What do consumers expect for government subsidies on low-carbon products in China?

What do consumers expect for government subsidies on low-carbon products in China? Abstract Based on the 873 questionnaires collected from six cities in four provinces in China, we made a quantitative analysis of different types of consumers’ expectations for government low-carbon subsidies by using the SPSS. The results indicate that: (1) Significant differences exist in the ‘expectations for government subsidies on low-carbon products’ from different types of consumers; there are significant differences among consumers with different monthly income, educational attainment and age on the ‘expectations for government subsidies on low-carbon products’, but the difference is not significant for the consumers in different regions or gender. (2) Monthly income and educational attainment exert significant influences on consumer ‘expectations for government subsidies on low-carbon products’, and the influence of monthly income is the biggest. Finally, we put forward policy recommendations accordingly. INTRODUCTION With the global climate change and increasing pressure for GHG emissions reduction, the development of low-carbon economy has become the top priority for all countries throughout the world. It is inevitable for developing low-carbon economy to achieve low-carbon consumption. Currently, the economic development and people’s income in China are still under development. Therefore, government subsidies are an important means to promote low-carbon consumption [1, 2]. Government subsidies for low-carbon products can stimulate consumers’ willingness to buy low-carbon products, to a certain extent [3–5]. So, what is the general expectation of government subsidies on low-carbon products? Which properties of consumers have significant impacts on their expectation? Are there any differences among different types of consumers in their expectations? The research literature related to government subsidies on low-carbon products and development of low-carbon economy mainly includes the following four aspects: (1) research on the role of government subsidies to the development of low-carbon economy [6]; (2) research on the mechanism of government subsidies [7, 8]; (3) research on the directions and objects of government subsidies [9, 10]; and (4) research on the forms of government subsidies [11, 12]. Most of the researchers believe that government subsidies play an important role in the carbon reduction process [13], and will be effective in promoting low-carbon consumption and energy-saving and emissions reduction [14, 15]. Sun [16] in his research point out that a country can achieve the goal of developing a low-carbon economy and the promotion of energy conservation and emissions reduction through adopting government subsidies on low-carbon products. Shen and He [17] consider that subsidies on low-carbon resources and products are important fiscal instrument to reducing carbon emissions, as well as increasing the consumption of low-carbon products. Bollino [18] explores the existence of consumer’s Willingness to Pay (WTP) in order to use renewable energy in the electricity production, and finds that the aggregate WTP for RES in Italy is not enough to attain the Government goal in 2010, which implies that the development of low-carbon industry, such as renewable energy, needs government subsidies. Moreover, low-carbon energy technologies are of increasing importance to India for reducing emissions and diversifying its energy supply mix [19]. Jin and Zheng [20] present a game analysis of both supply and demand sides in the green agricultural product market, and suggest that government compensation policy towards low-carbon agricultural consumers can help promote the development of low-carbon products. Roe et al. [21] and Zarnikau [22] suggest that public policy can have a positive and significant impact on consumers’ voluntarily payment of the additional costs for low-carbon products. Through the study of low-carbon product subsidies of the United States, Galinato and Yoder [23] find that the price subsidies for low-carbon products can be more efficient on carbon emissions than the tax on high-carbon products, which has similar findings with Bajona and Kelly’s [24] research on Chinese market in 1997. If China removes low-carbon subsidies, it will have negative impact on world carbon emissions [25]. Lapan and Moschini [26] construct a general equilibrium economy model to study on both the positive and normative implications of alternative policy instruments, including the subsidies and mandates. Their research shows that government subsidies for clean energy can implement carbon reduction goals effectively. Due to the imperfection of market mechanism and the lack of legal binding, the enterprises misrepresent the cost for more subsidies [27]. To solve this problem, He and Wang [28] established a cost reporting model, which can ensure the authenticity of the cost reports. Using GCAM, an integrated assessment model, Shukla and Chaturvedi [19] analyze a targets approach for pushing solar, wind and nuclear technologies in the Indian electricity generation sector from 2005 to 2095 in different scenarios, and find that the subsidy is still necessary in the short run when the carbon price is low. In terms of government subsidy directions and objects, Yang [29] summarizes the effective policies to promote low-carbon economy formulated by different countries, and then analyzes the government, enterprises and consumers in low-carbon economy through Game theory. He puts forward that government subsidies should focus on the consumers. Similarly, Sun’s [16] research proposes that government subsidies should pay more attention to the consumption link rather than the circulation chain. Shen and He [17] find that government subsidies for enterprises mainly focus on the industry of renewable power generation in the developed countries, such as Germany and the United States. But the government subsidies for consumers primarily involve the auto, real estate and household appliance industries, etc. Other scholars argue that government subsidy focus should depend on different situations. Brzeskot and Haupt [30] detect that an industry-friendly government levies an energy tax to supplement a lax product standard, but shies away from subsidies to firms; by contrast, a consumer-friendly government depends heavily on a strict product standard and additionally implements a moderate subsidy to firms, but avoids energy taxes. Xiong [31] points out that the government should take full advantage of the alternative role of financial subsidies to encourage city residents to buy energy-efficient appliances. As for government subsidy forms, Bansal and Gangopadhyay [32] think that differentiated subsidies can be more efficient than the unified subsidy on the improvement of environmental quality and social welfare, which is supported by Zhou and Zheng [33]. Li and Zhao [34] construct a Nash game and Stackelberg game model of manufacturer and retailer, and analyze the impact of R&D cost allocation coefficient and subsidies on the low-carbon investment of supply chain, and finally find out the strategies of low-carbon R&D cooperation and government subsidies under different game scenarios. Xie et al. [35] suggest that government should encourage various types of capital to the rural areas to promote the development of agricultural carbon sinks, as well as subsidize domestic agricultural carbon sinks by imposing carbon tariffs on imported agricultural products. Fan et al. [36] analyze the optimal strategy for the government to supervise low-carbon subsidies. Li et al. [37] find that consumers’ willingness to purchase low-carbon products differed significantly with respect to values, age, income and education. Literature review shows that: So far, there are little empirical study on the consumer differences of government subsidies from the perspective of demography. Meanwhile, the persuasion theories point out that the characteristics of the recipient, namely, the differences of consumers’ educational background, income, attitude, gender and age, may affect the interpretation and reaction of recipient to the information [38]. Therefore, a meticulous research on different consumer groups should be conducted to develop targeted measures as well as achieve maximum persuasion effect. In this study, based on the demographic view of different consumers, we make a quantitative analysis of the expectation differences of the low-carbon subsidies for different types of consumers by adopting the methodologies of one-way ANOVA, Bonferroni test, Dunnett’s T3 and regression model, in an attempt to reveal the potential trends of the subsidy expectations of different consumers. The reminder of this article is organized as follows: Research Hypothesis describes the empirical methodologies and data sources. Testing Results and Analyses analyzes the test results. Regression Analysis of Consumers’ Expectation conducts the regression analysis. Conclusions and Recommendations is the conclusions and recommendations of this study. RESEARCH HYPOTHESIS In this paper, we are analyzing what consumers expect for low-carbon product subsidies from the government. In this case, the definition of consumers’ expectations for government subsidies refers to ‘public demand for subsidies for low-carbon products’. In light of the technical and socio-economic conditions in present China, low-carbon products generally have higher costs than conventional high-carbon products. Therefore, in the beginning stage, government subsidies are necessary to stimulate consumers to accept and procure low-carbon products. Therefore, we proposed the following hypothesis for this research: different types of consumers have different expectations for government subsidies which could encourage them to buy low-carbon products; to be more specific, people differing in monthly income and educational levels, ages, regions and genders may have different subsidy expectations. METHODOLOGIES AND SAMPLE DATA Methodologies Firstly, we acquired the first-hand data through questionnaire survey, in an attempt to find out the ‘expectations for government subsidies on low-carbon products’ of different consumers in different regions, ages, educational backgrounds, incomes and genders. Individuals are randomly sampled from selected cities in China, and then the respondents completed the questionnaires at specific locations. Based on the questionnaire data of 873 consumers in six cities of China, we utilized the SPSS 20.0 software to conduct a one-way ANOVA, Bonferroni test and Dunnett’s T3 test to quantitatively analyze the differences in consumer expectations for government subsidies on low-carbon products. Then, by adopting the linear regression model, we further verified the influence of region and demographic variables on their expectations. Sample data The sample data come from the field questionnaire survey in six cities of four provinces in China. The geographical distribution of the samples covers the southern, central and northern part of China, with broad representation in terms of economic development and population distribution. In accordance with the random principle, we have taken the geographical distribution of various types of consumer groups into account and given out questionnaires in the densely populated areas, such as supermarkets, office buildings or residential districts. A total of 950 questionnaires were handed out and 876 received, providing a response rate of 92.21%, among which 873 are valid, accounting for 99.66%. As shown in Table 1, in the demographic variable of gender, females outnumber males with 58.08–41.92%. The samples cover six regions, whose frequencies are basically balanced: ranging from 13.06% (Wuhan) to 19.36% (Shenzhen). In the demographic variable of age, the youths and the middle aged are the main consumers of low-carbon products when those aged between 20 and 49 constitute 81.30%. In the dimension of income, the percentage of those earning <USD 761 per month is 86.40% and those earning 151–455 USD per month is 45.70%. Therefore, the sample is fairly typical and representative for Chinese consumers from a general perspective. Table 1. The distribution of samples. Features  Frequency  Ratio (%)  1. Gender  Male  366  41.92  Female  507  58.08  2. Age  Under 19  30  3.44  20–29  341  39.06  30–39  198  22.68  40–49  173  19.82  Over 50  131  15.01  3. Monthly income  Under 150 USD  167  19.13  151–455 USD  399  45.70  456–760 USD  189  21.65  761–1060 USD  73  8.36  Above 1061 USD  45  5.15  4. Educational attainment  Secondary school  122  13.97  Vocational school/high school  217  24.86  Professional training/college  146  16.72  University  289  33.10  Graduate school  99  11.34  5. City (region)  Xiangyang  147  16.84  Xuchang  154  17.64  Shenzhen  169  19.36  Daqing  142  16.27  Gucheng  147  16.84  Wuhan  114  13.06  Features  Frequency  Ratio (%)  1. Gender  Male  366  41.92  Female  507  58.08  2. Age  Under 19  30  3.44  20–29  341  39.06  30–39  198  22.68  40–49  173  19.82  Over 50  131  15.01  3. Monthly income  Under 150 USD  167  19.13  151–455 USD  399  45.70  456–760 USD  189  21.65  761–1060 USD  73  8.36  Above 1061 USD  45  5.15  4. Educational attainment  Secondary school  122  13.97  Vocational school/high school  217  24.86  Professional training/college  146  16.72  University  289  33.10  Graduate school  99  11.34  5. City (region)  Xiangyang  147  16.84  Xuchang  154  17.64  Shenzhen  169  19.36  Daqing  142  16.27  Gucheng  147  16.84  Wuhan  114  13.06  Source: Derived from the sample data by the authors using SPSS 20.0. Table 1. The distribution of samples. Features  Frequency  Ratio (%)  1. Gender  Male  366  41.92  Female  507  58.08  2. Age  Under 19  30  3.44  20–29  341  39.06  30–39  198  22.68  40–49  173  19.82  Over 50  131  15.01  3. Monthly income  Under 150 USD  167  19.13  151–455 USD  399  45.70  456–760 USD  189  21.65  761–1060 USD  73  8.36  Above 1061 USD  45  5.15  4. Educational attainment  Secondary school  122  13.97  Vocational school/high school  217  24.86  Professional training/college  146  16.72  University  289  33.10  Graduate school  99  11.34  5. City (region)  Xiangyang  147  16.84  Xuchang  154  17.64  Shenzhen  169  19.36  Daqing  142  16.27  Gucheng  147  16.84  Wuhan  114  13.06  Features  Frequency  Ratio (%)  1. Gender  Male  366  41.92  Female  507  58.08  2. Age  Under 19  30  3.44  20–29  341  39.06  30–39  198  22.68  40–49  173  19.82  Over 50  131  15.01  3. Monthly income  Under 150 USD  167  19.13  151–455 USD  399  45.70  456–760 USD  189  21.65  761–1060 USD  73  8.36  Above 1061 USD  45  5.15  4. Educational attainment  Secondary school  122  13.97  Vocational school/high school  217  24.86  Professional training/college  146  16.72  University  289  33.10  Graduate school  99  11.34  5. City (region)  Xiangyang  147  16.84  Xuchang  154  17.64  Shenzhen  169  19.36  Daqing  142  16.27  Gucheng  147  16.84  Wuhan  114  13.06  Source: Derived from the sample data by the authors using SPSS 20.0. TESTING RESULTS AND ANALYSES Firstly, we conducted Homogeneity tests for sample variances based on the monthly income level, educational background, age, region and gender data for these six cities in China. If the variances are homogeneous, we could use the Bonferroni test to analyze the samples; otherwise, we should use the Dunnett’s T3 test. Analysis by monthly income In order to facilitate the analysis, we regard the consumers with a monthly income of 1061 USD or above as high-income consumers, those ranging from 456 to 1060 USD as middle-income consumers, and those with <455 USD as low-income consumers. Analyzing monthly income differences reveals the following findings (Tables 2 and 3): The low-income consumers have the highest expectations for government subsidies on low-carbon products. The differences between high-income and low-income consumers are significant, so are the differences between middle-income and low-income consumers; but there is little difference between high-income and middle-income consumers. Table 2 displays that the difference of low-carbon subsidy expectations between high-income consumers (3.00) and low-income consumers (3.68) is large (22.67%), and that between middle-income consumers (3.24) and low-income consumers (3.68) is considerable (13.58%). However, the difference between high-income consumers (3.00) and middle-income consumers (3.24) is small (8% only) (Table 2). Currently, China’s average monthly income of urban residents is 420 USD in China, and the majority of them are still at low levels of income. Therefore, it still needs some more years to realize low-carbon consumption in China. What is more, because of the large income gap of Chinese residents, the high-income group should take the lead in low-carbon consumption, and the low-income group should get more low-carbon subsidies from the government. Consumers’ expectations for government subsidies on low-carbon products have a negative correlation with the monthly income. The average expectation of those consumers with a monthly income of 1061 USD or above is minimum (3.00), but that of the under 999 group is maximum (3.68), as shown in Table 2. Actually, government subsidies on low-carbon products can offset the additional costs incurred by the producer to produce low-carbon products. For the low-income consumers, the lower their income, the higher their expectations. In the case of high-income consumers, their expectations for subsidies are low because their living pressure is relatively low. The middle-income group has the moderate expectations. There are significant differences among different income levels on their expectations for government subsidies on low-carbon products. In order to verify the significance of the differences, firstly we made the Homogeneity tests for variances. From Table 2, we can observe that the Sig. value of the Homogeneity tests for variances is 0.065 (>0.05), which means that the sample has the homogeneity of variance. In addition, according to the result of variance analysis among groups in Table 2, the significance value is 0.000 (<0.05), showing that the differences among consumers with different incomes are significant. Table 2. Test results by monthly income and the ranking of expectation mean values. Monthly income    N  Mean value  Homogeneity tests for variances  Variance analysis  Significance  Significance (among groups)  Low-income consumers  Under 999 USD  167  3.68  0.065  0.000  151–455 USD  399  3.55  Middle-income consumers  5000–1060 USD  73  3.26  456–760 USD  189  3.24  High-income consumers  Above 1061 USD  45  3.00  Monthly income    N  Mean value  Homogeneity tests for variances  Variance analysis  Significance  Significance (among groups)  Low-income consumers  Under 999 USD  167  3.68  0.065  0.000  151–455 USD  399  3.55  Middle-income consumers  5000–1060 USD  73  3.26  456–760 USD  189  3.24  High-income consumers  Above 1061 USD  45  3.00  Note: The significance level of Homogeneity tests and Variance analysis is 0.05. Source: Derived from the sample data by the authors using SPSS 20.0. Table 2. Test results by monthly income and the ranking of expectation mean values. Monthly income    N  Mean value  Homogeneity tests for variances  Variance analysis  Significance  Significance (among groups)  Low-income consumers  Under 999 USD  167  3.68  0.065  0.000  151–455 USD  399  3.55  Middle-income consumers  5000–1060 USD  73  3.26  456–760 USD  189  3.24  High-income consumers  Above 1061 USD  45  3.00  Monthly income    N  Mean value  Homogeneity tests for variances  Variance analysis  Significance  Significance (among groups)  Low-income consumers  Under 999 USD  167  3.68  0.065  0.000  151–455 USD  399  3.55  Middle-income consumers  5000–1060 USD  73  3.26  456–760 USD  189  3.24  High-income consumers  Above 1061 USD  45  3.00  Note: The significance level of Homogeneity tests and Variance analysis is 0.05. Source: Derived from the sample data by the authors using SPSS 20.0. Table 3. Bonferroni test for different monthly income. (I) Monthly income  (J) Monthly income  Mean difference  Std. error  Significance  95% Confidence interval  (I–J)  Lower bound  Upper bound  Under 999 USD  151–455 USD  0.128  0.095  1.000  −0.14  0.40  456–760 USD  0.433*  0.110  0.001  0.12  0.74  5000–1060 USD  0.416*  0.145  0.041  0.01  0.82  Above 1061 USD  0.677*  0.173  0.001  0.19  1.16  151–455 USD  Under 999 USD  −0.128  0.095  1.000  −0.40  0.14  456–760 USD  0.305*  0.091  0.008  0.05  0.56  5000–1060 USD  0.289  0.131  0.283  −0.08  0.66  Above 1061 USD  0.549*  0.162  0.008  0.09  1.01  456–760 USD  Under 999 USD  −0.433*  0.110  0.001  −0.74  −0.12  151–455 USD  −0.305*  0.091  0.008  −0.56  −0.05  5000–1060 USD  −0.017  0.142  1.000  −0.42  0.38  Above 1061 USD  0.243  0.171  1.000  −0.24  0.73  5000–1060 USD  Under 999 USD  −0.416*  0.145  0.041  −0.82  −0.01  151–455 USD  −0.289  0.131  0.283  −0.66  0.08  456–760 USD  0.017  0.142  1.000  −0.38  0.42  Above 1061 USD  0.260  0.196  1.000  −0.29  0.81  Above 1061 USD  Under 999 USD  −0.677*  0.173  0.001  −1.16  −0.19  151–455 USD  −0.549*  0.162  0.008  −1.01  −0.09  456–760 USD  −0.243  0.171  1.000  −0.73  0.24  5000–1060 USD  −0.260  0.196  1.000  −0.81  0.29  (I) Monthly income  (J) Monthly income  Mean difference  Std. error  Significance  95% Confidence interval  (I–J)  Lower bound  Upper bound  Under 999 USD  151–455 USD  0.128  0.095  1.000  −0.14  0.40  456–760 USD  0.433*  0.110  0.001  0.12  0.74  5000–1060 USD  0.416*  0.145  0.041  0.01  0.82  Above 1061 USD  0.677*  0.173  0.001  0.19  1.16  151–455 USD  Under 999 USD  −0.128  0.095  1.000  −0.40  0.14  456–760 USD  0.305*  0.091  0.008  0.05  0.56  5000–1060 USD  0.289  0.131  0.283  −0.08  0.66  Above 1061 USD  0.549*  0.162  0.008  0.09  1.01  456–760 USD  Under 999 USD  −0.433*  0.110  0.001  −0.74  −0.12  151–455 USD  −0.305*  0.091  0.008  −0.56  −0.05  5000–1060 USD  −0.017  0.142  1.000  −0.42  0.38  Above 1061 USD  0.243  0.171  1.000  −0.24  0.73  5000–1060 USD  Under 999 USD  −0.416*  0.145  0.041  −0.82  −0.01  151–455 USD  −0.289  0.131  0.283  −0.66  0.08  456–760 USD  0.017  0.142  1.000  −0.38  0.42  Above 1061 USD  0.260  0.196  1.000  −0.29  0.81  Above 1061 USD  Under 999 USD  −0.677*  0.173  0.001  −1.16  −0.19  151–455 USD  −0.549*  0.162  0.008  −1.01  −0.09  456–760 USD  −0.243  0.171  1.000  −0.73  0.24  5000–1060 USD  −0.260  0.196  1.000  −0.81  0.29  *The mean difference is significant at the 0.05 level. Source: Derived from the sample data by the authors using SPSS 20.0. Table 3. Bonferroni test for different monthly income. (I) Monthly income  (J) Monthly income  Mean difference  Std. error  Significance  95% Confidence interval  (I–J)  Lower bound  Upper bound  Under 999 USD  151–455 USD  0.128  0.095  1.000  −0.14  0.40  456–760 USD  0.433*  0.110  0.001  0.12  0.74  5000–1060 USD  0.416*  0.145  0.041  0.01  0.82  Above 1061 USD  0.677*  0.173  0.001  0.19  1.16  151–455 USD  Under 999 USD  −0.128  0.095  1.000  −0.40  0.14  456–760 USD  0.305*  0.091  0.008  0.05  0.56  5000–1060 USD  0.289  0.131  0.283  −0.08  0.66  Above 1061 USD  0.549*  0.162  0.008  0.09  1.01  456–760 USD  Under 999 USD  −0.433*  0.110  0.001  −0.74  −0.12  151–455 USD  −0.305*  0.091  0.008  −0.56  −0.05  5000–1060 USD  −0.017  0.142  1.000  −0.42  0.38  Above 1061 USD  0.243  0.171  1.000  −0.24  0.73  5000–1060 USD  Under 999 USD  −0.416*  0.145  0.041  −0.82  −0.01  151–455 USD  −0.289  0.131  0.283  −0.66  0.08  456–760 USD  0.017  0.142  1.000  −0.38  0.42  Above 1061 USD  0.260  0.196  1.000  −0.29  0.81  Above 1061 USD  Under 999 USD  −0.677*  0.173  0.001  −1.16  −0.19  151–455 USD  −0.549*  0.162  0.008  −1.01  −0.09  456–760 USD  −0.243  0.171  1.000  −0.73  0.24  5000–1060 USD  −0.260  0.196  1.000  −0.81  0.29  (I) Monthly income  (J) Monthly income  Mean difference  Std. error  Significance  95% Confidence interval  (I–J)  Lower bound  Upper bound  Under 999 USD  151–455 USD  0.128  0.095  1.000  −0.14  0.40  456–760 USD  0.433*  0.110  0.001  0.12  0.74  5000–1060 USD  0.416*  0.145  0.041  0.01  0.82  Above 1061 USD  0.677*  0.173  0.001  0.19  1.16  151–455 USD  Under 999 USD  −0.128  0.095  1.000  −0.40  0.14  456–760 USD  0.305*  0.091  0.008  0.05  0.56  5000–1060 USD  0.289  0.131  0.283  −0.08  0.66  Above 1061 USD  0.549*  0.162  0.008  0.09  1.01  456–760 USD  Under 999 USD  −0.433*  0.110  0.001  −0.74  −0.12  151–455 USD  −0.305*  0.091  0.008  −0.56  −0.05  5000–1060 USD  −0.017  0.142  1.000  −0.42  0.38  Above 1061 USD  0.243  0.171  1.000  −0.24  0.73  5000–1060 USD  Under 999 USD  −0.416*  0.145  0.041  −0.82  −0.01  151–455 USD  −0.289  0.131  0.283  −0.66  0.08  456–760 USD  0.017  0.142  1.000  −0.38  0.42  Above 1061 USD  0.260  0.196  1.000  −0.29  0.81  Above 1061 USD  Under 999 USD  −0.677*  0.173  0.001  −1.16  −0.19  151–455 USD  −0.549*  0.162  0.008  −1.01  −0.09  456–760 USD  −0.243  0.171  1.000  −0.73  0.24  5000–1060 USD  −0.260  0.196  1.000  −0.81  0.29  *The mean difference is significant at the 0.05 level. Source: Derived from the sample data by the authors using SPSS 20.0. In order to figure out the differences of subsidy expectations among different types of consumers, we tested the expectation differences in this dimension by adopting the Bonferroni Test. Those consumers with a monthly income below 455 USD are significantly different from the 456–760 USD and 1061 or more USD groups, and have certain differences with other groups, but do not reach the significant level (Table 3). Analysis by educational attainment The expectations of consumers with different educational attainment are analyzed with one-way ANOVA as well, with the following findings. The mean values of the consumer expectations for government subsidies on low-carbon products by educational attainment are as follows in a descending order: vocational school/high school (3.54), university (3.53), professional training/college (3.52), secondary school (3.50) and graduate school (2.88). From Table 4, we can easily observe that the maximum mean value (vocational school/high school) is 22.92% higher than the minimum mean value (graduate school), which indicates a significant difference in the expectation by educational level. Moreover, the lower their educational attainment, the stronger their expectations for subsidies. This may be caused by the fact that the government subsidies can effectively reduce the price of low-carbon products, and consumers of different educational attainment have different income and life stress. Consumers with less education possess lower income and face greater pressure, so they are more willing to get subsidies on low-carbon products from the government. Table 4. Test results by educational background and the ranking of expectation mean values. Educational background  N  Mean value  Homogeneity tests for variances Significance  Variance analysis Significance (among groups)  Vocational school/high school  217  3.54  0.065  0.000  University  289  3.53  Professional training/college  146  3.52  Secondary school  122  3.50  Graduate school  99  2.88  Educational background  N  Mean value  Homogeneity tests for variances Significance  Variance analysis Significance (among groups)  Vocational school/high school  217  3.54  0.065  0.000  University  289  3.53  Professional training/college  146  3.52  Secondary school  122  3.50  Graduate school  99  2.88  Note: The significance level of Homogeneity tests and Variance analysis is 0.05. Source: Derived from the sample data by the authors using SPSS 20.0. Table 4. Test results by educational background and the ranking of expectation mean values. Educational background  N  Mean value  Homogeneity tests for variances Significance  Variance analysis Significance (among groups)  Vocational school/high school  217  3.54  0.065  0.000  University  289  3.53  Professional training/college  146  3.52  Secondary school  122  3.50  Graduate school  99  2.88  Educational background  N  Mean value  Homogeneity tests for variances Significance  Variance analysis Significance (among groups)  Vocational school/high school  217  3.54  0.065  0.000  University  289  3.53  Professional training/college  146  3.52  Secondary school  122  3.50  Graduate school  99  2.88  Note: The significance level of Homogeneity tests and Variance analysis is 0.05. Source: Derived from the sample data by the authors using SPSS 20.0. There are significant differences among consumers with different educational attainment on the expectations for government subsidies on low-carbon products. In order to verify the significance of the differences, firstly we made the Homogeneity tests for variances. Table 4 reports that the sample has the homogeneity of variance (Sig. 0.065 > 0.05), and the differences is significant according to the result of variance analysis (P < 0.05). To find out the differences of subsidy expectations among different consumers with different educational background, we tested the expectation differences in this dimension by adopting the Bonferroni test (results are presented in Table 5). The results indicate that the consumers with graduate school background are significantly different from other groups on subsidy expectation, but the differences among other groups are not significant. Table 5. Bonferroni Test for different educational background. (I) Educational background  (J) Educational background  Mean difference  Std. error  Significance  95% Confidence interval  (I−J)  Lower bound  Upper bound  Secondary school  Vocational school/high school  −0.044  0.116  1.000  −0.37  0.28  Professional training/college  −0.021  0.126  1.000  −0.38  0.33  University  −0.033  0.111  1.000  −0.35  0.28  Graduate school  0.621*  0.139  0.000  0.23  1.01  Vocational school/high school  Secondary school  0.044  0.116  1.000  −0.28  0.37  Professional training/college  0.023  0.110  1.000  −0.29  0.33  University  0.011  0.092  1.000  −0.25  0.27  Graduate school  0.665*  0.125  0.000  0.31  1.02  Professional training/college  Secondary school  0.021  0.126  1.000  −0.33  0.38  Vocational school/high school  −0.023  0.110  1.000  −0.33  0.29  University  −0.012  0.104  1.000  −0.31  0.28  Graduate school  0.642*  0.134  0.000  0.26  1.02  University  Secondary school  0.033  0.111  1.000  −0.28  0.35  Vocational school/high school  −0.011  0.092  1.000  −0.27  0.25  Professional training/college  0.012  0.104  1.000  −0.28  0.31  Graduate school  0.654*  0.120  0.000  0.32  0.99  Graduate school  Secondary school  −0.621*  0.139  0.000  −1.01  −0.23  Vocational school/high school  −0.665*  0.125  0.000  −1.02  −0.31  Professional training/college  −0.642*  0.134  0.000  −1.02  −0.26  University  −0.654*  0.120  0.000  −0.99  −0.32  (I) Educational background  (J) Educational background  Mean difference  Std. error  Significance  95% Confidence interval  (I−J)  Lower bound  Upper bound  Secondary school  Vocational school/high school  −0.044  0.116  1.000  −0.37  0.28  Professional training/college  −0.021  0.126  1.000  −0.38  0.33  University  −0.033  0.111  1.000  −0.35  0.28  Graduate school  0.621*  0.139  0.000  0.23  1.01  Vocational school/high school  Secondary school  0.044  0.116  1.000  −0.28  0.37  Professional training/college  0.023  0.110  1.000  −0.29  0.33  University  0.011  0.092  1.000  −0.25  0.27  Graduate school  0.665*  0.125  0.000  0.31  1.02  Professional training/college  Secondary school  0.021  0.126  1.000  −0.33  0.38  Vocational school/high school  −0.023  0.110  1.000  −0.33  0.29  University  −0.012  0.104  1.000  −0.31  0.28  Graduate school  0.642*  0.134  0.000  0.26  1.02  University  Secondary school  0.033  0.111  1.000  −0.28  0.35  Vocational school/high school  −0.011  0.092  1.000  −0.27  0.25  Professional training/college  0.012  0.104  1.000  −0.28  0.31  Graduate school  0.654*  0.120  0.000  0.32  0.99  Graduate school  Secondary school  −0.621*  0.139  0.000  −1.01  −0.23  Vocational school/high school  −0.665*  0.125  0.000  −1.02  −0.31  Professional training/college  −0.642*  0.134  0.000  −1.02  −0.26  University  −0.654*  0.120  0.000  −0.99  −0.32  *The mean difference is significant at the 0.05 level. Source: Derived from the sample data by the authors using SPSS 20.0. Table 5. Bonferroni Test for different educational background. (I) Educational background  (J) Educational background  Mean difference  Std. error  Significance  95% Confidence interval  (I−J)  Lower bound  Upper bound  Secondary school  Vocational school/high school  −0.044  0.116  1.000  −0.37  0.28  Professional training/college  −0.021  0.126  1.000  −0.38  0.33  University  −0.033  0.111  1.000  −0.35  0.28  Graduate school  0.621*  0.139  0.000  0.23  1.01  Vocational school/high school  Secondary school  0.044  0.116  1.000  −0.28  0.37  Professional training/college  0.023  0.110  1.000  −0.29  0.33  University  0.011  0.092  1.000  −0.25  0.27  Graduate school  0.665*  0.125  0.000  0.31  1.02  Professional training/college  Secondary school  0.021  0.126  1.000  −0.33  0.38  Vocational school/high school  −0.023  0.110  1.000  −0.33  0.29  University  −0.012  0.104  1.000  −0.31  0.28  Graduate school  0.642*  0.134  0.000  0.26  1.02  University  Secondary school  0.033  0.111  1.000  −0.28  0.35  Vocational school/high school  −0.011  0.092  1.000  −0.27  0.25  Professional training/college  0.012  0.104  1.000  −0.28  0.31  Graduate school  0.654*  0.120  0.000  0.32  0.99  Graduate school  Secondary school  −0.621*  0.139  0.000  −1.01  −0.23  Vocational school/high school  −0.665*  0.125  0.000  −1.02  −0.31  Professional training/college  −0.642*  0.134  0.000  −1.02  −0.26  University  −0.654*  0.120  0.000  −0.99  −0.32  (I) Educational background  (J) Educational background  Mean difference  Std. error  Significance  95% Confidence interval  (I−J)  Lower bound  Upper bound  Secondary school  Vocational school/high school  −0.044  0.116  1.000  −0.37  0.28  Professional training/college  −0.021  0.126  1.000  −0.38  0.33  University  −0.033  0.111  1.000  −0.35  0.28  Graduate school  0.621*  0.139  0.000  0.23  1.01  Vocational school/high school  Secondary school  0.044  0.116  1.000  −0.28  0.37  Professional training/college  0.023  0.110  1.000  −0.29  0.33  University  0.011  0.092  1.000  −0.25  0.27  Graduate school  0.665*  0.125  0.000  0.31  1.02  Professional training/college  Secondary school  0.021  0.126  1.000  −0.33  0.38  Vocational school/high school  −0.023  0.110  1.000  −0.33  0.29  University  −0.012  0.104  1.000  −0.31  0.28  Graduate school  0.642*  0.134  0.000  0.26  1.02  University  Secondary school  0.033  0.111  1.000  −0.28  0.35  Vocational school/high school  −0.011  0.092  1.000  −0.27  0.25  Professional training/college  0.012  0.104  1.000  −0.28  0.31  Graduate school  0.654*  0.120  0.000  0.32  0.99  Graduate school  Secondary school  −0.621*  0.139  0.000  −1.01  −0.23  Vocational school/high school  −0.665*  0.125  0.000  −1.02  −0.31  Professional training/college  −0.642*  0.134  0.000  −1.02  −0.26  University  −0.654*  0.120  0.000  −0.99  −0.32  *The mean difference is significant at the 0.05 level. Source: Derived from the sample data by the authors using SPSS 20.0. Analysis by age Then, we went on to test the expectation differences in the age dimension by adopting one-way ANOVA, with the following findings. Middle-aged consumers have the lowest expectations for government subsidies on low-carbon products, but the youth group has the highest. The maximum mean value (30–39 and 40–49 groups) is 9.76% higher than the minimum mean value (20–29 group), demonstrating a significant difference in the expectations by age (Table 6). Middle-aged consumers (30–39 and 40–49 groups) always have stable jobs and high income; therefore, they are not sensitive to the subsidies and have the lowest expectations. Additionally, consumers between 20 and 29 years old exhibit highest expectations for the government subsidies on low-carbon products, which may be due to the large income elasticity and great life pressure. There are significant differences among consumers with different age on the expectations for government subsidies on low-carbon products. After the Homogeneity tests for variances, the results show that the sample has the homogeneity of variance (Sig. 0.243 > 0.05), and the differences is significant according to the result of Variance analysis (P < 0.05). Table 6. Test results by age and the ranking of expectation mean values. Age  N  Mean value  Homogeneity tests for variances Significance  Variance analysis Significance (among groups)  20–29  341  3.60  0.243  0.001  Over 50  131  3.57  Under 19  30  3.47  30–39  198  3.28  40–49  173  3.28  Age  N  Mean value  Homogeneity tests for variances Significance  Variance analysis Significance (among groups)  20–29  341  3.60  0.243  0.001  Over 50  131  3.57  Under 19  30  3.47  30–39  198  3.28  40–49  173  3.28  Note: The significance level of Homogeneity tests and Variance analysis is 0.05. Source: Derived from the sample data by the authors using SPSS 20.0. Table 6. Test results by age and the ranking of expectation mean values. Age  N  Mean value  Homogeneity tests for variances Significance  Variance analysis Significance (among groups)  20–29  341  3.60  0.243  0.001  Over 50  131  3.57  Under 19  30  3.47  30–39  198  3.28  40–49  173  3.28  Age  N  Mean value  Homogeneity tests for variances Significance  Variance analysis Significance (among groups)  20–29  341  3.60  0.243  0.001  Over 50  131  3.57  Under 19  30  3.47  30–39  198  3.28  40–49  173  3.28  Note: The significance level of Homogeneity tests and Variance analysis is 0.05. Source: Derived from the sample data by the authors using SPSS 20.0. Then, by adopting the Bonferroni test, we tested the expectation differences in this dimension (results are shown in Table 7). The results indicate that the consumers in age 20–29 are significantly different from the 30–39 and 40–49 groups on subsidy expectations, but the differences among other groups are not significant. In addition, the mean values of consumer expectations for government subsidies on low-carbon products are presented in Z-shape ranked by age on an ascending order. Table 7. Bonferroni test for different ages. (I) Age  (J) Age  Mean difference (I−J)  Std. error  Significance  95% Confidence interval  Lower bound  Upper bound  Under 19  20–29  −0.135  0.198  1.000  −0.69  0.42  30–39  0.189  0.203  1.000  −0.38  0.76  40–49  0.189  0.205  1.000  −0.39  0.77  Over 50  −0.106  0.210  1.000  −0.70  0.49  20–29  Under 19  0.135  0.198  1.000  −0.42  0.69  30–39  0.323*  0.093  0.005  0.06  0.58  40–49  0.324*  0.097  0.009  0.05  0.60  Over 50  0.029  0.107  1.000  −0.27  0.33  30–39  Under 19  −0.189  0.203  1.000  −0.76  0.38  20–29  −0.323*  0.093  0.005  −0.58  −0.06  40–49  0.000  0.108  1.000  −0.30  0.30  Over 50  −0.295  0.117  0.119  −0.62  0.03  40–49  Under 19  −0.189  0.205  1.000  −0.77  0.39  20–29  −0.324*  0.097  0.009  −0.60  −0.05  30–39  0.000  0.108  1.000  −0.30  0.30  Over 50  −0.295  0.120  0.143  −0.63  0.04  Over 50  Under 19  0.106  0.210  1.000  −0.49  0.70  20–29  −0.029  0.107  1.000  −0.33  0.27  30–39  0.295  0.117  0.119  −0.03  0.62  40–49  0.295  0.120  0.143  −0.04  0.63  (I) Age  (J) Age  Mean difference (I−J)  Std. error  Significance  95% Confidence interval  Lower bound  Upper bound  Under 19  20–29  −0.135  0.198  1.000  −0.69  0.42  30–39  0.189  0.203  1.000  −0.38  0.76  40–49  0.189  0.205  1.000  −0.39  0.77  Over 50  −0.106  0.210  1.000  −0.70  0.49  20–29  Under 19  0.135  0.198  1.000  −0.42  0.69  30–39  0.323*  0.093  0.005  0.06  0.58  40–49  0.324*  0.097  0.009  0.05  0.60  Over 50  0.029  0.107  1.000  −0.27  0.33  30–39  Under 19  −0.189  0.203  1.000  −0.76  0.38  20–29  −0.323*  0.093  0.005  −0.58  −0.06  40–49  0.000  0.108  1.000  −0.30  0.30  Over 50  −0.295  0.117  0.119  −0.62  0.03  40–49  Under 19  −0.189  0.205  1.000  −0.77  0.39  20–29  −0.324*  0.097  0.009  −0.60  −0.05  30–39  0.000  0.108  1.000  −0.30  0.30  Over 50  −0.295  0.120  0.143  −0.63  0.04  Over 50  Under 19  0.106  0.210  1.000  −0.49  0.70  20–29  −0.029  0.107  1.000  −0.33  0.27  30–39  0.295  0.117  0.119  −0.03  0.62  40–49  0.295  0.120  0.143  −0.04  0.63  *The mean difference is significant at the 0.05 level. Source: Derived from the sample data by the authors using SPSS 20.0. Table 7. Bonferroni test for different ages. (I) Age  (J) Age  Mean difference (I−J)  Std. error  Significance  95% Confidence interval  Lower bound  Upper bound  Under 19  20–29  −0.135  0.198  1.000  −0.69  0.42  30–39  0.189  0.203  1.000  −0.38  0.76  40–49  0.189  0.205  1.000  −0.39  0.77  Over 50  −0.106  0.210  1.000  −0.70  0.49  20–29  Under 19  0.135  0.198  1.000  −0.42  0.69  30–39  0.323*  0.093  0.005  0.06  0.58  40–49  0.324*  0.097  0.009  0.05  0.60  Over 50  0.029  0.107  1.000  −0.27  0.33  30–39  Under 19  −0.189  0.203  1.000  −0.76  0.38  20–29  −0.323*  0.093  0.005  −0.58  −0.06  40–49  0.000  0.108  1.000  −0.30  0.30  Over 50  −0.295  0.117  0.119  −0.62  0.03  40–49  Under 19  −0.189  0.205  1.000  −0.77  0.39  20–29  −0.324*  0.097  0.009  −0.60  −0.05  30–39  0.000  0.108  1.000  −0.30  0.30  Over 50  −0.295  0.120  0.143  −0.63  0.04  Over 50  Under 19  0.106  0.210  1.000  −0.49  0.70  20–29  −0.029  0.107  1.000  −0.33  0.27  30–39  0.295  0.117  0.119  −0.03  0.62  40–49  0.295  0.120  0.143  −0.04  0.63  (I) Age  (J) Age  Mean difference (I−J)  Std. error  Significance  95% Confidence interval  Lower bound  Upper bound  Under 19  20–29  −0.135  0.198  1.000  −0.69  0.42  30–39  0.189  0.203  1.000  −0.38  0.76  40–49  0.189  0.205  1.000  −0.39  0.77  Over 50  −0.106  0.210  1.000  −0.70  0.49  20–29  Under 19  0.135  0.198  1.000  −0.42  0.69  30–39  0.323*  0.093  0.005  0.06  0.58  40–49  0.324*  0.097  0.009  0.05  0.60  Over 50  0.029  0.107  1.000  −0.27  0.33  30–39  Under 19  −0.189  0.203  1.000  −0.76  0.38  20–29  −0.323*  0.093  0.005  −0.58  −0.06  40–49  0.000  0.108  1.000  −0.30  0.30  Over 50  −0.295  0.117  0.119  −0.62  0.03  40–49  Under 19  −0.189  0.205  1.000  −0.77  0.39  20–29  −0.324*  0.097  0.009  −0.60  −0.05  30–39  0.000  0.108  1.000  −0.30  0.30  Over 50  −0.295  0.120  0.143  −0.63  0.04  Over 50  Under 19  0.106  0.210  1.000  −0.49  0.70  20–29  −0.029  0.107  1.000  −0.33  0.27  30–39  0.295  0.117  0.119  −0.03  0.62  40–49  0.295  0.120  0.143  −0.04  0.63  *The mean difference is significant at the 0.05 level. Source: Derived from the sample data by the authors using SPSS 20.0. Analysis by region Based on the field survey in six cities (regions), we analyzed the expectation differences in the “Region” dimension by using the methodology of one-way ANOVA and obtained the following results. The expectations for government subsidies on low-carbon products in different regions are closely related to the level of economic development. The mean values of expectations for government subsidies on low-carbon products of different regions are ranked as follows based on a descending order: Xiangyang (3.59), Xuchang (3.53), Shenzhen (3.49), Daqing (3.41), Gucheng (3.37) and Wuhan (3.30), and the maximum mean value is 8.79% higher than the minimum value (Table 8). The differences of expectations for government subsidies on low-carbon products of different regions do not reach a significant level. The results of Homogeneity tests for variances show that it is non-homogeneity of variance (Sig. 0.022 < 0.05, see Table 8). The results of one-way ANOVA indicate that the differences are not significant in different cities (P > 0.05, see Table 8). Then, we conducted Dunnett’s T3 test to further analyze the differences. The results show that there is no significant difference among the six cities (regions), which is consistent with the previous results. Table 8. Test results by region and the ranking of expectation mean values. Region  N  Mean value  Homogeneity tests for variances Significance  Variance analysis Significance (among groups)  Xiangyang  147  3.59  0.022  0.216  Xuchang  154  3.53  Shenzhen  169  3.49  Daqing  142  3.41  Gucheng  147  3.37  Wuhan  114  3.30  Region  N  Mean value  Homogeneity tests for variances Significance  Variance analysis Significance (among groups)  Xiangyang  147  3.59  0.022  0.216  Xuchang  154  3.53  Shenzhen  169  3.49  Daqing  142  3.41  Gucheng  147  3.37  Wuhan  114  3.30  Note: The significance level of Homogeneity tests and variance analysis is 0.05. Source: Derived from the sample data by the authors using SPSS 20.0. Table 8. Test results by region and the ranking of expectation mean values. Region  N  Mean value  Homogeneity tests for variances Significance  Variance analysis Significance (among groups)  Xiangyang  147  3.59  0.022  0.216  Xuchang  154  3.53  Shenzhen  169  3.49  Daqing  142  3.41  Gucheng  147  3.37  Wuhan  114  3.30  Region  N  Mean value  Homogeneity tests for variances Significance  Variance analysis Significance (among groups)  Xiangyang  147  3.59  0.022  0.216  Xuchang  154  3.53  Shenzhen  169  3.49  Daqing  142  3.41  Gucheng  147  3.37  Wuhan  114  3.30  Note: The significance level of Homogeneity tests and variance analysis is 0.05. Source: Derived from the sample data by the authors using SPSS 20.0. Analysis by gender We made the one-way ANOVA and independent samples T test on different genders and found that: The mean value of men’s expectations is a little higher than that of women. Because the number of groups is smaller than 3, we cannot conduct the Homogeneity tests for variances. According to the results of descriptive statistical analysis in Table 9, we can find that the mean value of men’s expectations for government subsidies on low-carbon products is 3.47, which is higher than that of women (3.45), but the disparity is quite small (by just 0.58%). The expectation differences between men and women are not significant. In the result of one-way ANOVA and independent samples T test, the Sig. value is 0.765 (0.765 > 0.05, see Tables 9 and 10). Therefore, we can conclude that people of different genders have pretty much the same expectations for government subsidies on low-carbon products. Table 9. Test results by gender and the ranking of expectation mean values. Gender  N  Mean value  Variance analysis  Significance (among groups)  Male  366  3.47  0.765  Female  507  3.45  Sum  873  3.45  Gender  N  Mean value  Variance analysis  Significance (among groups)  Male  366  3.47  0.765  Female  507  3.45  Sum  873  3.45  Note: The significance level of variance analysis is 0.05. Source: Derived from the sample data by the authors using SPSS 20.0. Table 9. Test results by gender and the ranking of expectation mean values. Gender  N  Mean value  Variance analysis  Significance (among groups)  Male  366  3.47  0.765  Female  507  3.45  Sum  873  3.45  Gender  N  Mean value  Variance analysis  Significance (among groups)  Male  366  3.47  0.765  Female  507  3.45  Sum  873  3.45  Note: The significance level of variance analysis is 0.05. Source: Derived from the sample data by the authors using SPSS 20.0. Table 10. Independent-samples T test for different genders.   Levene’s test for equality of variances  T-test for equality of means  F  Significance  t  df  Significance (2-tailed)  Mean difference  Std. error difference  95% Confidence interval of the difference  Lower  Upper  Expectation for government subsidies  Equal variances assumed  0.941  0.332  0.298  871  0.765  0.021  0.072  −0.120  0.163  Equal variances not assumed      0.297  772.150  0.767  0.021  0.072  −0.120  0.163    Levene’s test for equality of variances  T-test for equality of means  F  Significance  t  df  Significance (2-tailed)  Mean difference  Std. error difference  95% Confidence interval of the difference  Lower  Upper  Expectation for government subsidies  Equal variances assumed  0.941  0.332  0.298  871  0.765  0.021  0.072  −0.120  0.163  Equal variances not assumed      0.297  772.150  0.767  0.021  0.072  −0.120  0.163  Source: Derived from the sample data by the authors using SPSS 20.0. Table 10. Independent-samples T test for different genders.   Levene’s test for equality of variances  T-test for equality of means  F  Significance  t  df  Significance (2-tailed)  Mean difference  Std. error difference  95% Confidence interval of the difference  Lower  Upper  Expectation for government subsidies  Equal variances assumed  0.941  0.332  0.298  871  0.765  0.021  0.072  −0.120  0.163  Equal variances not assumed      0.297  772.150  0.767  0.021  0.072  −0.120  0.163    Levene’s test for equality of variances  T-test for equality of means  F  Significance  t  df  Significance (2-tailed)  Mean difference  Std. error difference  95% Confidence interval of the difference  Lower  Upper  Expectation for government subsidies  Equal variances assumed  0.941  0.332  0.298  871  0.765  0.021  0.072  −0.120  0.163  Equal variances not assumed      0.297  772.150  0.767  0.021  0.072  −0.120  0.163  Source: Derived from the sample data by the authors using SPSS 20.0. REGRESSION ANALYSIS OF CONSUMERS’ EXPECTATION FACTORS In order to figure out the city (region) and population factors on people’s expectations for government subsidies on low-carbon products and further define their degrees and significance of effect, we conducted a linear regression and set the ‘expectations for government subsidies on low-carbon products’ as the dependent variable, taking different cities (regions) and population factors as independent variables. According to the regression coefficients and significance level, we can find that educational attainment and monthly income of consumers have important effects on their expectations for government subsidies on low-carbon products. The regression coefficients of educational attainment and monthly income are significantly associated with the consumers’ subsidy expectations. Monthly income exerts the greatest influence (−0.167) on consumers’ expectations, which is significant at the 1% level and the coefficient is negative. Educational attainment is significant at the 10% level and the coefficient (−0.062) is negative as well. The results in Table 11 indicate that the lower their monthly income and educational attainment, the stronger their expectations. The regression results also further verify the conclusions that consumer expectations for government subsidies on low-carbon products has a negative correlation with their educational attainment and monthly income as stated earlier. Table 11. The regression results. Model  Unstandardized coefficients  Standardized coefficients  t  Significance  B  Std. error  Beta  (Constant)  4.160  0.220    18.941  0.000  Region  0.026  0.020  0.044  1.293  0.196  Age  −0.024  0.032  −0.027  −0.753  0.452  Gender  −0.111  0.073  −0.052  −1.524  0.128  Educational background  −0.051  0.031  −0.062  −1.652  0.099*  Monthly income  −0.168  0.038  −0.167  −4.404  0.000***  Model  Unstandardized coefficients  Standardized coefficients  t  Significance  B  Std. error  Beta  (Constant)  4.160  0.220    18.941  0.000  Region  0.026  0.020  0.044  1.293  0.196  Age  −0.024  0.032  −0.027  −0.753  0.452  Gender  −0.111  0.073  −0.052  −1.524  0.128  Educational background  −0.051  0.031  −0.062  −1.652  0.099*  Monthly income  −0.168  0.038  −0.167  −4.404  0.000***  *The mean difference is significant at the 0.1 level, **means 0.05, ***means 0.01. Source: Derived from the sample data by the authors using SPSS 20.0. Table 11. The regression results. Model  Unstandardized coefficients  Standardized coefficients  t  Significance  B  Std. error  Beta  (Constant)  4.160  0.220    18.941  0.000  Region  0.026  0.020  0.044  1.293  0.196  Age  −0.024  0.032  −0.027  −0.753  0.452  Gender  −0.111  0.073  −0.052  −1.524  0.128  Educational background  −0.051  0.031  −0.062  −1.652  0.099*  Monthly income  −0.168  0.038  −0.167  −4.404  0.000***  Model  Unstandardized coefficients  Standardized coefficients  t  Significance  B  Std. error  Beta  (Constant)  4.160  0.220    18.941  0.000  Region  0.026  0.020  0.044  1.293  0.196  Age  −0.024  0.032  −0.027  −0.753  0.452  Gender  −0.111  0.073  −0.052  −1.524  0.128  Educational background  −0.051  0.031  −0.062  −1.652  0.099*  Monthly income  −0.168  0.038  −0.167  −4.404  0.000***  *The mean difference is significant at the 0.1 level, **means 0.05, ***means 0.01. Source: Derived from the sample data by the authors using SPSS 20.0. CONCLUSIONS AND RECOMMENDATIONS Conclusions Based on the above statistical analyses, we can draw the following conclusions: Significant differences exist in the expectations for government subsidies on low-carbon products from different types of consumers. The maximum mean values of expectations are 22.92, 22.67, 9.76, 8.79 and 0.58% higher than the minimum values for educational attainment, monthly income, age, region and gender, respectively. There are significant differences among consumers with different monthly income, educational attainment and age on the expectations for government subsidies on low-carbon products, but the difference is not significant for the consumers in different regions or genders. Monthly income and educational attainment exert significant influences on consumer expectations for government subsidies on low-carbon products, and the influence of monthly income is the biggest. The results of one-way ANOVA and regression both indicate that there is a negative relationship between monthly income and subsidy expectations, and similarly that between educational attainment and subsidy expectations. That means, ceteris paribus, the lower their monthly income, the stronger their expectations for government subsidies on low-carbon products, so is their educational attainment. Therefore, for the current average income and education levels of the Chinese residents, it is not a mature time for an overall implementation of low-carbon consumption. Policy recommendations From the above conclusions, we propose the following policy recommendations. Fully play the demonstrative role of specific groups of consumers on low-carbon consumption. Take middle-aged female consumers that have high income and good education in the developed areas as the breakthrough, let them take the lead in implementing the strategy of low-carbon consumption throughout the country. According to the differences in purchase behavior of different types of consumers, the government should develop different strategies to promote low-carbon products, make the consumers with robust purchasing power to drive the consumption demand of low-carbon products. Develop the economy, strengthen education and improve consumers’ income. The prices of low-carbon products are always higher than conventional products. If the consumers do not have a fair income, they could not afford or are not willing to pay for low-carbon products. Educated consumers are often more aware of the significance of low-carbon consumption and environmental protection. Therefore, the subsidy policy is often the short-term tactic, but the development of the economy and education can essentially provide continuous power for low-carbon consumption in the long run. ACKNOWLEDGEMENTS The study is supported by the National Natural Science Foundation of China (NSFC) (Nos. 71473231, 71773119 and 71173201) ‘Research on the Impact Mechanism of Carbon Tariff and Carbon Labeling on Agri-trade and Carbon Reduction’. REFERENCES 1 Lewis JI. The evolving role of carbon finance in promoting renewable energy development in China. Energy Policy  2010; 38: 2875– 86. Google Scholar CrossRef Search ADS   2 Dong L, Fujita T, Zhang H et al.  . Promoting low-carbon city through industrial symbiosis: a case in China by applying HPIMO model. Energy policy  2013; 61: 864– 73. Google Scholar CrossRef Search ADS   3 Du S, Tang W, Song M. Low-carbon production with low-carbon premium in cap-and-trade regulation. J Cleaner Prod  2016; 134: 652– 62. Google Scholar CrossRef Search ADS   4 Gupta M. Willingness to pay for carbon tax: a study of Indian road passenger transport. Transp Policy  2016; 45: 46– 54. Google Scholar CrossRef Search ADS   5 Bigerna S, Bollino CA, Micheli S et al.  . Revealed and stated preferences for CO2 emissions reduction: the missing link. Renew Sustain Energy Rev  2017; 68: 1213– 21. Google Scholar CrossRef Search ADS   6 Newbery DM. Towards a green energy economy? The EU Energy Union’s transition to a low-carbon zero subsidy electricity system–lessons from the UK’s Electricity Market Reform. Appl Energy  2016; 179: 1321– 30. Google Scholar CrossRef Search ADS   7 Li Y, Fan J, Zhao D et al.  . Tiered gasoline pricing: a personal carbon trading perspective. Energy Policy  2016; 89: 194– 201. Google Scholar CrossRef Search ADS   8 Bunn DW, Muñoz JI. Supporting the externality of intermittency in policies for renewable energy. Energy Policy  2016; 88: 594– 602. Google Scholar CrossRef Search ADS   9 Sun J, Xiao Z, Zhou G. Two stage supply chain enterprises’ production and reduction decision-making mechanism research considering emission trading. J Ind Intell Inf  2016; 4: 46– 50. 10 Campiglio E. Beyond carbon pricing: the role of banking and monetary policy in financing the transition to a low-carbon economy. Ecol Econ  2016; 121: 220– 30. Google Scholar CrossRef Search ADS   11 Alonso PM, Hewitt R, Pacheco JD et al.  . Losing the roadmap: renewable energy paralysis in Spain and its implications for the EU low carbon economy. Renew Energy  2016; 89: 680– 94. Google Scholar CrossRef Search ADS   12 Liu W, Qin B. Low-carbon city initiatives in China: a review from the policy paradigm perspective. Cities  2016; 51: 131– 8. Google Scholar CrossRef Search ADS   13 Ling Y, Xu J, 2016. Low Carbon Consumption Preference and the Related Enterprises’ Optimal Strategies. Available at SSRN 2745693. Available at SSRN: http://ssrn.com/abstract=2745693. 14 He Y, Xu Y, Pang Y et al.  . A regulatory policy to promote renewable energy consumption in China: review and future evolutionary path. Renewable Energy  2016; 89: 695– 705. Google Scholar CrossRef Search ADS   15 Huang B, Mauerhofer V, Geng Y. Analysis of existing building energy saving policies in Japan and China. J Cleaner Prod  2016; 112: 1510– 8. Google Scholar CrossRef Search ADS   16 Sun Y. Research on the subsidy policy for the development of low-carbon economy. Public Finance Res  2010; 4: 59– 60. 17 Shen M, He Z. The experience of foreign fiscal policy on low-carbon economy. Ecol Econ  2011; 3: 83– 9. 18 Bollino CA. The willingness to pay for renewable energy sources: the case of Italy with socio-demographic determinants. Energy J  2009; 30: 81– 96. 19 Shukla PR, Chaturvedi V. Low carbon and clean energy scenarios for India: analysis of targets approach. Energy Econ  2012; 34: S487– 95. Google Scholar CrossRef Search ADS   20 Jin M, Zheng S. Analysis of the game behavior in green agricultural products market of China. Finance Trade Econ  2006; 6: 38– 41. 21 Roe B, Teisl MF, Levy A et al.  . US consumers’ willingness to pay for green electricity. Energy policy  2001; 29: 917– 25. Google Scholar CrossRef Search ADS   22 Zarnikau J. Consumer demand for ‘green power’ and energy efficiency. Energy Policy  2003; 31: 1661– 72. Google Scholar CrossRef Search ADS   23 Galinato GI, Yoder JK. An integrated tax-subsidy policy for carbon emission reduction. Resour Energy Econ  2010; 32: 310– 26. Google Scholar CrossRef Search ADS   24 Bajona C, Kelly DL. Trade and the environment with pre-existing subsidies: a dynamic general equilibrium analysis. J Environ Econ Manage  2012; 64: 253– 78. Google Scholar CrossRef Search ADS   25 Lin B, Li A. Impacts of removing fossil fuel subsidies on China: how large and how to mitigate? Energy  2012; 44: 741– 9. Google Scholar CrossRef Search ADS   26 Lapan H, Moschini G. Second-best biofuel policies and the welfare effects of quantity mandates and subsidies. J Environ Econ Manage  2012; 63: 224– 41. Google Scholar CrossRef Search ADS   27 Song L, Hua B. A comparative analysis of the fiscal policy system for different countries to develop low carbon economy. J Yunnan Univ Finance Econ  2011; 27: 98– 105. 28 He S, Wang F. Cost report model based on the mechanism of low-carbon subsidies. Energy Procedia  2011; 5: 1869– 73. Google Scholar CrossRef Search ADS   29 Yang J. The Game analysis of government subsidies in low carbon economy. Commer Res  2010; 8: 109– 12. 30 Brzeskot M, Haupt A. Environmental policy and the energy efficiency of vertically differentiated consumer products. Energy Econ  2013; 36: 444– 53. Google Scholar CrossRef Search ADS   31 Xiong D. Research on fiscal policies guiding low-carbon consumption of urban residents. J Finance Econ  2012; 2: 72– 5. 32 Bansal S, Gangopadhyay S. Tax/subsidy policies in the presence of environmentally aware consumers. J Environ Econ Manage  2003; 45: 333– 55. Google Scholar CrossRef Search ADS   33 Zhou L, Zheng X. Green subsidy policy effect assessment based on farmers’ willingness to pay the low-carbon elements: an empirical study in pig industry. J Nanjing Agric Univ  2012; 12: 85– 91. 34 Li Y, Zhao D. Research on R&D cost allocation comparison for low-carbon supply chain based on government’ subsidies. Soft Sci  2014; 28: 21– 6. 35 Xie S, Kuang Y, Huang N. Main paths and policy proposals for the development of carbon sinking agriculture in China. China Popul Resour Environ  2010; 20: 46– 51. 36 Fan R, Dong L, Yang W et al.  . Study on the optimal supervision strategy of government low-carbon subsidy and the corresponding efficiency and stability in the small-world network context. J Cleaner Prod  2017; 168. DOI:10.1016/j.jclepro.2017.09.044. 37 Li Q, Long R, Chen H. Empirical study of the willingness of consumers to purchase low-carbon products by considering carbon labels: a case study. J Cleaner Prod  2017. DOI: 10.1016/j.jclepro. 2017.04.154. 38 Foxall GR, Goldsmith RE, Brown S. Consumer Psychology for Marketing ; Vol. 1. Cengage Learning EMEA, Oxford, UK, 1994. © The Author(s) 2018. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png International Journal of Low-Carbon Technologies Oxford University Press

What do consumers expect for government subsidies on low-carbon products in China?

Loading next page...
 
/lp/ou_press/what-do-consumers-expect-for-government-subsidies-on-low-carbon-kljAef2jlU
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press.
ISSN
1748-1317
eISSN
1748-1325
D.O.I.
10.1093/ijlct/cty005
Publisher site
See Article on Publisher Site

Abstract

Abstract Based on the 873 questionnaires collected from six cities in four provinces in China, we made a quantitative analysis of different types of consumers’ expectations for government low-carbon subsidies by using the SPSS. The results indicate that: (1) Significant differences exist in the ‘expectations for government subsidies on low-carbon products’ from different types of consumers; there are significant differences among consumers with different monthly income, educational attainment and age on the ‘expectations for government subsidies on low-carbon products’, but the difference is not significant for the consumers in different regions or gender. (2) Monthly income and educational attainment exert significant influences on consumer ‘expectations for government subsidies on low-carbon products’, and the influence of monthly income is the biggest. Finally, we put forward policy recommendations accordingly. INTRODUCTION With the global climate change and increasing pressure for GHG emissions reduction, the development of low-carbon economy has become the top priority for all countries throughout the world. It is inevitable for developing low-carbon economy to achieve low-carbon consumption. Currently, the economic development and people’s income in China are still under development. Therefore, government subsidies are an important means to promote low-carbon consumption [1, 2]. Government subsidies for low-carbon products can stimulate consumers’ willingness to buy low-carbon products, to a certain extent [3–5]. So, what is the general expectation of government subsidies on low-carbon products? Which properties of consumers have significant impacts on their expectation? Are there any differences among different types of consumers in their expectations? The research literature related to government subsidies on low-carbon products and development of low-carbon economy mainly includes the following four aspects: (1) research on the role of government subsidies to the development of low-carbon economy [6]; (2) research on the mechanism of government subsidies [7, 8]; (3) research on the directions and objects of government subsidies [9, 10]; and (4) research on the forms of government subsidies [11, 12]. Most of the researchers believe that government subsidies play an important role in the carbon reduction process [13], and will be effective in promoting low-carbon consumption and energy-saving and emissions reduction [14, 15]. Sun [16] in his research point out that a country can achieve the goal of developing a low-carbon economy and the promotion of energy conservation and emissions reduction through adopting government subsidies on low-carbon products. Shen and He [17] consider that subsidies on low-carbon resources and products are important fiscal instrument to reducing carbon emissions, as well as increasing the consumption of low-carbon products. Bollino [18] explores the existence of consumer’s Willingness to Pay (WTP) in order to use renewable energy in the electricity production, and finds that the aggregate WTP for RES in Italy is not enough to attain the Government goal in 2010, which implies that the development of low-carbon industry, such as renewable energy, needs government subsidies. Moreover, low-carbon energy technologies are of increasing importance to India for reducing emissions and diversifying its energy supply mix [19]. Jin and Zheng [20] present a game analysis of both supply and demand sides in the green agricultural product market, and suggest that government compensation policy towards low-carbon agricultural consumers can help promote the development of low-carbon products. Roe et al. [21] and Zarnikau [22] suggest that public policy can have a positive and significant impact on consumers’ voluntarily payment of the additional costs for low-carbon products. Through the study of low-carbon product subsidies of the United States, Galinato and Yoder [23] find that the price subsidies for low-carbon products can be more efficient on carbon emissions than the tax on high-carbon products, which has similar findings with Bajona and Kelly’s [24] research on Chinese market in 1997. If China removes low-carbon subsidies, it will have negative impact on world carbon emissions [25]. Lapan and Moschini [26] construct a general equilibrium economy model to study on both the positive and normative implications of alternative policy instruments, including the subsidies and mandates. Their research shows that government subsidies for clean energy can implement carbon reduction goals effectively. Due to the imperfection of market mechanism and the lack of legal binding, the enterprises misrepresent the cost for more subsidies [27]. To solve this problem, He and Wang [28] established a cost reporting model, which can ensure the authenticity of the cost reports. Using GCAM, an integrated assessment model, Shukla and Chaturvedi [19] analyze a targets approach for pushing solar, wind and nuclear technologies in the Indian electricity generation sector from 2005 to 2095 in different scenarios, and find that the subsidy is still necessary in the short run when the carbon price is low. In terms of government subsidy directions and objects, Yang [29] summarizes the effective policies to promote low-carbon economy formulated by different countries, and then analyzes the government, enterprises and consumers in low-carbon economy through Game theory. He puts forward that government subsidies should focus on the consumers. Similarly, Sun’s [16] research proposes that government subsidies should pay more attention to the consumption link rather than the circulation chain. Shen and He [17] find that government subsidies for enterprises mainly focus on the industry of renewable power generation in the developed countries, such as Germany and the United States. But the government subsidies for consumers primarily involve the auto, real estate and household appliance industries, etc. Other scholars argue that government subsidy focus should depend on different situations. Brzeskot and Haupt [30] detect that an industry-friendly government levies an energy tax to supplement a lax product standard, but shies away from subsidies to firms; by contrast, a consumer-friendly government depends heavily on a strict product standard and additionally implements a moderate subsidy to firms, but avoids energy taxes. Xiong [31] points out that the government should take full advantage of the alternative role of financial subsidies to encourage city residents to buy energy-efficient appliances. As for government subsidy forms, Bansal and Gangopadhyay [32] think that differentiated subsidies can be more efficient than the unified subsidy on the improvement of environmental quality and social welfare, which is supported by Zhou and Zheng [33]. Li and Zhao [34] construct a Nash game and Stackelberg game model of manufacturer and retailer, and analyze the impact of R&D cost allocation coefficient and subsidies on the low-carbon investment of supply chain, and finally find out the strategies of low-carbon R&D cooperation and government subsidies under different game scenarios. Xie et al. [35] suggest that government should encourage various types of capital to the rural areas to promote the development of agricultural carbon sinks, as well as subsidize domestic agricultural carbon sinks by imposing carbon tariffs on imported agricultural products. Fan et al. [36] analyze the optimal strategy for the government to supervise low-carbon subsidies. Li et al. [37] find that consumers’ willingness to purchase low-carbon products differed significantly with respect to values, age, income and education. Literature review shows that: So far, there are little empirical study on the consumer differences of government subsidies from the perspective of demography. Meanwhile, the persuasion theories point out that the characteristics of the recipient, namely, the differences of consumers’ educational background, income, attitude, gender and age, may affect the interpretation and reaction of recipient to the information [38]. Therefore, a meticulous research on different consumer groups should be conducted to develop targeted measures as well as achieve maximum persuasion effect. In this study, based on the demographic view of different consumers, we make a quantitative analysis of the expectation differences of the low-carbon subsidies for different types of consumers by adopting the methodologies of one-way ANOVA, Bonferroni test, Dunnett’s T3 and regression model, in an attempt to reveal the potential trends of the subsidy expectations of different consumers. The reminder of this article is organized as follows: Research Hypothesis describes the empirical methodologies and data sources. Testing Results and Analyses analyzes the test results. Regression Analysis of Consumers’ Expectation conducts the regression analysis. Conclusions and Recommendations is the conclusions and recommendations of this study. RESEARCH HYPOTHESIS In this paper, we are analyzing what consumers expect for low-carbon product subsidies from the government. In this case, the definition of consumers’ expectations for government subsidies refers to ‘public demand for subsidies for low-carbon products’. In light of the technical and socio-economic conditions in present China, low-carbon products generally have higher costs than conventional high-carbon products. Therefore, in the beginning stage, government subsidies are necessary to stimulate consumers to accept and procure low-carbon products. Therefore, we proposed the following hypothesis for this research: different types of consumers have different expectations for government subsidies which could encourage them to buy low-carbon products; to be more specific, people differing in monthly income and educational levels, ages, regions and genders may have different subsidy expectations. METHODOLOGIES AND SAMPLE DATA Methodologies Firstly, we acquired the first-hand data through questionnaire survey, in an attempt to find out the ‘expectations for government subsidies on low-carbon products’ of different consumers in different regions, ages, educational backgrounds, incomes and genders. Individuals are randomly sampled from selected cities in China, and then the respondents completed the questionnaires at specific locations. Based on the questionnaire data of 873 consumers in six cities of China, we utilized the SPSS 20.0 software to conduct a one-way ANOVA, Bonferroni test and Dunnett’s T3 test to quantitatively analyze the differences in consumer expectations for government subsidies on low-carbon products. Then, by adopting the linear regression model, we further verified the influence of region and demographic variables on their expectations. Sample data The sample data come from the field questionnaire survey in six cities of four provinces in China. The geographical distribution of the samples covers the southern, central and northern part of China, with broad representation in terms of economic development and population distribution. In accordance with the random principle, we have taken the geographical distribution of various types of consumer groups into account and given out questionnaires in the densely populated areas, such as supermarkets, office buildings or residential districts. A total of 950 questionnaires were handed out and 876 received, providing a response rate of 92.21%, among which 873 are valid, accounting for 99.66%. As shown in Table 1, in the demographic variable of gender, females outnumber males with 58.08–41.92%. The samples cover six regions, whose frequencies are basically balanced: ranging from 13.06% (Wuhan) to 19.36% (Shenzhen). In the demographic variable of age, the youths and the middle aged are the main consumers of low-carbon products when those aged between 20 and 49 constitute 81.30%. In the dimension of income, the percentage of those earning <USD 761 per month is 86.40% and those earning 151–455 USD per month is 45.70%. Therefore, the sample is fairly typical and representative for Chinese consumers from a general perspective. Table 1. The distribution of samples. Features  Frequency  Ratio (%)  1. Gender  Male  366  41.92  Female  507  58.08  2. Age  Under 19  30  3.44  20–29  341  39.06  30–39  198  22.68  40–49  173  19.82  Over 50  131  15.01  3. Monthly income  Under 150 USD  167  19.13  151–455 USD  399  45.70  456–760 USD  189  21.65  761–1060 USD  73  8.36  Above 1061 USD  45  5.15  4. Educational attainment  Secondary school  122  13.97  Vocational school/high school  217  24.86  Professional training/college  146  16.72  University  289  33.10  Graduate school  99  11.34  5. City (region)  Xiangyang  147  16.84  Xuchang  154  17.64  Shenzhen  169  19.36  Daqing  142  16.27  Gucheng  147  16.84  Wuhan  114  13.06  Features  Frequency  Ratio (%)  1. Gender  Male  366  41.92  Female  507  58.08  2. Age  Under 19  30  3.44  20–29  341  39.06  30–39  198  22.68  40–49  173  19.82  Over 50  131  15.01  3. Monthly income  Under 150 USD  167  19.13  151–455 USD  399  45.70  456–760 USD  189  21.65  761–1060 USD  73  8.36  Above 1061 USD  45  5.15  4. Educational attainment  Secondary school  122  13.97  Vocational school/high school  217  24.86  Professional training/college  146  16.72  University  289  33.10  Graduate school  99  11.34  5. City (region)  Xiangyang  147  16.84  Xuchang  154  17.64  Shenzhen  169  19.36  Daqing  142  16.27  Gucheng  147  16.84  Wuhan  114  13.06  Source: Derived from the sample data by the authors using SPSS 20.0. Table 1. The distribution of samples. Features  Frequency  Ratio (%)  1. Gender  Male  366  41.92  Female  507  58.08  2. Age  Under 19  30  3.44  20–29  341  39.06  30–39  198  22.68  40–49  173  19.82  Over 50  131  15.01  3. Monthly income  Under 150 USD  167  19.13  151–455 USD  399  45.70  456–760 USD  189  21.65  761–1060 USD  73  8.36  Above 1061 USD  45  5.15  4. Educational attainment  Secondary school  122  13.97  Vocational school/high school  217  24.86  Professional training/college  146  16.72  University  289  33.10  Graduate school  99  11.34  5. City (region)  Xiangyang  147  16.84  Xuchang  154  17.64  Shenzhen  169  19.36  Daqing  142  16.27  Gucheng  147  16.84  Wuhan  114  13.06  Features  Frequency  Ratio (%)  1. Gender  Male  366  41.92  Female  507  58.08  2. Age  Under 19  30  3.44  20–29  341  39.06  30–39  198  22.68  40–49  173  19.82  Over 50  131  15.01  3. Monthly income  Under 150 USD  167  19.13  151–455 USD  399  45.70  456–760 USD  189  21.65  761–1060 USD  73  8.36  Above 1061 USD  45  5.15  4. Educational attainment  Secondary school  122  13.97  Vocational school/high school  217  24.86  Professional training/college  146  16.72  University  289  33.10  Graduate school  99  11.34  5. City (region)  Xiangyang  147  16.84  Xuchang  154  17.64  Shenzhen  169  19.36  Daqing  142  16.27  Gucheng  147  16.84  Wuhan  114  13.06  Source: Derived from the sample data by the authors using SPSS 20.0. TESTING RESULTS AND ANALYSES Firstly, we conducted Homogeneity tests for sample variances based on the monthly income level, educational background, age, region and gender data for these six cities in China. If the variances are homogeneous, we could use the Bonferroni test to analyze the samples; otherwise, we should use the Dunnett’s T3 test. Analysis by monthly income In order to facilitate the analysis, we regard the consumers with a monthly income of 1061 USD or above as high-income consumers, those ranging from 456 to 1060 USD as middle-income consumers, and those with <455 USD as low-income consumers. Analyzing monthly income differences reveals the following findings (Tables 2 and 3): The low-income consumers have the highest expectations for government subsidies on low-carbon products. The differences between high-income and low-income consumers are significant, so are the differences between middle-income and low-income consumers; but there is little difference between high-income and middle-income consumers. Table 2 displays that the difference of low-carbon subsidy expectations between high-income consumers (3.00) and low-income consumers (3.68) is large (22.67%), and that between middle-income consumers (3.24) and low-income consumers (3.68) is considerable (13.58%). However, the difference between high-income consumers (3.00) and middle-income consumers (3.24) is small (8% only) (Table 2). Currently, China’s average monthly income of urban residents is 420 USD in China, and the majority of them are still at low levels of income. Therefore, it still needs some more years to realize low-carbon consumption in China. What is more, because of the large income gap of Chinese residents, the high-income group should take the lead in low-carbon consumption, and the low-income group should get more low-carbon subsidies from the government. Consumers’ expectations for government subsidies on low-carbon products have a negative correlation with the monthly income. The average expectation of those consumers with a monthly income of 1061 USD or above is minimum (3.00), but that of the under 999 group is maximum (3.68), as shown in Table 2. Actually, government subsidies on low-carbon products can offset the additional costs incurred by the producer to produce low-carbon products. For the low-income consumers, the lower their income, the higher their expectations. In the case of high-income consumers, their expectations for subsidies are low because their living pressure is relatively low. The middle-income group has the moderate expectations. There are significant differences among different income levels on their expectations for government subsidies on low-carbon products. In order to verify the significance of the differences, firstly we made the Homogeneity tests for variances. From Table 2, we can observe that the Sig. value of the Homogeneity tests for variances is 0.065 (>0.05), which means that the sample has the homogeneity of variance. In addition, according to the result of variance analysis among groups in Table 2, the significance value is 0.000 (<0.05), showing that the differences among consumers with different incomes are significant. Table 2. Test results by monthly income and the ranking of expectation mean values. Monthly income    N  Mean value  Homogeneity tests for variances  Variance analysis  Significance  Significance (among groups)  Low-income consumers  Under 999 USD  167  3.68  0.065  0.000  151–455 USD  399  3.55  Middle-income consumers  5000–1060 USD  73  3.26  456–760 USD  189  3.24  High-income consumers  Above 1061 USD  45  3.00  Monthly income    N  Mean value  Homogeneity tests for variances  Variance analysis  Significance  Significance (among groups)  Low-income consumers  Under 999 USD  167  3.68  0.065  0.000  151–455 USD  399  3.55  Middle-income consumers  5000–1060 USD  73  3.26  456–760 USD  189  3.24  High-income consumers  Above 1061 USD  45  3.00  Note: The significance level of Homogeneity tests and Variance analysis is 0.05. Source: Derived from the sample data by the authors using SPSS 20.0. Table 2. Test results by monthly income and the ranking of expectation mean values. Monthly income    N  Mean value  Homogeneity tests for variances  Variance analysis  Significance  Significance (among groups)  Low-income consumers  Under 999 USD  167  3.68  0.065  0.000  151–455 USD  399  3.55  Middle-income consumers  5000–1060 USD  73  3.26  456–760 USD  189  3.24  High-income consumers  Above 1061 USD  45  3.00  Monthly income    N  Mean value  Homogeneity tests for variances  Variance analysis  Significance  Significance (among groups)  Low-income consumers  Under 999 USD  167  3.68  0.065  0.000  151–455 USD  399  3.55  Middle-income consumers  5000–1060 USD  73  3.26  456–760 USD  189  3.24  High-income consumers  Above 1061 USD  45  3.00  Note: The significance level of Homogeneity tests and Variance analysis is 0.05. Source: Derived from the sample data by the authors using SPSS 20.0. Table 3. Bonferroni test for different monthly income. (I) Monthly income  (J) Monthly income  Mean difference  Std. error  Significance  95% Confidence interval  (I–J)  Lower bound  Upper bound  Under 999 USD  151–455 USD  0.128  0.095  1.000  −0.14  0.40  456–760 USD  0.433*  0.110  0.001  0.12  0.74  5000–1060 USD  0.416*  0.145  0.041  0.01  0.82  Above 1061 USD  0.677*  0.173  0.001  0.19  1.16  151–455 USD  Under 999 USD  −0.128  0.095  1.000  −0.40  0.14  456–760 USD  0.305*  0.091  0.008  0.05  0.56  5000–1060 USD  0.289  0.131  0.283  −0.08  0.66  Above 1061 USD  0.549*  0.162  0.008  0.09  1.01  456–760 USD  Under 999 USD  −0.433*  0.110  0.001  −0.74  −0.12  151–455 USD  −0.305*  0.091  0.008  −0.56  −0.05  5000–1060 USD  −0.017  0.142  1.000  −0.42  0.38  Above 1061 USD  0.243  0.171  1.000  −0.24  0.73  5000–1060 USD  Under 999 USD  −0.416*  0.145  0.041  −0.82  −0.01  151–455 USD  −0.289  0.131  0.283  −0.66  0.08  456–760 USD  0.017  0.142  1.000  −0.38  0.42  Above 1061 USD  0.260  0.196  1.000  −0.29  0.81  Above 1061 USD  Under 999 USD  −0.677*  0.173  0.001  −1.16  −0.19  151–455 USD  −0.549*  0.162  0.008  −1.01  −0.09  456–760 USD  −0.243  0.171  1.000  −0.73  0.24  5000–1060 USD  −0.260  0.196  1.000  −0.81  0.29  (I) Monthly income  (J) Monthly income  Mean difference  Std. error  Significance  95% Confidence interval  (I–J)  Lower bound  Upper bound  Under 999 USD  151–455 USD  0.128  0.095  1.000  −0.14  0.40  456–760 USD  0.433*  0.110  0.001  0.12  0.74  5000–1060 USD  0.416*  0.145  0.041  0.01  0.82  Above 1061 USD  0.677*  0.173  0.001  0.19  1.16  151–455 USD  Under 999 USD  −0.128  0.095  1.000  −0.40  0.14  456–760 USD  0.305*  0.091  0.008  0.05  0.56  5000–1060 USD  0.289  0.131  0.283  −0.08  0.66  Above 1061 USD  0.549*  0.162  0.008  0.09  1.01  456–760 USD  Under 999 USD  −0.433*  0.110  0.001  −0.74  −0.12  151–455 USD  −0.305*  0.091  0.008  −0.56  −0.05  5000–1060 USD  −0.017  0.142  1.000  −0.42  0.38  Above 1061 USD  0.243  0.171  1.000  −0.24  0.73  5000–1060 USD  Under 999 USD  −0.416*  0.145  0.041  −0.82  −0.01  151–455 USD  −0.289  0.131  0.283  −0.66  0.08  456–760 USD  0.017  0.142  1.000  −0.38  0.42  Above 1061 USD  0.260  0.196  1.000  −0.29  0.81  Above 1061 USD  Under 999 USD  −0.677*  0.173  0.001  −1.16  −0.19  151–455 USD  −0.549*  0.162  0.008  −1.01  −0.09  456–760 USD  −0.243  0.171  1.000  −0.73  0.24  5000–1060 USD  −0.260  0.196  1.000  −0.81  0.29  *The mean difference is significant at the 0.05 level. Source: Derived from the sample data by the authors using SPSS 20.0. Table 3. Bonferroni test for different monthly income. (I) Monthly income  (J) Monthly income  Mean difference  Std. error  Significance  95% Confidence interval  (I–J)  Lower bound  Upper bound  Under 999 USD  151–455 USD  0.128  0.095  1.000  −0.14  0.40  456–760 USD  0.433*  0.110  0.001  0.12  0.74  5000–1060 USD  0.416*  0.145  0.041  0.01  0.82  Above 1061 USD  0.677*  0.173  0.001  0.19  1.16  151–455 USD  Under 999 USD  −0.128  0.095  1.000  −0.40  0.14  456–760 USD  0.305*  0.091  0.008  0.05  0.56  5000–1060 USD  0.289  0.131  0.283  −0.08  0.66  Above 1061 USD  0.549*  0.162  0.008  0.09  1.01  456–760 USD  Under 999 USD  −0.433*  0.110  0.001  −0.74  −0.12  151–455 USD  −0.305*  0.091  0.008  −0.56  −0.05  5000–1060 USD  −0.017  0.142  1.000  −0.42  0.38  Above 1061 USD  0.243  0.171  1.000  −0.24  0.73  5000–1060 USD  Under 999 USD  −0.416*  0.145  0.041  −0.82  −0.01  151–455 USD  −0.289  0.131  0.283  −0.66  0.08  456–760 USD  0.017  0.142  1.000  −0.38  0.42  Above 1061 USD  0.260  0.196  1.000  −0.29  0.81  Above 1061 USD  Under 999 USD  −0.677*  0.173  0.001  −1.16  −0.19  151–455 USD  −0.549*  0.162  0.008  −1.01  −0.09  456–760 USD  −0.243  0.171  1.000  −0.73  0.24  5000–1060 USD  −0.260  0.196  1.000  −0.81  0.29  (I) Monthly income  (J) Monthly income  Mean difference  Std. error  Significance  95% Confidence interval  (I–J)  Lower bound  Upper bound  Under 999 USD  151–455 USD  0.128  0.095  1.000  −0.14  0.40  456–760 USD  0.433*  0.110  0.001  0.12  0.74  5000–1060 USD  0.416*  0.145  0.041  0.01  0.82  Above 1061 USD  0.677*  0.173  0.001  0.19  1.16  151–455 USD  Under 999 USD  −0.128  0.095  1.000  −0.40  0.14  456–760 USD  0.305*  0.091  0.008  0.05  0.56  5000–1060 USD  0.289  0.131  0.283  −0.08  0.66  Above 1061 USD  0.549*  0.162  0.008  0.09  1.01  456–760 USD  Under 999 USD  −0.433*  0.110  0.001  −0.74  −0.12  151–455 USD  −0.305*  0.091  0.008  −0.56  −0.05  5000–1060 USD  −0.017  0.142  1.000  −0.42  0.38  Above 1061 USD  0.243  0.171  1.000  −0.24  0.73  5000–1060 USD  Under 999 USD  −0.416*  0.145  0.041  −0.82  −0.01  151–455 USD  −0.289  0.131  0.283  −0.66  0.08  456–760 USD  0.017  0.142  1.000  −0.38  0.42  Above 1061 USD  0.260  0.196  1.000  −0.29  0.81  Above 1061 USD  Under 999 USD  −0.677*  0.173  0.001  −1.16  −0.19  151–455 USD  −0.549*  0.162  0.008  −1.01  −0.09  456–760 USD  −0.243  0.171  1.000  −0.73  0.24  5000–1060 USD  −0.260  0.196  1.000  −0.81  0.29  *The mean difference is significant at the 0.05 level. Source: Derived from the sample data by the authors using SPSS 20.0. In order to figure out the differences of subsidy expectations among different types of consumers, we tested the expectation differences in this dimension by adopting the Bonferroni Test. Those consumers with a monthly income below 455 USD are significantly different from the 456–760 USD and 1061 or more USD groups, and have certain differences with other groups, but do not reach the significant level (Table 3). Analysis by educational attainment The expectations of consumers with different educational attainment are analyzed with one-way ANOVA as well, with the following findings. The mean values of the consumer expectations for government subsidies on low-carbon products by educational attainment are as follows in a descending order: vocational school/high school (3.54), university (3.53), professional training/college (3.52), secondary school (3.50) and graduate school (2.88). From Table 4, we can easily observe that the maximum mean value (vocational school/high school) is 22.92% higher than the minimum mean value (graduate school), which indicates a significant difference in the expectation by educational level. Moreover, the lower their educational attainment, the stronger their expectations for subsidies. This may be caused by the fact that the government subsidies can effectively reduce the price of low-carbon products, and consumers of different educational attainment have different income and life stress. Consumers with less education possess lower income and face greater pressure, so they are more willing to get subsidies on low-carbon products from the government. Table 4. Test results by educational background and the ranking of expectation mean values. Educational background  N  Mean value  Homogeneity tests for variances Significance  Variance analysis Significance (among groups)  Vocational school/high school  217  3.54  0.065  0.000  University  289  3.53  Professional training/college  146  3.52  Secondary school  122  3.50  Graduate school  99  2.88  Educational background  N  Mean value  Homogeneity tests for variances Significance  Variance analysis Significance (among groups)  Vocational school/high school  217  3.54  0.065  0.000  University  289  3.53  Professional training/college  146  3.52  Secondary school  122  3.50  Graduate school  99  2.88  Note: The significance level of Homogeneity tests and Variance analysis is 0.05. Source: Derived from the sample data by the authors using SPSS 20.0. Table 4. Test results by educational background and the ranking of expectation mean values. Educational background  N  Mean value  Homogeneity tests for variances Significance  Variance analysis Significance (among groups)  Vocational school/high school  217  3.54  0.065  0.000  University  289  3.53  Professional training/college  146  3.52  Secondary school  122  3.50  Graduate school  99  2.88  Educational background  N  Mean value  Homogeneity tests for variances Significance  Variance analysis Significance (among groups)  Vocational school/high school  217  3.54  0.065  0.000  University  289  3.53  Professional training/college  146  3.52  Secondary school  122  3.50  Graduate school  99  2.88  Note: The significance level of Homogeneity tests and Variance analysis is 0.05. Source: Derived from the sample data by the authors using SPSS 20.0. There are significant differences among consumers with different educational attainment on the expectations for government subsidies on low-carbon products. In order to verify the significance of the differences, firstly we made the Homogeneity tests for variances. Table 4 reports that the sample has the homogeneity of variance (Sig. 0.065 > 0.05), and the differences is significant according to the result of variance analysis (P < 0.05). To find out the differences of subsidy expectations among different consumers with different educational background, we tested the expectation differences in this dimension by adopting the Bonferroni test (results are presented in Table 5). The results indicate that the consumers with graduate school background are significantly different from other groups on subsidy expectation, but the differences among other groups are not significant. Table 5. Bonferroni Test for different educational background. (I) Educational background  (J) Educational background  Mean difference  Std. error  Significance  95% Confidence interval  (I−J)  Lower bound  Upper bound  Secondary school  Vocational school/high school  −0.044  0.116  1.000  −0.37  0.28  Professional training/college  −0.021  0.126  1.000  −0.38  0.33  University  −0.033  0.111  1.000  −0.35  0.28  Graduate school  0.621*  0.139  0.000  0.23  1.01  Vocational school/high school  Secondary school  0.044  0.116  1.000  −0.28  0.37  Professional training/college  0.023  0.110  1.000  −0.29  0.33  University  0.011  0.092  1.000  −0.25  0.27  Graduate school  0.665*  0.125  0.000  0.31  1.02  Professional training/college  Secondary school  0.021  0.126  1.000  −0.33  0.38  Vocational school/high school  −0.023  0.110  1.000  −0.33  0.29  University  −0.012  0.104  1.000  −0.31  0.28  Graduate school  0.642*  0.134  0.000  0.26  1.02  University  Secondary school  0.033  0.111  1.000  −0.28  0.35  Vocational school/high school  −0.011  0.092  1.000  −0.27  0.25  Professional training/college  0.012  0.104  1.000  −0.28  0.31  Graduate school  0.654*  0.120  0.000  0.32  0.99  Graduate school  Secondary school  −0.621*  0.139  0.000  −1.01  −0.23  Vocational school/high school  −0.665*  0.125  0.000  −1.02  −0.31  Professional training/college  −0.642*  0.134  0.000  −1.02  −0.26  University  −0.654*  0.120  0.000  −0.99  −0.32  (I) Educational background  (J) Educational background  Mean difference  Std. error  Significance  95% Confidence interval  (I−J)  Lower bound  Upper bound  Secondary school  Vocational school/high school  −0.044  0.116  1.000  −0.37  0.28  Professional training/college  −0.021  0.126  1.000  −0.38  0.33  University  −0.033  0.111  1.000  −0.35  0.28  Graduate school  0.621*  0.139  0.000  0.23  1.01  Vocational school/high school  Secondary school  0.044  0.116  1.000  −0.28  0.37  Professional training/college  0.023  0.110  1.000  −0.29  0.33  University  0.011  0.092  1.000  −0.25  0.27  Graduate school  0.665*  0.125  0.000  0.31  1.02  Professional training/college  Secondary school  0.021  0.126  1.000  −0.33  0.38  Vocational school/high school  −0.023  0.110  1.000  −0.33  0.29  University  −0.012  0.104  1.000  −0.31  0.28  Graduate school  0.642*  0.134  0.000  0.26  1.02  University  Secondary school  0.033  0.111  1.000  −0.28  0.35  Vocational school/high school  −0.011  0.092  1.000  −0.27  0.25  Professional training/college  0.012  0.104  1.000  −0.28  0.31  Graduate school  0.654*  0.120  0.000  0.32  0.99  Graduate school  Secondary school  −0.621*  0.139  0.000  −1.01  −0.23  Vocational school/high school  −0.665*  0.125  0.000  −1.02  −0.31  Professional training/college  −0.642*  0.134  0.000  −1.02  −0.26  University  −0.654*  0.120  0.000  −0.99  −0.32  *The mean difference is significant at the 0.05 level. Source: Derived from the sample data by the authors using SPSS 20.0. Table 5. Bonferroni Test for different educational background. (I) Educational background  (J) Educational background  Mean difference  Std. error  Significance  95% Confidence interval  (I−J)  Lower bound  Upper bound  Secondary school  Vocational school/high school  −0.044  0.116  1.000  −0.37  0.28  Professional training/college  −0.021  0.126  1.000  −0.38  0.33  University  −0.033  0.111  1.000  −0.35  0.28  Graduate school  0.621*  0.139  0.000  0.23  1.01  Vocational school/high school  Secondary school  0.044  0.116  1.000  −0.28  0.37  Professional training/college  0.023  0.110  1.000  −0.29  0.33  University  0.011  0.092  1.000  −0.25  0.27  Graduate school  0.665*  0.125  0.000  0.31  1.02  Professional training/college  Secondary school  0.021  0.126  1.000  −0.33  0.38  Vocational school/high school  −0.023  0.110  1.000  −0.33  0.29  University  −0.012  0.104  1.000  −0.31  0.28  Graduate school  0.642*  0.134  0.000  0.26  1.02  University  Secondary school  0.033  0.111  1.000  −0.28  0.35  Vocational school/high school  −0.011  0.092  1.000  −0.27  0.25  Professional training/college  0.012  0.104  1.000  −0.28  0.31  Graduate school  0.654*  0.120  0.000  0.32  0.99  Graduate school  Secondary school  −0.621*  0.139  0.000  −1.01  −0.23  Vocational school/high school  −0.665*  0.125  0.000  −1.02  −0.31  Professional training/college  −0.642*  0.134  0.000  −1.02  −0.26  University  −0.654*  0.120  0.000  −0.99  −0.32  (I) Educational background  (J) Educational background  Mean difference  Std. error  Significance  95% Confidence interval  (I−J)  Lower bound  Upper bound  Secondary school  Vocational school/high school  −0.044  0.116  1.000  −0.37  0.28  Professional training/college  −0.021  0.126  1.000  −0.38  0.33  University  −0.033  0.111  1.000  −0.35  0.28  Graduate school  0.621*  0.139  0.000  0.23  1.01  Vocational school/high school  Secondary school  0.044  0.116  1.000  −0.28  0.37  Professional training/college  0.023  0.110  1.000  −0.29  0.33  University  0.011  0.092  1.000  −0.25  0.27  Graduate school  0.665*  0.125  0.000  0.31  1.02  Professional training/college  Secondary school  0.021  0.126  1.000  −0.33  0.38  Vocational school/high school  −0.023  0.110  1.000  −0.33  0.29  University  −0.012  0.104  1.000  −0.31  0.28  Graduate school  0.642*  0.134  0.000  0.26  1.02  University  Secondary school  0.033  0.111  1.000  −0.28  0.35  Vocational school/high school  −0.011  0.092  1.000  −0.27  0.25  Professional training/college  0.012  0.104  1.000  −0.28  0.31  Graduate school  0.654*  0.120  0.000  0.32  0.99  Graduate school  Secondary school  −0.621*  0.139  0.000  −1.01  −0.23  Vocational school/high school  −0.665*  0.125  0.000  −1.02  −0.31  Professional training/college  −0.642*  0.134  0.000  −1.02  −0.26  University  −0.654*  0.120  0.000  −0.99  −0.32  *The mean difference is significant at the 0.05 level. Source: Derived from the sample data by the authors using SPSS 20.0. Analysis by age Then, we went on to test the expectation differences in the age dimension by adopting one-way ANOVA, with the following findings. Middle-aged consumers have the lowest expectations for government subsidies on low-carbon products, but the youth group has the highest. The maximum mean value (30–39 and 40–49 groups) is 9.76% higher than the minimum mean value (20–29 group), demonstrating a significant difference in the expectations by age (Table 6). Middle-aged consumers (30–39 and 40–49 groups) always have stable jobs and high income; therefore, they are not sensitive to the subsidies and have the lowest expectations. Additionally, consumers between 20 and 29 years old exhibit highest expectations for the government subsidies on low-carbon products, which may be due to the large income elasticity and great life pressure. There are significant differences among consumers with different age on the expectations for government subsidies on low-carbon products. After the Homogeneity tests for variances, the results show that the sample has the homogeneity of variance (Sig. 0.243 > 0.05), and the differences is significant according to the result of Variance analysis (P < 0.05). Table 6. Test results by age and the ranking of expectation mean values. Age  N  Mean value  Homogeneity tests for variances Significance  Variance analysis Significance (among groups)  20–29  341  3.60  0.243  0.001  Over 50  131  3.57  Under 19  30  3.47  30–39  198  3.28  40–49  173  3.28  Age  N  Mean value  Homogeneity tests for variances Significance  Variance analysis Significance (among groups)  20–29  341  3.60  0.243  0.001  Over 50  131  3.57  Under 19  30  3.47  30–39  198  3.28  40–49  173  3.28  Note: The significance level of Homogeneity tests and Variance analysis is 0.05. Source: Derived from the sample data by the authors using SPSS 20.0. Table 6. Test results by age and the ranking of expectation mean values. Age  N  Mean value  Homogeneity tests for variances Significance  Variance analysis Significance (among groups)  20–29  341  3.60  0.243  0.001  Over 50  131  3.57  Under 19  30  3.47  30–39  198  3.28  40–49  173  3.28  Age  N  Mean value  Homogeneity tests for variances Significance  Variance analysis Significance (among groups)  20–29  341  3.60  0.243  0.001  Over 50  131  3.57  Under 19  30  3.47  30–39  198  3.28  40–49  173  3.28  Note: The significance level of Homogeneity tests and Variance analysis is 0.05. Source: Derived from the sample data by the authors using SPSS 20.0. Then, by adopting the Bonferroni test, we tested the expectation differences in this dimension (results are shown in Table 7). The results indicate that the consumers in age 20–29 are significantly different from the 30–39 and 40–49 groups on subsidy expectations, but the differences among other groups are not significant. In addition, the mean values of consumer expectations for government subsidies on low-carbon products are presented in Z-shape ranked by age on an ascending order. Table 7. Bonferroni test for different ages. (I) Age  (J) Age  Mean difference (I−J)  Std. error  Significance  95% Confidence interval  Lower bound  Upper bound  Under 19  20–29  −0.135  0.198  1.000  −0.69  0.42  30–39  0.189  0.203  1.000  −0.38  0.76  40–49  0.189  0.205  1.000  −0.39  0.77  Over 50  −0.106  0.210  1.000  −0.70  0.49  20–29  Under 19  0.135  0.198  1.000  −0.42  0.69  30–39  0.323*  0.093  0.005  0.06  0.58  40–49  0.324*  0.097  0.009  0.05  0.60  Over 50  0.029  0.107  1.000  −0.27  0.33  30–39  Under 19  −0.189  0.203  1.000  −0.76  0.38  20–29  −0.323*  0.093  0.005  −0.58  −0.06  40–49  0.000  0.108  1.000  −0.30  0.30  Over 50  −0.295  0.117  0.119  −0.62  0.03  40–49  Under 19  −0.189  0.205  1.000  −0.77  0.39  20–29  −0.324*  0.097  0.009  −0.60  −0.05  30–39  0.000  0.108  1.000  −0.30  0.30  Over 50  −0.295  0.120  0.143  −0.63  0.04  Over 50  Under 19  0.106  0.210  1.000  −0.49  0.70  20–29  −0.029  0.107  1.000  −0.33  0.27  30–39  0.295  0.117  0.119  −0.03  0.62  40–49  0.295  0.120  0.143  −0.04  0.63  (I) Age  (J) Age  Mean difference (I−J)  Std. error  Significance  95% Confidence interval  Lower bound  Upper bound  Under 19  20–29  −0.135  0.198  1.000  −0.69  0.42  30–39  0.189  0.203  1.000  −0.38  0.76  40–49  0.189  0.205  1.000  −0.39  0.77  Over 50  −0.106  0.210  1.000  −0.70  0.49  20–29  Under 19  0.135  0.198  1.000  −0.42  0.69  30–39  0.323*  0.093  0.005  0.06  0.58  40–49  0.324*  0.097  0.009  0.05  0.60  Over 50  0.029  0.107  1.000  −0.27  0.33  30–39  Under 19  −0.189  0.203  1.000  −0.76  0.38  20–29  −0.323*  0.093  0.005  −0.58  −0.06  40–49  0.000  0.108  1.000  −0.30  0.30  Over 50  −0.295  0.117  0.119  −0.62  0.03  40–49  Under 19  −0.189  0.205  1.000  −0.77  0.39  20–29  −0.324*  0.097  0.009  −0.60  −0.05  30–39  0.000  0.108  1.000  −0.30  0.30  Over 50  −0.295  0.120  0.143  −0.63  0.04  Over 50  Under 19  0.106  0.210  1.000  −0.49  0.70  20–29  −0.029  0.107  1.000  −0.33  0.27  30–39  0.295  0.117  0.119  −0.03  0.62  40–49  0.295  0.120  0.143  −0.04  0.63  *The mean difference is significant at the 0.05 level. Source: Derived from the sample data by the authors using SPSS 20.0. Table 7. Bonferroni test for different ages. (I) Age  (J) Age  Mean difference (I−J)  Std. error  Significance  95% Confidence interval  Lower bound  Upper bound  Under 19  20–29  −0.135  0.198  1.000  −0.69  0.42  30–39  0.189  0.203  1.000  −0.38  0.76  40–49  0.189  0.205  1.000  −0.39  0.77  Over 50  −0.106  0.210  1.000  −0.70  0.49  20–29  Under 19  0.135  0.198  1.000  −0.42  0.69  30–39  0.323*  0.093  0.005  0.06  0.58  40–49  0.324*  0.097  0.009  0.05  0.60  Over 50  0.029  0.107  1.000  −0.27  0.33  30–39  Under 19  −0.189  0.203  1.000  −0.76  0.38  20–29  −0.323*  0.093  0.005  −0.58  −0.06  40–49  0.000  0.108  1.000  −0.30  0.30  Over 50  −0.295  0.117  0.119  −0.62  0.03  40–49  Under 19  −0.189  0.205  1.000  −0.77  0.39  20–29  −0.324*  0.097  0.009  −0.60  −0.05  30–39  0.000  0.108  1.000  −0.30  0.30  Over 50  −0.295  0.120  0.143  −0.63  0.04  Over 50  Under 19  0.106  0.210  1.000  −0.49  0.70  20–29  −0.029  0.107  1.000  −0.33  0.27  30–39  0.295  0.117  0.119  −0.03  0.62  40–49  0.295  0.120  0.143  −0.04  0.63  (I) Age  (J) Age  Mean difference (I−J)  Std. error  Significance  95% Confidence interval  Lower bound  Upper bound  Under 19  20–29  −0.135  0.198  1.000  −0.69  0.42  30–39  0.189  0.203  1.000  −0.38  0.76  40–49  0.189  0.205  1.000  −0.39  0.77  Over 50  −0.106  0.210  1.000  −0.70  0.49  20–29  Under 19  0.135  0.198  1.000  −0.42  0.69  30–39  0.323*  0.093  0.005  0.06  0.58  40–49  0.324*  0.097  0.009  0.05  0.60  Over 50  0.029  0.107  1.000  −0.27  0.33  30–39  Under 19  −0.189  0.203  1.000  −0.76  0.38  20–29  −0.323*  0.093  0.005  −0.58  −0.06  40–49  0.000  0.108  1.000  −0.30  0.30  Over 50  −0.295  0.117  0.119  −0.62  0.03  40–49  Under 19  −0.189  0.205  1.000  −0.77  0.39  20–29  −0.324*  0.097  0.009  −0.60  −0.05  30–39  0.000  0.108  1.000  −0.30  0.30  Over 50  −0.295  0.120  0.143  −0.63  0.04  Over 50  Under 19  0.106  0.210  1.000  −0.49  0.70  20–29  −0.029  0.107  1.000  −0.33  0.27  30–39  0.295  0.117  0.119  −0.03  0.62  40–49  0.295  0.120  0.143  −0.04  0.63  *The mean difference is significant at the 0.05 level. Source: Derived from the sample data by the authors using SPSS 20.0. Analysis by region Based on the field survey in six cities (regions), we analyzed the expectation differences in the “Region” dimension by using the methodology of one-way ANOVA and obtained the following results. The expectations for government subsidies on low-carbon products in different regions are closely related to the level of economic development. The mean values of expectations for government subsidies on low-carbon products of different regions are ranked as follows based on a descending order: Xiangyang (3.59), Xuchang (3.53), Shenzhen (3.49), Daqing (3.41), Gucheng (3.37) and Wuhan (3.30), and the maximum mean value is 8.79% higher than the minimum value (Table 8). The differences of expectations for government subsidies on low-carbon products of different regions do not reach a significant level. The results of Homogeneity tests for variances show that it is non-homogeneity of variance (Sig. 0.022 < 0.05, see Table 8). The results of one-way ANOVA indicate that the differences are not significant in different cities (P > 0.05, see Table 8). Then, we conducted Dunnett’s T3 test to further analyze the differences. The results show that there is no significant difference among the six cities (regions), which is consistent with the previous results. Table 8. Test results by region and the ranking of expectation mean values. Region  N  Mean value  Homogeneity tests for variances Significance  Variance analysis Significance (among groups)  Xiangyang  147  3.59  0.022  0.216  Xuchang  154  3.53  Shenzhen  169  3.49  Daqing  142  3.41  Gucheng  147  3.37  Wuhan  114  3.30  Region  N  Mean value  Homogeneity tests for variances Significance  Variance analysis Significance (among groups)  Xiangyang  147  3.59  0.022  0.216  Xuchang  154  3.53  Shenzhen  169  3.49  Daqing  142  3.41  Gucheng  147  3.37  Wuhan  114  3.30  Note: The significance level of Homogeneity tests and variance analysis is 0.05. Source: Derived from the sample data by the authors using SPSS 20.0. Table 8. Test results by region and the ranking of expectation mean values. Region  N  Mean value  Homogeneity tests for variances Significance  Variance analysis Significance (among groups)  Xiangyang  147  3.59  0.022  0.216  Xuchang  154  3.53  Shenzhen  169  3.49  Daqing  142  3.41  Gucheng  147  3.37  Wuhan  114  3.30  Region  N  Mean value  Homogeneity tests for variances Significance  Variance analysis Significance (among groups)  Xiangyang  147  3.59  0.022  0.216  Xuchang  154  3.53  Shenzhen  169  3.49  Daqing  142  3.41  Gucheng  147  3.37  Wuhan  114  3.30  Note: The significance level of Homogeneity tests and variance analysis is 0.05. Source: Derived from the sample data by the authors using SPSS 20.0. Analysis by gender We made the one-way ANOVA and independent samples T test on different genders and found that: The mean value of men’s expectations is a little higher than that of women. Because the number of groups is smaller than 3, we cannot conduct the Homogeneity tests for variances. According to the results of descriptive statistical analysis in Table 9, we can find that the mean value of men’s expectations for government subsidies on low-carbon products is 3.47, which is higher than that of women (3.45), but the disparity is quite small (by just 0.58%). The expectation differences between men and women are not significant. In the result of one-way ANOVA and independent samples T test, the Sig. value is 0.765 (0.765 > 0.05, see Tables 9 and 10). Therefore, we can conclude that people of different genders have pretty much the same expectations for government subsidies on low-carbon products. Table 9. Test results by gender and the ranking of expectation mean values. Gender  N  Mean value  Variance analysis  Significance (among groups)  Male  366  3.47  0.765  Female  507  3.45  Sum  873  3.45  Gender  N  Mean value  Variance analysis  Significance (among groups)  Male  366  3.47  0.765  Female  507  3.45  Sum  873  3.45  Note: The significance level of variance analysis is 0.05. Source: Derived from the sample data by the authors using SPSS 20.0. Table 9. Test results by gender and the ranking of expectation mean values. Gender  N  Mean value  Variance analysis  Significance (among groups)  Male  366  3.47  0.765  Female  507  3.45  Sum  873  3.45  Gender  N  Mean value  Variance analysis  Significance (among groups)  Male  366  3.47  0.765  Female  507  3.45  Sum  873  3.45  Note: The significance level of variance analysis is 0.05. Source: Derived from the sample data by the authors using SPSS 20.0. Table 10. Independent-samples T test for different genders.   Levene’s test for equality of variances  T-test for equality of means  F  Significance  t  df  Significance (2-tailed)  Mean difference  Std. error difference  95% Confidence interval of the difference  Lower  Upper  Expectation for government subsidies  Equal variances assumed  0.941  0.332  0.298  871  0.765  0.021  0.072  −0.120  0.163  Equal variances not assumed      0.297  772.150  0.767  0.021  0.072  −0.120  0.163    Levene’s test for equality of variances  T-test for equality of means  F  Significance  t  df  Significance (2-tailed)  Mean difference  Std. error difference  95% Confidence interval of the difference  Lower  Upper  Expectation for government subsidies  Equal variances assumed  0.941  0.332  0.298  871  0.765  0.021  0.072  −0.120  0.163  Equal variances not assumed      0.297  772.150  0.767  0.021  0.072  −0.120  0.163  Source: Derived from the sample data by the authors using SPSS 20.0. Table 10. Independent-samples T test for different genders.   Levene’s test for equality of variances  T-test for equality of means  F  Significance  t  df  Significance (2-tailed)  Mean difference  Std. error difference  95% Confidence interval of the difference  Lower  Upper  Expectation for government subsidies  Equal variances assumed  0.941  0.332  0.298  871  0.765  0.021  0.072  −0.120  0.163  Equal variances not assumed      0.297  772.150  0.767  0.021  0.072  −0.120  0.163    Levene’s test for equality of variances  T-test for equality of means  F  Significance  t  df  Significance (2-tailed)  Mean difference  Std. error difference  95% Confidence interval of the difference  Lower  Upper  Expectation for government subsidies  Equal variances assumed  0.941  0.332  0.298  871  0.765  0.021  0.072  −0.120  0.163  Equal variances not assumed      0.297  772.150  0.767  0.021  0.072  −0.120  0.163  Source: Derived from the sample data by the authors using SPSS 20.0. REGRESSION ANALYSIS OF CONSUMERS’ EXPECTATION FACTORS In order to figure out the city (region) and population factors on people’s expectations for government subsidies on low-carbon products and further define their degrees and significance of effect, we conducted a linear regression and set the ‘expectations for government subsidies on low-carbon products’ as the dependent variable, taking different cities (regions) and population factors as independent variables. According to the regression coefficients and significance level, we can find that educational attainment and monthly income of consumers have important effects on their expectations for government subsidies on low-carbon products. The regression coefficients of educational attainment and monthly income are significantly associated with the consumers’ subsidy expectations. Monthly income exerts the greatest influence (−0.167) on consumers’ expectations, which is significant at the 1% level and the coefficient is negative. Educational attainment is significant at the 10% level and the coefficient (−0.062) is negative as well. The results in Table 11 indicate that the lower their monthly income and educational attainment, the stronger their expectations. The regression results also further verify the conclusions that consumer expectations for government subsidies on low-carbon products has a negative correlation with their educational attainment and monthly income as stated earlier. Table 11. The regression results. Model  Unstandardized coefficients  Standardized coefficients  t  Significance  B  Std. error  Beta  (Constant)  4.160  0.220    18.941  0.000  Region  0.026  0.020  0.044  1.293  0.196  Age  −0.024  0.032  −0.027  −0.753  0.452  Gender  −0.111  0.073  −0.052  −1.524  0.128  Educational background  −0.051  0.031  −0.062  −1.652  0.099*  Monthly income  −0.168  0.038  −0.167  −4.404  0.000***  Model  Unstandardized coefficients  Standardized coefficients  t  Significance  B  Std. error  Beta  (Constant)  4.160  0.220    18.941  0.000  Region  0.026  0.020  0.044  1.293  0.196  Age  −0.024  0.032  −0.027  −0.753  0.452  Gender  −0.111  0.073  −0.052  −1.524  0.128  Educational background  −0.051  0.031  −0.062  −1.652  0.099*  Monthly income  −0.168  0.038  −0.167  −4.404  0.000***  *The mean difference is significant at the 0.1 level, **means 0.05, ***means 0.01. Source: Derived from the sample data by the authors using SPSS 20.0. Table 11. The regression results. Model  Unstandardized coefficients  Standardized coefficients  t  Significance  B  Std. error  Beta  (Constant)  4.160  0.220    18.941  0.000  Region  0.026  0.020  0.044  1.293  0.196  Age  −0.024  0.032  −0.027  −0.753  0.452  Gender  −0.111  0.073  −0.052  −1.524  0.128  Educational background  −0.051  0.031  −0.062  −1.652  0.099*  Monthly income  −0.168  0.038  −0.167  −4.404  0.000***  Model  Unstandardized coefficients  Standardized coefficients  t  Significance  B  Std. error  Beta  (Constant)  4.160  0.220    18.941  0.000  Region  0.026  0.020  0.044  1.293  0.196  Age  −0.024  0.032  −0.027  −0.753  0.452  Gender  −0.111  0.073  −0.052  −1.524  0.128  Educational background  −0.051  0.031  −0.062  −1.652  0.099*  Monthly income  −0.168  0.038  −0.167  −4.404  0.000***  *The mean difference is significant at the 0.1 level, **means 0.05, ***means 0.01. Source: Derived from the sample data by the authors using SPSS 20.0. CONCLUSIONS AND RECOMMENDATIONS Conclusions Based on the above statistical analyses, we can draw the following conclusions: Significant differences exist in the expectations for government subsidies on low-carbon products from different types of consumers. The maximum mean values of expectations are 22.92, 22.67, 9.76, 8.79 and 0.58% higher than the minimum values for educational attainment, monthly income, age, region and gender, respectively. There are significant differences among consumers with different monthly income, educational attainment and age on the expectations for government subsidies on low-carbon products, but the difference is not significant for the consumers in different regions or genders. Monthly income and educational attainment exert significant influences on consumer expectations for government subsidies on low-carbon products, and the influence of monthly income is the biggest. The results of one-way ANOVA and regression both indicate that there is a negative relationship between monthly income and subsidy expectations, and similarly that between educational attainment and subsidy expectations. That means, ceteris paribus, the lower their monthly income, the stronger their expectations for government subsidies on low-carbon products, so is their educational attainment. Therefore, for the current average income and education levels of the Chinese residents, it is not a mature time for an overall implementation of low-carbon consumption. Policy recommendations From the above conclusions, we propose the following policy recommendations. Fully play the demonstrative role of specific groups of consumers on low-carbon consumption. Take middle-aged female consumers that have high income and good education in the developed areas as the breakthrough, let them take the lead in implementing the strategy of low-carbon consumption throughout the country. According to the differences in purchase behavior of different types of consumers, the government should develop different strategies to promote low-carbon products, make the consumers with robust purchasing power to drive the consumption demand of low-carbon products. Develop the economy, strengthen education and improve consumers’ income. The prices of low-carbon products are always higher than conventional products. If the consumers do not have a fair income, they could not afford or are not willing to pay for low-carbon products. Educated consumers are often more aware of the significance of low-carbon consumption and environmental protection. Therefore, the subsidy policy is often the short-term tactic, but the development of the economy and education can essentially provide continuous power for low-carbon consumption in the long run. ACKNOWLEDGEMENTS The study is supported by the National Natural Science Foundation of China (NSFC) (Nos. 71473231, 71773119 and 71173201) ‘Research on the Impact Mechanism of Carbon Tariff and Carbon Labeling on Agri-trade and Carbon Reduction’. REFERENCES 1 Lewis JI. The evolving role of carbon finance in promoting renewable energy development in China. Energy Policy  2010; 38: 2875– 86. Google Scholar CrossRef Search ADS   2 Dong L, Fujita T, Zhang H et al.  . Promoting low-carbon city through industrial symbiosis: a case in China by applying HPIMO model. Energy policy  2013; 61: 864– 73. Google Scholar CrossRef Search ADS   3 Du S, Tang W, Song M. Low-carbon production with low-carbon premium in cap-and-trade regulation. J Cleaner Prod  2016; 134: 652– 62. Google Scholar CrossRef Search ADS   4 Gupta M. Willingness to pay for carbon tax: a study of Indian road passenger transport. Transp Policy  2016; 45: 46– 54. Google Scholar CrossRef Search ADS   5 Bigerna S, Bollino CA, Micheli S et al.  . Revealed and stated preferences for CO2 emissions reduction: the missing link. Renew Sustain Energy Rev  2017; 68: 1213– 21. Google Scholar CrossRef Search ADS   6 Newbery DM. Towards a green energy economy? The EU Energy Union’s transition to a low-carbon zero subsidy electricity system–lessons from the UK’s Electricity Market Reform. Appl Energy  2016; 179: 1321– 30. Google Scholar CrossRef Search ADS   7 Li Y, Fan J, Zhao D et al.  . Tiered gasoline pricing: a personal carbon trading perspective. Energy Policy  2016; 89: 194– 201. Google Scholar CrossRef Search ADS   8 Bunn DW, Muñoz JI. Supporting the externality of intermittency in policies for renewable energy. Energy Policy  2016; 88: 594– 602. Google Scholar CrossRef Search ADS   9 Sun J, Xiao Z, Zhou G. Two stage supply chain enterprises’ production and reduction decision-making mechanism research considering emission trading. J Ind Intell Inf  2016; 4: 46– 50. 10 Campiglio E. Beyond carbon pricing: the role of banking and monetary policy in financing the transition to a low-carbon economy. Ecol Econ  2016; 121: 220– 30. Google Scholar CrossRef Search ADS   11 Alonso PM, Hewitt R, Pacheco JD et al.  . Losing the roadmap: renewable energy paralysis in Spain and its implications for the EU low carbon economy. Renew Energy  2016; 89: 680– 94. Google Scholar CrossRef Search ADS   12 Liu W, Qin B. Low-carbon city initiatives in China: a review from the policy paradigm perspective. Cities  2016; 51: 131– 8. Google Scholar CrossRef Search ADS   13 Ling Y, Xu J, 2016. Low Carbon Consumption Preference and the Related Enterprises’ Optimal Strategies. Available at SSRN 2745693. Available at SSRN: http://ssrn.com/abstract=2745693. 14 He Y, Xu Y, Pang Y et al.  . A regulatory policy to promote renewable energy consumption in China: review and future evolutionary path. Renewable Energy  2016; 89: 695– 705. Google Scholar CrossRef Search ADS   15 Huang B, Mauerhofer V, Geng Y. Analysis of existing building energy saving policies in Japan and China. J Cleaner Prod  2016; 112: 1510– 8. Google Scholar CrossRef Search ADS   16 Sun Y. Research on the subsidy policy for the development of low-carbon economy. Public Finance Res  2010; 4: 59– 60. 17 Shen M, He Z. The experience of foreign fiscal policy on low-carbon economy. Ecol Econ  2011; 3: 83– 9. 18 Bollino CA. The willingness to pay for renewable energy sources: the case of Italy with socio-demographic determinants. Energy J  2009; 30: 81– 96. 19 Shukla PR, Chaturvedi V. Low carbon and clean energy scenarios for India: analysis of targets approach. Energy Econ  2012; 34: S487– 95. Google Scholar CrossRef Search ADS   20 Jin M, Zheng S. Analysis of the game behavior in green agricultural products market of China. Finance Trade Econ  2006; 6: 38– 41. 21 Roe B, Teisl MF, Levy A et al.  . US consumers’ willingness to pay for green electricity. Energy policy  2001; 29: 917– 25. Google Scholar CrossRef Search ADS   22 Zarnikau J. Consumer demand for ‘green power’ and energy efficiency. Energy Policy  2003; 31: 1661– 72. Google Scholar CrossRef Search ADS   23 Galinato GI, Yoder JK. An integrated tax-subsidy policy for carbon emission reduction. Resour Energy Econ  2010; 32: 310– 26. Google Scholar CrossRef Search ADS   24 Bajona C, Kelly DL. Trade and the environment with pre-existing subsidies: a dynamic general equilibrium analysis. J Environ Econ Manage  2012; 64: 253– 78. Google Scholar CrossRef Search ADS   25 Lin B, Li A. Impacts of removing fossil fuel subsidies on China: how large and how to mitigate? Energy  2012; 44: 741– 9. Google Scholar CrossRef Search ADS   26 Lapan H, Moschini G. Second-best biofuel policies and the welfare effects of quantity mandates and subsidies. J Environ Econ Manage  2012; 63: 224– 41. Google Scholar CrossRef Search ADS   27 Song L, Hua B. A comparative analysis of the fiscal policy system for different countries to develop low carbon economy. J Yunnan Univ Finance Econ  2011; 27: 98– 105. 28 He S, Wang F. Cost report model based on the mechanism of low-carbon subsidies. Energy Procedia  2011; 5: 1869– 73. Google Scholar CrossRef Search ADS   29 Yang J. The Game analysis of government subsidies in low carbon economy. Commer Res  2010; 8: 109– 12. 30 Brzeskot M, Haupt A. Environmental policy and the energy efficiency of vertically differentiated consumer products. Energy Econ  2013; 36: 444– 53. Google Scholar CrossRef Search ADS   31 Xiong D. Research on fiscal policies guiding low-carbon consumption of urban residents. J Finance Econ  2012; 2: 72– 5. 32 Bansal S, Gangopadhyay S. Tax/subsidy policies in the presence of environmentally aware consumers. J Environ Econ Manage  2003; 45: 333– 55. Google Scholar CrossRef Search ADS   33 Zhou L, Zheng X. Green subsidy policy effect assessment based on farmers’ willingness to pay the low-carbon elements: an empirical study in pig industry. J Nanjing Agric Univ  2012; 12: 85– 91. 34 Li Y, Zhao D. Research on R&D cost allocation comparison for low-carbon supply chain based on government’ subsidies. Soft Sci  2014; 28: 21– 6. 35 Xie S, Kuang Y, Huang N. Main paths and policy proposals for the development of carbon sinking agriculture in China. China Popul Resour Environ  2010; 20: 46– 51. 36 Fan R, Dong L, Yang W et al.  . Study on the optimal supervision strategy of government low-carbon subsidy and the corresponding efficiency and stability in the small-world network context. J Cleaner Prod  2017; 168. DOI:10.1016/j.jclepro.2017.09.044. 37 Li Q, Long R, Chen H. Empirical study of the willingness of consumers to purchase low-carbon products by considering carbon labels: a case study. J Cleaner Prod  2017. DOI: 10.1016/j.jclepro. 2017.04.154. 38 Foxall GR, Goldsmith RE, Brown S. Consumer Psychology for Marketing ; Vol. 1. Cengage Learning EMEA, Oxford, UK, 1994. © The Author(s) 2018. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com

Journal

International Journal of Low-Carbon TechnologiesOxford University Press

Published: Feb 21, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off