provide funds for corporate R&D activities (Guellec & Van Pottelsberghe De La Potterie, 2003; Caloffi et al., 2018), but also reduce the risk of corporate R&D activities (Arqué-Castells & Mohnen, 2015), thereby increasing the profitability of R&D activities. However, companies can also reduce their original R&D expenditures and switch to investment and other projects after receiving government subsidies, thereby triggering crowding-out effects (Lichtenberg, 1984; Kauko, 1996). It suggests that the high government subsidies obtained by companies may lead to a substantial crowding-out effect (Görg & Strobl, 2007). We use the median of the government subsidies of the sample companies to divide them into companies with low government subsidies and companies with high government subsidies. The results are shown in columns 5–6 of Table 5.

According to columns 1–2 of Table 5, the coefficient of transfer in column 1 is significantly positive, while in column 2 it is significantly negative. It demonstrates that SNWTP's water delivery has a significant positive effect on the R&D expenditures of manufacturing companies in water-receiving cities; however, it has the opposite effect on non-manufacturing companies.

Furthermore, the coefficient of transfer in column 3 is significantly positive, while in column 4 it is insignificant. It demonstrates that SNWTP's water delivery has a more substantial promotion effect on the R&D activities of mature companies in water-receiving cities; however, this promotion effect is insignificant in young companies.

Finally, the coefficient of transfer in column 5 is significantly positive, while in column 6 it is insignificant. It demonstrates that SNWTP's water delivery promotes the R&D activities of companies with low government subsidies in water-receiving cities, but has no significant impact on companies with high government subsidies.

Few studies have examined the relationship between IBWT policies and corporate R&D activities. More importantly, as far as we know, there are no studies to examine the micro-mechanism of IBWT policies influencing corporate R&D activities. By focusing on the unintended impact of China's SNWTP on corporate R&D activities in water-receiving cities, this study contributes to our understanding of this complicated relationship and provides insights into the micro-mechanisms. Moreover, this study reveals new findings, contradicting previous claims that natural resource endowments inhibit corporate R&D activities. Therefore, it contributes a heuristic inquiry, speaking to the ongoing debate on the relationship between resource endowments and R&D activities.

We combine the panel data sets of 499 listed companies in 60 cities in China from 2007 to 2017 to examine the unintended impact of IBWT policies on corporate R&D activities. We determined the causality by adopting a DID design and using the geographic locations of water-receiving and non-water-receiving cities and the date of SNWTP's water delivery. Moreover, we also use the PSM approach to ensure the comparability of the treatment and control groups. The results are reliable for using different model settings. Other robustness checks, including verification of the parallel trend assumption and placebo test, can confirm our results.

The findings demonstrate that IBWT policies can directly improve the water endowment in water-receiving areas, thereby promoting corporate R&D activities. Notably, water resources are different from other natural resources since they may not cause a resource curse. On the contrary, the improvement of water endowment caused by IBWT policies may become a water blessing and promote corporate R&D activities. IBWT policies have changed the regional water endowment and provided more water support for corporate R&D activities in water-receiving areas. Additionally, the change in water endowments can also trigger the redistribution of labour and capital used for corporate R&D activities (Shao & Yang, 2014).

Furthermore, IBWT policies can also indirectly encourage local governments in water-receiving areas to strengthen the intensity of environmental regulations, ultimately promoting companies to improve R&D activities. To ensure SNWTP's water quality, China has adopted the principles of ‘first controlling pollution, then transferring water; first environmental protection, then using water’ as two of the three SNWTP water management objectives. To this end, SNWTP implements promotion tournaments for improving water quality, which links the promotion of local officials to SNWTP pollution control, thereby prompting local governments to compete to strengthen the intensity of environmental regulations (Sheng et al., 2018). The increase in the intensity of environmental regulations can lead to increased pollution costs in production (Štreimikiene & Esekina, 2008), which encourages companies to improve production technology by increasing R&D expenditures, thereby increasing productivity (Huang et al., 2019).

The findings also show that the unintended impacts of IBWT policies on corporate R&D activities in water-receiving areas are heterogeneous. First, IBWT policies can increase the R&D activities of manufacturing companies while inhibiting the R&D activities of non-manufacturing companies. Compared with the R&D-intensive strategies that manufacturing companies often use, non-manufacturing companies may not rely entirely on R&D expenditures but often use advertising investment to increase company visibility to create value (Yew Kee et al., 2005). Therefore, IBWT policies can encourage manufacturing companies to increase R&D expenditures to effectively use water and avoid the adverse effects of environmental regulations. Conversely, non-manufacturing companies may use brand effects instead of improving R&D activities to create reputation premiums and maximise profits (Erickson & Jacobson, 1992). Furthermore, the R&D activities of non-manufacturing companies are less dependent on water endowments than manufacturing companies (Jin et al., 2019). Therefore, the improvement in water endowment induced by the IBWT policies does not encourage the R&D activities of non-manufacturing companies. However, the increased intensity of environmental regulations in the water sector triggered by the IBWT policies increases the operating costs of non-manufacturing companies, which may crowd out their R&D expenditures and reduce their innovation success (Rennings & Rammer, 2011).

Second, IBWT policies can encourage the improvement of R&D activities in mature companies, but there is no significant impact on the R&D activities of young companies. Both young and mature companies will improve their company performance through R&D activities. However, young companies need to take high risks when engaging in R&D activities, while mature companies can predict their future returns and effectively avoid risks (Coad et al., 2016). Moreover, the R&D activities of young companies rely on the increase in market demand, which can stimulate these companies to increase R&D expenditures; in contrast, mature companies increase R&D expenditures in pursuit of market concentration and product diversification (García-Quevedo et al., 2014). Although IBWT policies provide a large amount of water to companies in water-receiving areas, it does not directly drive market demand. Moreover, strict environmental regulations have also put pressure on young companies, putting them in a dilemma of development and transformation. However, mature companies can make full use of external water for R&D activities to create an innovation compensation effect (Herriott et al., 1985), which helps to achieve the goals of increasing market share and improving company performance.

Third, IBWT policies help to improve the R&D activities of companies with low government subsidies, but have no significant impact on the R&D activities of companies with high government subsidies. Since companies with high government subsidies receive a large number of government subsidies, the crowding-out effect of their subsidies is significant (Görg & Strobl, 2007). Even though IBWT policies can help to improve corporate R&D activities, this effect is insignificant relative to the crowding-out effect of subsidies. Therefore, it suggests that moderate government subsidies are conducive to incentivising corporate R&D activities, while excessive government subsidies can trigger a crowding-out effect and negatively affect corporate R&D activities.

These findings have implications for understanding corporate R&D activities in the context of IBWT policies. We have confirmed that IBWT policies promote corporate R&D activities by improving water endowment and triggering improvements in the intensity of environmental regulations. However, the unintended impacts of IBWT policies on the R&D activities of different companies are heterogeneous. Therefore, it is necessary to consider its impact on different companies when designing IBWT policies. The previous findings indicate that IBWT policies contribute to the R&D activities of manufacturing companies while suppressing the R&D activities of non-manufacturing companies. The second phase of China's SNWTP is still in the design stage, which means that the urban industrial structure needs to be considered when selecting future water-receiving cities. Moreover, since IBWT policies are more conducive to promoting the R&D activities of companies with low government subsidies, it is possible to consider reducing government subsidies to encourage corporate R&D activities in water-receiving cities.

Finally, we note that some previous studies argued that water scarcity might encourage corporate R&D behaviour (Debaere & Kapral, 2021), suggesting that the improved water endowment triggered by IBWT policies may dissipate this incentive effect. Therefore, we invite additional case studies from other spatiotemporal contexts to verify whether IBWT policies can positively influence corporate R&D behaviour. This can extend our understanding of the complex relationship between IBWT policies and corporate R&D behaviour, and increase the cross-scale understanding of the underlying mechanisms involved. In addition, another limitation is the focus on the impact of the IBWT policies on the corporate R&D activities in water-receiving areas but not on water-source areas. IBWT policies may reduce the water endowment of water-source areas and, therefore, may inhibit the corporate R&D activities. However, due to the abundance of water resources, this impact in water-source areas remains uncertain and unknown. Therefore, more studies are necessary to respond to this gap.

This study was supported by the National Natural Science Foundation of China under Grants 72074119 and 71774088.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.

Aivazian
V. A.
,
Ge
Y.
&
Qiu
J.
(
2005
).
The impact of leverage on firm investment: Canadian evidence
.
Journal of Corporate Finance
11
(
1
),
277
291
.
Arqué-Castells
P.
&
Mohnen
P.
(
2015
).
Sunk costs, extensive R&D subsidies and permanent inducement effects
.
The Journal of Industrial Economics
63
(
3
),
458
494
.
Aw
B. Y.
,
Roberts
M. J.
&
Xu
D. Y.
(
2011
).
R&D investment, exporting, and productivity dynamics
.
American Economic Review
101
(
4
),
1312
1344
.
Bertrand
M.
&
Mullainathan
S.
(
2003
).
Enjoying the quiet life? corporate governance and managerial preferences
.
Journal of Political Economy
111
(
5
),
1043
1075
.
Bertrand
M.
,
Duflo
E.
&
Mullainathan
S.
(
2004
).
How much should we trust differences-in-differences estimates?
.
The Quarterly Journal of Economics
119
(
1
),
249
275
.
Biesanz
J. C.
,
Falk
C. F.
&
Savalei
V.
(
2010
).
Assessing mediational models: testing and interval estimation for indirect effects
.
Multivariate Behavioral Research
45
(
4
),
661
701
.
Bradley
D.
,
Kim
I.
&
Tian
X.
(
2017
).
Do Unions Affect Innovation?
Management Science
63
(
7
),
2251
2271
.
Cao
J.
,
Cumming
D.
&
Zhou
S.
(
2020
).
State ownership and corporate innovative efficiency
.
Emerging Markets Review
44
,
100699
.
China Central Television
(
2013
).
Danjiangkou: The Water Source That is Being Polluted (丹江口:正被污染的水源地)
.
China Central Television
.
China Water
(
2010
).
The Eastern Route of the South-North Water Transfer Project has Been Postponed due to Pollution
.
China Water
.
Coad
A.
,
Segarra
A.
&
Teruel
M.
(
2016
).
Innovation and firm growth: does firm age play a role?
Research Policy
45
(
2
),
387
400
.
Distefano
T.
&
Kelly
S.
(
2017
).
Are we in deep water? water scarcity and its limits to economic growth
.
Ecological Economics
142
,
130
147
.
Economic Daily
(
2019
).
South-North Water Transfer Project has benefited more than 120 million people in the past 5 years (南水北调工程通水5周年超1.2亿人直接受益). Economic Daily. Beijing, China
.
Fields
L. P.
,
Fraser
D. R.
&
Wilkins
M. S.
(
2004
).
An investigation of the pricing of audit services for financial institutions
.
Journal of Accounting and Public Policy
23
(
1
),
53
77
.
García-Quevedo
J.
,
Pellegrino
G.
&
Vivarelli
M.
(
2014
).
R&D drivers and age: are young firms different?
Research Policy
43
(
9
),
1544
1556
.
Gasbarro
F.
,
Rizzi
F.
&
Frey
M.
(
2016
).
Adaptation measures of energy and utility companies to cope with water scarcity induced by climate change
.
Business Strategy and the Environment
25
(
1
),
54
72
.
Gaur
A. S.
,
Kumar
V.
&
Singh
D.
(
2014
).
Institutions, resources, and internationalization of emerging economy firms
.
Journal of World Business
49
(
1
),
12
20
.
GOCPCCC
(
2015
).
Circular of the General Office of the CPC Central Committee and the General Office of the State Council on Issuing the Measures for Accountability of Ecological and Environmental Damage of Party and Government Leading Cadres (中共中央办公厅、国务院办公厅关于印发《党政领导干部生态环境损害责任追究办法(试行)》的通知) (中办发(2015)45号)
.
General Office of the CPC Central Committee
,
Beijing, China
.
Görg
H.
&
Strobl
E.
(
2007
).
The effect of R&D subsidies on private R&D
.
Economica
74
(
294
),
215
234
.
GOSC
(
2013
).
Circular of the General Office of the State Council on Issuing the Measures for Assessment of Implementation of the Strictest Management System for Water Resources (国务院办公厅关于印发实行最严格水资源管理制度考核办法的通知) (Official Communication 国办发(2013) 2号)
.
General Office of the State Council
,
Beijing, China
.
Griffin
G.
(
2003
).
Selling the snowy: the snowy mountains scheme and national mythmaking
.
Journal of Australian Studies
27
(
79
),
39
49
.
Guellec
D.
&
Van Pottelsberghe De La Potterie
B.
(
2003
).
The impact of public R&D expenditure on business R&D*
.
Economics of Innovation and New Technology
12
(
3
),
225
243
.
Hecker
A.
,
Huber
F.
, (
2017
).
The future of the management of innovation: trends and challenges
. In
Handbook Of The Management Of Creativity And Innovation: Theory And Practice
.
Tang
L. M.
&
Werner
C. H.
(eds.).
World Scientific
,
Hackensack, NJ
, pp.
331
346
.
Herriott
S. R.
,
Levinthal
D.
&
March
J. G.
(
1985
).
Learning from experience in organizations
.
The American Economic Review
75
(
2
),
298
302
.
Hirshleifer
D.
,
Hsu
P. -H.
&
Li
D.
(
2013
).
Innovative efficiency and stock returns
.
Journal of Financial Economics
107
(
3
),
632
654
.
Huang
J.
,
Cai
X.
,
Huang
S.
,
Tian
S.
&
Lei
H.
(
2019
).
Technological factors and total factor productivity in China: evidence based on a panel threshold model
.
China Economic Review
54
,
271
285
.
Huergo
E.
&
Jaumandreu
J.
(
2004
).
How does probability of innovation change with firm Age?
Small Business Economics
22
(
3
),
193
207
.
Jaffe
A. B.
&
Palmer
K.
(
1997
).
Environmental regulation and innovation: a panel data study
.
The Review of Economics and Statistics
79
(
4
),
610
619
.
Jiang
X.
,
Eaton
S.
&
Kostka
G.
(
2021
).
Not at the table but stuck paying the bill: perceptions of injustice in China's Xin'anjiang eco-compensation program
.
Journal of Environmental Policy & Planning
1
17
.
Jin
W.
,
Zhang
H. -q.
,
Liu
S. -s.
&
Zhang
H. -b.
(
2019
).
Technological innovation, environmental regulation, and green total factor efficiency of industrial water resources
.
Journal of Cleaner Production
211
,
61
69
.
Larson
W. M.
,
Freedman
P. L.
,
Passinsky
V.
,
Grubb
E.
&
Adriaens
P.
(
2012
).
Mitigating corporate water risk: financial market tools and supply management strategies
.
Water Alternatives
5
(
3
),
582
602
.
Lichtenberg
F. R.
(
1984
).
The relationship between federal contract R&D and company R&D
.
The American Economic Review
74
(
2
),
73
78
.
Ma
Y. J.
,
Li
X. Y.
,
Wilson
M.
,
Wu
X. C.
,
Smith
A.
&
Wu
J. G.
(
2016
).
Water loss by evaporation from China's south-north water transfer project
.
Ecological Engineering
95
,
206
215
.
McDonald
R. I.
,
Green
P.
,
Balk
D.
,
Fekete
B. M.
,
Revenga
C.
,
Todd
M.
&
Montgomery
M.
(
2011
).
Urban growth, climate change, and freshwater availability
.
Proceedings of the National Academy of Sciences
108
(
15
),
6312
6317
.
Minetti
R.
,
Murro
P.
&
Paiella
M.
(
2015
).
Ownership structure, governance, and innovation
.
European Economic Review
80
,
165
193
.
MWR
(
2002
).
South-North Water Transfer Project Masterplan (Summary)
.
Ministry of Water Resources
,
Beijing, China
.
OhIsson
L.
(
2000
).
Water conflicts and social resource scarcity
.
Physics and Chemistry of the Earth, Part B: Hydrology, Oceans and Atmosphere
25
(
3
),
213
220
.
People's Daily
(
2014
).
The Middle Route of South-North Water Transfer Project is Ready for Tansferring Water (穿黄隧洞充水试验成功 南水北调中线全线具备通水条件)
.
People's Daily
,
Beijing, China
.
Porter
M. E.
&
van der Linde
C.
(
1995
).
Toward a new conception of the environment-competitiveness relationship
.
Journal of Economic Perspectives
9
,
97
118
.
Rogers
S.
,
Barnett
J.
,
Webber
M.
,
Finlayson
B.
&
Wang
M.
(
2016
).
Governmentality and the conduct of water: China's south-north water transfer project
.
Transactions of the Institute of British Geographers
41
(
4
),
429
441
.
Rucker
D. D.
,
Preacher
K. J.
,
Tormala
Z. L.
&
Petty
R. E.
(
2011
).
Mediation analysis in social psychology: current practices and new recommendations
.
Social and Personality Psychology Compass
5
(
6
),
359
371
.
Shefer
D.
&
Frenkel
A.
(
2005
).
R&D, firm size and innovation: an empirical analysis
.
Technovation
25
(
1
),
25
32
.
Sheng
J.
,
Webber
M.
&
Han
X.
(
2018
).
Governmentality within China's south-north water transfer project: tournaments, markets and water pollution
.
Journal of Environmental Policy & Planning
20
(
4
),
533
549
.
State Council
(
2003
).
Circular of Sate Council on Approving the Opinions of Construction Committee Office of South to North Water Transfer Project on the Implementation of Pollution Control Plan for the SNWTP-ER
.
State Council
,
Beijing
.
State Council
(
2006
).
Reply of the State Council on Plan of Water Pollution Control and Water and Soil Conservation in Danjiangkou Reservoir Area and Upstream Region (国务院关于丹江口库区及上游水污染防治和水土保持规划的批复) (Official Communication 国函[2006]10号)
.
State Council
,
Beijing
.
State Council
(
2014a
).
Regulations on Water Supply and Water use for the SNWTP.
State Council
,
Beijing
.
State Council
(
2014b
).
Regulation on Urban Drainage and Sewage Treatment (城镇排水与污水处理条例)
.
State Council
,
Beijing
.
Stijns
J. -P.
(
2006
).
Natural resource abundance and human capital accumulation
.
World Development
34
(
6
),
1060
1083
.
Štreimikiene
D.
&
Esekina
B.
(
2008
).
EU pollution reduction strategies and their impact on atmospheric emissions in Lithuania
.
Ukio Technologinis ir Ekonominis Vystymas
14
(
2
),
162
170
.
Sun
X.
,
Li
H.
&
Ghosal
V.
(
2020
).
Firm-level human capital and innovation: evidence from China
.
China Economic Review
59
,
101388
.
Verma
S.
,
Kampman
D. A.
,
van der Zaag
P.
&
Hoekstra
A. Y.
(
2009
).
Going against the flow: a critical analysis of inter-state virtual water trade in the context of India's National River Linking Program
.
Physics Chemistry of the Earth, Parts A/B/C
34
(
4–5
),
261
269
.
Yew Kee
H.
,
Keh
H. T.
&
Jin Mei
O.
(
2005
).
The effects of R&D and advertising on firm value: an examination of manufacturing and nonmanufacturing firms
.
IEEE Transactions on Engineering Management
52
(
1
),
3
14
.
Zhuang
W.
(
2016
).
Eco-environmental impact of inter-basin water transfer projects: a review
.
Environmental Science and Pollution Research
23
(
13
),
12867
12879
.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY 4.0), which permits copying, adaptation and redistribution, provided the original work is properly cited (http://creativecommons.org/licenses/by/4.0/).

Supplementary data