Improving potential economic growing rate of china in the new normal based on artificial intelligence with fuzzy integration

Abstract

As we know, AI is deeply integrated with all walks of life. While upgrading traditional products and transforming traditional industries, a large number of new products, new industries and new formats have emerged in large numbers, and new industrial space has been further expanded. The potential for steady economic growth remains huge. However, the influence of merelygrowing capital asset and labor force dimension in terms of boosting economic progress has been waning globally. Considering the China’s data panel data outlying areas during 1978–2017 as the research sample, possible economic growing rate of Chinahas been calculated in detail according to the labor-intensive structural time-varying elastic model. It shows a U-shaped development track during 2018–2027, with an average growth rate between 5.00%and 6.00%. There are also great variances in the probable economy growing rates in different regions in China. It shows that relying solely on traditional factors of production to drive potential economic growth, China has been unable to maintain the prosperity of stable development over the past few decades. Based on the experience of typical representative countries, this paper also puts forward some relevant countermeasures and suggestions including focusing on a new round of technological revolutions such as big data, new generation AI, 5 G, IOT, cloud computing, robotics, and blockchain to progress economical growing rate of China in the future.

Keywords

Economic growth potential output AI china fuzzy rule artificial neural network (ANN)

1. Introduction

Entering the new normal period, economy of China now is in a acuteage from high-speed development to high-quality progress. At this development stage, China’s economic growth is mainly reflected in the shift from low efficiency and high-cost growth mode driven by factors to medium and high-speed growth driven by innovation and consumption upgrading. China’s transformation and start-up has come into the subterranean aquatic area and the critical period from now on. The growth model driven solely by factors is increasingly unsustainable, which requires us to change the mode of economic growth, further stimulate its power, fully release its potential and drive it by innovation.

The new normal is a necessary stage for China’s economic development to a higher level. At present, China is in a critical period of in-depth development of industrialization and urbanization. The economic growth has entered a period of shift adjustment, and the trend of structural slowdown is very obvious. The new normal of the economy is not only the direct embodiment of the continuous deepening of China’s industrialization process but also the unavoidableconsequence of the revolution of the connotation and intensity of the late development advantage and the mode of technological progress. Its essence is the continuous evolution and upgrading of China’s economy to a higher form, a more complex division of labor and a more rational structure.

The new normal is an important basis for formulating China’s economic development strategies and policies. The connotation and conditions of China’s primary stage of socialism have developed profound changes. New industries, new models and new products are emerging in an endless stream, and new technologies, new formats and new demands are emerging as well. However, it means that it will not only have new opportunities but also face new challenges. So, it requires us to have new understandings, new ideas and new methods to deal with it. The new normal is not only the high generalization on the characteristics of the current stage of China’s economic development but also the regular understanding of China’s economic transformation and upgrading. It is also an important basis for formulating China’s economic development strategies and policies at present and in the future.

In the study of potential output, more scholars have done some relevant research [3 , 22]; In recent years, more experts, scholars, and related academic institutions have also conducted relevant research on China’s medium and long-term potential economic growth rate. The representative ones include [23], Research Group of Macroeconomic Research Institute of China National Development and Reform Commission (2008), [7], Fang and Yang (2013), [1 , 25], and so on. The relevant research is based on different methods and data sources and predicts China’s potential economic growth rate during different periods. Because the methods and data sources used are not the same, the predictions of relevant research on China’s potential economic growth rate are also inconsistent, and most of them are based on a macro perspective. In addition, [10 , 19], also did related research on the relationship between AI and economic growth, but the research topics and contents are relatively scattered.

The afore-mentioned methods and relevant conclusions of those researches provide important ideas and inspiration for the smooth development of this paper, but there are still some shortcomings exiting as follows: a) the research perspective is too macro, only limited to China’s physical capital, potential employment, human capital and other factors. It ignores the actual development of different regions and provinces in China, especially the differences in industrial structure, population structure and education level. So, inevitably there is a problem that “explaining all the problems with one question” in the analysis and prediction of China’s potential economic growth rate in the future; b) the prediction period is too long. In the specific prediction of China’s potential economic growth rate, the relevant research is mainly based on the models and parameters obtained from historical data, but the background conditions of the simulation are mostly based on the stage development characteristics of China’s recent economy. In prediction examining ability Ang et al. [1] surveyed estimation ability of supplementary forecasting method of families including non-linear approach, hybridization of various forecasting and prediction techniques. It is difficult to accurately estimate and measure the uncertain factors such as fertility policy, employment policy, investment policy and trade policy and so on, so it is inevitable to fall into the misunderstanding of “covering the long term with the short term” when analyzing and predicting potential economical growing rate of China in the research; c) related research mainly focuses on the relationship between the potential economic growth rate and the actual economic growth rate, while less attention is paid to the growth factors of the potential economic growth rate, especially in AI. From recent research response for economic growth and financial prediction and forecasting Fuzzy integrated neural network gain so much popularity and attention [13]. Furthermore, Adaptive Neuro-Fuzzy Inference System (ANFIS) play vital role to in predicting econometric boundaries in the form of non-linear connotations, thus ANFIS can considered as e-cient approach for analyzing non-linear economic behavior [4].

2. Estimation of potential economical growing rate of China in the new normal

As the Cobb-Douglas production function method is relatively simple and easy to obtain the data related to the model, it is widely used in the research of specific calculation of potential economic growth rate. Based on the relevant research method which was presented by [20] and considering China’s development practice, this paper further expands the Cobb-Douglas production function and establishes the labor-intensive structural time-varying elastic model as follows: $Y = A^{λ} K^{\sum_{i = 1}^{n} α_{it} k_{it}} L^{\sum_{i = 1}^{n} β_{it} l_{it}} H^{\sum_{j = 1}^{m} γ_{jt} h_{jt}} e^{u}$ (1)

Where Y refers to the actual output of different provinces in China. A refers to the technical level of different provinces in China, K is the capital stock of different provinces in China, L refers to the labor input of different provinces in China, that is the number of employees, and H refers to the human capital stock of different provinces in China. By taking the logarithm of variables on both sides of the equation respectively, the labor-intensive structural time-varying elastic can be obtained as follows:

$\begin{matrix} ln Y = & λ ln A + \sum_{i = 1}^{n} α_{it} k_{it} lnK + \sum_{i = 1}^{n} β_{it} l_{it} lnL \\ + \sum_{j = 1}^{m} γ_{jt} h_{jt} lnH + μ \end{matrix}$ (2)

Where α _it and k _it represent respectively the elasticity of the capital-output and the proportion of the capital stock by sectors (i = 1, 2, 3) in a given year. β _it and l _it represent respectively the labor output elasticity and labor proportion by sectors (i = 1, 2, 3) in a given year. γ _jt and h _jt represent respectively the output elasticity and the proportion of the human capital-output by types of human capital (j = 1, 2, 3, 4, 5) in a given year. λ represents the elasticity of the technological output of different provinces in China.

2.1 Model selection and analysis of regression results

By examining the spatial autoregression coefficient, spatial error coefficient and their significance, and combining the symbols and significance of the correlation coefficient in the regression results, this paper uses the regression results of the modified two-way fixed effect model (FGLS), the individual fixed effect spatial error model (SEM) and the generalized spatial panel random-effects model (GSPRE) to analyze and estimate China’s potential economic growth rate (see Table 1).

Table 1
Model specific regression results

Dependent vaiable: lnY FGLS model SEM model GSPRE model

lnA 0.00385^* (6.41) 0.0262^* (7.72) 0.0268^*** (7.91)

k₁lnK 0.463^* (42.88) 0.516^* (24.52) 0.507^*** (24.11)

k₂lnK 0.613^* (74.48) 0.528^* (44.36) 0.523^*** (44.24)

k₃lnK 0.596^* (76.91) 0.470^* (39.61) 0.466^*** (39.55)

l₁lnL 0.253^* (31.39) 0.391^* (8.93) 0.414^*** (13.01)

l₂lnL 0.390^* (37.55) 0.445^* (9.73) 0.474^*** (13.80)

l₃lnL 0.357^* (36.56) 0.465^* (10.52) 0.489^*** (13.88)

h₁lnH 0.0698^* (4.09) 0.825^* (9.60) 0.820^*** (9.55)

h₂lnH 0.0850^* (6.19) 0.375^* (3.66) 0.397^*** (3.86)

h₃lnH 0.0752^* (7.78) 0.656^* (10.98) 0.644^*** (10.70)

h₄lnH –0.0221^** (–2.45) 0.0235 (0.60) 0.0172 (0.44)

h₅lnH –0.0291^* (–3.14) 0.103^* (2.77) 0.103^*** (2.75)

_cons –0.202^* (–4.56) –1.394^*

λ 0.529^* (15.13) 0.534^* (15.40)

φ 0.256 (1.09)

Dependent vaiable: lnY	FGLS model	SEM model	GSPRE model
lnA	0.00385^*** (6.41)	0.0262^*** (7.72)	0.0268^*** (7.91)
k₁lnK	0.463^*** (42.88)	0.516^*** (24.52)	0.507^*** (24.11)
k₂lnK	0.613^*** (74.48)	0.528^*** (44.36)	0.523^*** (44.24)
k₃lnK	0.596^*** (76.91)	0.470^*** (39.61)	0.466^*** (39.55)
l₁lnL	0.253^*** (31.39)	0.391^*** (8.93)	0.414^*** (13.01)
l₂lnL	0.390^*** (37.55)	0.445^*** (9.73)	0.474^*** (13.80)
l₃lnL	0.357^*** (36.56)	0.465^*** (10.52)	0.489^*** (13.88)
h₁lnH	0.0698^*** (4.09)	0.825^*** (9.60)	0.820^*** (9.55)
h₂lnH	0.0850^*** (6.19)	0.375^*** (3.66)	0.397^*** (3.86)
h₃lnH	0.0752^*** (7.78)	0.656^*** (10.98)	0.644^*** (10.70)
h₄lnH	–0.0221^** (–2.45)	0.0235 (0.60)	0.0172 (0.44)
h₅lnH	–0.0291^*** (–3.14)	0.103^*** (2.77)	0.103^*** (2.75)
_cons	–0.202^*** (–4.56)		–1.394^***
λ		0.529^*** (15.13)	0.534^*** (15.40)
φ			0.256 (1.09)

Note: ^***significant at 1%level of significance; ^**significant at 5%level of significance; ^*significant at 10%level of significance; T value in parentheses; All variables are measured in logs. Source: Author own estimations.

From the specific regression results of these models, it can be seen that: a) technology has a significant positive impact on the actual output, but its coefficient is relatively small, indicating that technology has a relatively small impact on the actual output of China; b) in the aspect of capital stock structure, the elasticity coefficients of capital-output of three industries in China are all positive, rejecting the original hypothesis at the significance level of 1%, which shows that China’s capital stock has a significant positive effect on the actual output. The elastic coefficient of capital-output in the secondary industry is the highest. In the SEM model and the GSPRE model, the elastic coefficient of capital-output in the third industry is the lowest, while it is the lowest in the primary industry in the FGLS model. It can be seen that the capital stock factor plays a significant role in promoting economic growth in the secondary industry in China; c)in the aspect of labor input structure, the elasticity coefficients of labor output of three industries are all positive, and they all reject the original hypothesis at the level of 1%significance, which shows that China’s labor input has a significant positive effect on the actual output. In the SEM model and the GSPRE model, the elasticity coefficient of labor output of the third, secondary and primary industries decreases in turn, while the elasticity coefficient of labor output of the secondary, third and primary industries decreases in turn in the FGLS model; d) in terms of the structural factors of human capital stock, the elasticity coefficient of humancapital-output of junior college and above, senior high school and junior high school is positive, and they all reject the original hypothesis at a significant level of 1%, indicating that China’s acceptance of human capital such as higher education and secondary education has a significant positive effect on the actual output. The elasticity coefficient of human capital-output of primary school and illiterate is negative in the FGLS model, which shows that only primary school education and illiterate human capital have a significant negative effect on the actual output. However, the elasticity coefficient of human capital-output of primary school and illiterate turns to be positive in the SEM model and the GSPRE model, which indicates that considering the spatial heterogeneity at the provincial level and the spatial correlation degree of an individual effect, primary school education or illiterate human capital plays a positive role in the actual output of China.

2.2 Overall analysis of potential economical growing rate of China

According to the regression results of the relevant models, the potential output of six regions in China has been calculated. On this basis, the potential output of China during 1978–2017 could be calculated by cumulative calculation, and the potential economic growth rate of China during 1979–2017 has been calculated accordingly. By calculating China’s real economic growth rate, it can be seen that the change of China’s economic growth rate during 1979–2017 (as shown in Fig. 1).

Fig. 1

China’s economic growth rate: 1979–2017. Source: Author own estimations.

It can be seen from Fig. 1 that: a) China’s potential economic growth rate generally shows a downward trend in the FGLS model, especially during the period of 1979–1991. In 1979, China’s potential economic growth rate is 14.91%. However, it drops rapidly to 7.05%in 1991, with a decline rate of more than 50%. On the contrary, it has entered a rapid growth stage during the period of 1992–2010. It is 7.95%in 1992, but it rises to 10.62%in 2010. However, it has entered the downward channel from 10.12%in 2011 to 6.25%in 2017 with a significant decline. In addition, its overall trend in the SEM model and the GSPRE model is similar to that in the FGLS model, and the overall trend is basically the same; b) the real economic growth rate of China fluctuates around the potential growth rate in the FGLS model, and they basically show a similar law of change. Generally speaking, the change of China’s real economic growth lags behind the trend of potential economic growth, which further indicates that the change of China’s real economic growth mainly depends on the change of potential economic growth. In other words, the change of China’s real economic growth rate is mainly caused by the change of potential economic growth rate; c) in the SEM model and the GSPRE model, the change of China’s real economic growth rate and potential economic growth rate is similar to that in the FGLS model. But during the period of 1981–1986, 1995–1997 and 1999–2009, the change between China’s real economic growth rate and its potential economic growth rate basically shows the opposite development direction, which indicates that in these development stages, the change of China’s real economic growth rate is not only affected by the internal influence of potential economic growth rate, but also by external factors such as demand and policy change. It also means that if the external stimulus of expansionary demand policy is adopted during these periods, China’s real economic growth may get rid of the internal constraints of potential economic growth, so as to achieve stable economic growth.

3. Forecast of China’s potential economic growth rate: 2018–2027

This paper only forecasts and analyzes the potential economic growth rates of different regions and provinces in China in the short term using Fuzzy Logic based Artificial Neural Network [20, 22]. On the basis of the above analysis, this paper specifically forecasts the input of technology, capital stock, labor input, and human capital stock at the interprovincial level in the next 10 years (2018–2027), and then calculates the input of corresponding factors in different regions in China respectively by summing up [2]. For potential economic growth rates, we present layered architecture that consist of five layer: Fuzzy Input layer, Fuzzy Rule Base Interference Layer, Fuzzy Output Layer, Aggregation Layer, and Defuzzification Layer.

3.1 Fuzzy Input layer

In this layer crisp input data is converted into fuzzy linguistic variable based o rules in the form of membership function. Fuzzy membership unction evaluation equation is described using following equation: $u_{mf} (x, p, q, r) = {\begin{matrix} 0, & if x \leq p; \\ \frac{x - p}{r - p}, & if p < x < r; \\ \frac{q - x}{q - r}, & if r < x < q; \\ 0, & if x \geq q \end{matrix}$ (3)

Where, x, p, q, r represent input of technology, capital stock, labor input, and human capital stock which is converted into fuzzy input using Equation (3). Each input value of all input variables is converted into fuzzy input sets, define as: u_mf (x, p, q, r) = ${\begin{matrix} Very Low (VL), Low (L), Medium (M), High (H), \\ VeryHigh (VH) \end{matrix}}$ All input variable is represent based on Gaussian-model membership functions.

3.2 Fuzzy rule base interference layer

This layer consists of neural node that represent fuzzy rule based engine and perform min-max operation over the inputs received from fuzzy input layer min-max operation based rule validation is performed using following predefined conditional equation: $ρ_{j} = \min_{k = 1}^{m} [max_{X_{k}} (U_{k}^{'^{}} (x_{k})^U_{jk} (x_{k}))]$ (4)

Where, represent new autonomous values of input variable. Here, we use Mamdani rule implication that generate individual fuzzy output in the form of membership function. The general format of rule implication is as follows: $\begin{matrix} IF x_{1} is U_{11} and \dots \dots x_{m} is U_{1 m} THEN y is O_{1}; \\ IF x_{1} is U_{m 1} and \dots \dots x_{m} is U_{nm} THEN y is O_{m}; \end{matrix}$

Where x_k = 1 … m represent input and output sets.

3.3 Fuzzy output layer

This layer contain fuzzy output generated at Fuzzy Rule Base Interference Layer using rule base knowledge.

3.4 Aggregation layer

At this layer, aggregation of fuzzy output from various nodes is performed using following equation: $O^{'} (y) = \max (O_{1}^{'^{}} (y), O_{2}^{'^{}} (y), . . . . . ., O_{m}^{'^{}} (y))$ (5)

3.5 Defuzzification layer

The output that is received from aggregation layer is fuzzy form which is converted into real crisp values using following equation of center of gravity (COG) of defuzzification method: $y = \int_{sp} O (y) ydy / \int_{sp} O (y) dy$ (6)

Where, sp represent fuzzy output set support of O (y).

In accordance with the regression outcomes of the former relevant models, the specific forecast of potential economical growing rate of China in 2018–2027 is shown in Table 2.

Table 2

Forecast of China’s potential economic growth rate (%): 2018–2027

FGLS model	2018	2019	2020	2021	2022	2023	2024	2025	2026	2027
North China	3.18	2.96	2.78	2.62	2.44	2.44	2.40	2.40	2.42	2.47
Northeast China	1.92	1.69	1.51	1.37	1.24	1.42	1.17	1.17	1.20	1.26
East China	6.38	6.25	6.13	5.99	5.85	5.60	5.78	5.72	5.68	5.65
Central South China	6.92	6.76	6.61	6.47	6.31	6.02	6.20	6.15	6.11	6.08
Southwest China	7.92	7.91	7.91	7.92	7.96	10.02	8.44	8.81	9.32	10.01
Northwest China	7.75	7.66	7.68	7.75	7.83	8.62	8.09	8.25	8.44	8.63
China	6.21	6.10	6.03	5.97	5.91	5.57	5.97	6.06	6.19	6.38
SEM model	2018	2019	2020	2021	2022	2023	2024	2025	2026	2027
North China	3.70	3.54	3.40	3.28	2.93	3.43	3.19	3.23	3.29	3.38
Northeast China	2.04	1.91	1.81	1.74	1.56	3.24	1.69	1.74	1.82	1.92
East China	5.91	5.81	5.73	5.62	5.54	4.30	5.86	5.84	5.82	5.82
Central South China	6.76	6.64	6.53	6.42	6.19	4.68	6.25	6.23	6.21	6.20
Southwest China	7.09	7.26	7.28	7.31	7.20	20.34	8.03	8.40	8.90	9.58
Northwest China	7.90	7.38	7.42	7.49	7.60	12.30	7.93	8.10	8.30	8.50
China	5.97	5.87	5.84	5.81	5.70	3.37	5.75	5.85	5.98	6.17
GSPRE model	2018	2019	2020	2021	2022	2023	2024	2025	2026	2027
North China	3.70	3.54	3.40	3.29	2.94	3.45	3.20	3.25	3.31	3.41
Northeast China	2.01	1.88	1.78	1.72	1.55	3.26	1.68	1.74	1.82	1.92
East China	5.88	5.79	5.71	5.60	5.52	4.26	5.86	5.84	5.83	5.83
Central South China	6.73	6.61	6.51	6.40	6.18	4.63	6.25	6.23	6.21	6.20
Southwest China	7.05	7.23	7.26	7.29	7.18	20.61	8.01	8.38	8.88	9.56
Northwest China	7.90	7.38	7.42	7.50	7.61	12.42	7.95	8.12	8.32	8.52
China	5.94	5.85	5.82	5.79	5.68	3.30	5.74	5.84	5.97	6.17

Source: Author own estimations.

It can be seen from Table 2 that China’s potential economic growth rate in 2018–2027 presents a U-shaped development track: it declines steadily from 6.21%in 2018 to 5.57%in 2023 in the FGLS model, and then steadily rises to 6.38%in 2027; However, it declines steadily from 5.90%in 2018 to 3.40%in 2023, and then steadily increased to 6.20%in 2027 in the SEM model and the GSPRE model. It is also found that the fluctuation of China’s potential economic growth rate in 2018–2027 is small in the FGLS model, basically maintaining at about 0.5 percentage points, but it is comparatively large in the SEM model and the GSPRE model, reaching about 2 percentage points.

At the regional level, the potential economic growth rates of different regions in China are also quite different [8]. The potential economic growth rates of Northeast China, North China, East China and Central South China are generally in a steady downward trend in 2018–2027, but that of Northwest China and Southwest China are generally in a steady upward trend at the same time. The magnitude of the change is not large, basically maintained at 1 to 2 percentage points. In addition, it is found that the potential economic growth rates of Northeast China and North China will maintain a low-speed growth channel of 2.00%–3.00%in the short term.

In addition, the difference between the potential economic growth rates of different provinces in China are even larger. At the provincial level, the potential economic growth rates of most provinces in China are generally declining. However, the potential economic growth rates of several provinces such as Guizhou, Yunnan, Tibet, Gansu, Qinghai, Ningxia and Xinjiang are generally on the rise. The potential economic growth rate of Guizhou, Tibet, Xinjiang, Qinghai, Ningxia, Yunnan, Fujian will remain above 8.00%in 2018–2027, while that of Liaoning, Inner Mongolia, Shanxi, Heilongjiang, Tianjin will drop to 1.00%–3.00%. The average potential economic growth of Liaoning and Inner Mongolia will even achieve negative growth in the FGLS model. At the same time, the potential economic growth rate of Tibet, Xinjiang, Guizhou, Ningxia, Qinghai, Yunnan, Fujian will remain above 8.00%, while that of Liaoning, Inner Mongolia, Heilongjiang, Shanxi will decline to 1.00%–3.00%. Liaoning’s average potential economic growth will be negative (see Table 3) in the SEM model and the GSPRE model. It can be seen that the potential economic growth rates of different provinces in China in 2018–2027 are roughly the same in the FGLS model, the SEM model and the GSPRE model, which further explains the reliability and robustness of these models adopted in this paper from another point of view.

Table 3

Average potential economic growth rate (%) of different provinces in China: 2018–2027

Provinces	FGLS model			SEM model			GSPRE model
	2018–2022	2023–2027	2018–2027	2018–2022	2023–2027	2018–2027	2018–2022	2023–2027	2018–2027
Beijing	6.47	6.16	6.32	6.92	6.79	6.85	6.95	6.83	6.89
Tianjin	3.17	2.45	2.81	4.03	3.72	3.88	4.06	3.79	3.92
Hebei	3.44	2.64	3.04	4.34	3.76	4.05	4.34	3.78	4.06
Shanxi	0.94	1.41	1.17	2.38	2.94	2.66	2.42	2.98	2.70
Inner Mongolia	0.19	–0.58	–0.20	1.32	0.81	1.07	1.30	0.81	1.05
Liaoning	–0.20	–0.97	–0.59	0.06	–0.46	–0.20	–0.01	–0.51	–0.26
Jilin	3.38	3.22	3.30	3.97	4.09	4.03	4.01	4.15	4.08
Heilongjiang	3.01	2.54	2.77	2.66	2.32	2.49	2.61	2.28	2.45
Shanghai	6.02	6.19	6.10	6.05	6.24	6.15	6.02	6.23	6.12
Jiangsu	6.12	5.36	5.74	5.70	5.15	5.43	5.66	5.13	5.40
Zhejiang	5.59	5.32	5.45	5.38	5.27	5.32	5.37	5.29	5.33
Anhui	5.78	5.07	5.42	5.03	4.40	4.71	5.01	4.39	4.70
Fujian	8.41	8.12	8.27	8.38	8.32	8.35	8.39	8.35	8.37
Jiangxi	6.61	6.03	6.32	5.96	5.45	5.70	5.94	5.45	5.69
Shandong	5.30	4.59	4.95	5.30	4.69	5.00	5.27	4.67	4.97
Henan	6.54	5.82	6.18	6.16	5.66	5.91	6.17	5.68	5.93
Hubei	7.11	6.64	6.87	6.75	6.49	6.62	6.70	6.45	6.58
Hunan	5.27	4.40	4.84	5.24	4.60	4.92	5.17	4.55	4.86
Guangdong	7.65	7.35	7.50	7.83	7.64	7.74	7.81	7.64	7.73
Guangxi	3.94	2.92	3.43	3.87	3.13	3.50	3.87	3.15	3.51
Hainan	6.92	6.58	6.75	7.06	6.83	6.95	7.09	6.87	6.98
Sichuan	6.29	5.59	5.94	5.70	5.28	5.49	5.67	5.26	5.47
Guizhou	11.26	16.44	13.85	10.65	15.67	13.16	10.65	15.61	13.13
Yunnan	8.93	9.30	9.11	8.08	8.75	8.42	8.04	8.71	8.37
Tibet	11.49	12.15	11.82	13.00	13.87	13.44	13.12	14.03	13.58
Shaanxi	6.10	5.50	5.80	5.79	5.14	5.46	5.76	5.11	5.43
Gansu	5.23	5.74	5.48	5.02	5.74	5.38	5.01	5.74	5.37
Qinghai	9.55	9.71	9.63	9.42	8.88	9.15	9.41	8.86	9.14
Ningxia	9.18	9.90	9.54	9.81	10.63	10.22	9.82	10.66	10.24
Xinjiang	10.71	11.62	11.16	10.84	11.71	11.28	10.92	11.81	11.37

Source: Author own estimations.

4. Research on the growth factors of potential economical growing rate of China in the new normal

Since the reform and opening-up, China’s long-term sustained high growth has mainly benefited from the huge contribution of demographic dividend, resource dividend and globalization dividend [17]. However, with the gradual decline of the traditional dividend, potential economical growing rate of China has been entering the downward channel. The era of “low cost but high growth” is gradually drifting away. China’s economy is still in the “low adjustment period” of the post-financial crisis, which is not only the pain period of China’s economic transformation and development but also the key period of cultivating new growth points [18]. At the same time, while the traditional supply factors supporting the potential economic growth rate are tightening, a new round of growth factors is brewing.

4.1 Deepening reform and development

As a way of gradual adjustment and reconstruction of the system, reform is a powerful driving force for social development. At the 40th anniversary of the reform and opening-up, a series of major decision-making arrangements for reform and development will gradually enter a deep-water zone. The development of Western China, the revitalization of Northeast China, the rise of Central China and the lead development strategy in East China will be further coordinated. The “One Belt and One Road” construction, the coordinated development of Beijing-Tianjin-Hebei, the integrated development of the Yangtze River Delta, the construction of Guangdong-Hong Kong-Macau Bay Area, the ecological protection of Yellow River Basin and with high-quality development have steadily advanced. These will further bring China’s centralized release of institutional dividends. Major strategic initiatives such as Hebei Xiong’an New District, Shanghai Free Trade Zone Lingang New Area, Shenzhen Socialist Pioneer Demonstration Zone, as well as a large number of free trade pilot zones, cross-border E-commerce comprehensive pilot zones and national independent innovation demonstration zones will be deeply promoted and implemented. The reform of the state-owned assets management system will be continuously improved. Greater bay area, large gardens, large passageway, and metropolitan areas will be continuously promoted. China’s reform has shown a sound development trend of comprehensive exertion, multiple breakthroughs and deep progress, providing a strong impetus and institutional guarantee for the follow-up development.

4.2 Gradual improvement of human capital

As one of the decisive factors to promote economic development in the present society, the human capital dividend characterized by the continuous improvement of per capita education level gradually replaces the population dividend, which is characterized by the size and quantity of population, and plays an increasingly obvious role in China’s future economic development. It has been calculated that one year’s increase in the number of years of education for major labor force can drive the growth of GDP by 1 percentage point.

In 2017, there are 13,875 people with a college degree and 17,554 people with a high school level in every 100,000 people in China, compared with 7,287 and 13,797 in 2009, which increase by 90.41%and 27.23%respectively (as shown in Fig. 2). It can be seen that the growing scale of high-quality laborers has made a qualitative leap in China’s labor structure, and the continuous improvement and effective utilization of human capital will surely play a huge role in promoting China’s economy.

Fig. 2

The educational level of every 100,000 people in China: 2009–2017. Source: China Statistical Yearbook: 2010-2018.

4.3 Large growth space for innovation elements

After years of development, China’s innovation level has been continuously improved, and innovation factors such as environment, resources and opportunities have been continuously gathered, and international competitiveness has been continuously improved. It is undeniable that there is still need for perfection of China’s current innovation factors. Taking R&D investment as an example, China’s R&D expenditures and its share of GDP have steadily increased. China’s R&D expenditure is 5,802.10 billion Yuan in 2009, accounting for 1.70%of GDP. However, it reaches 1,955.7 billion Yuan in 2009, with an average annual growth rate of 14.53%, with its share of GDP rising to 2.18%(as shown in Fig. 3).

Fig. 3

R&D disbursement of China and its GDP share: 2009–2018. Source: Statistical Bulletin on National Science and Technology Funding: 2009-2018.

According to PPP, it is 370.6 billion US dollars in 2018, ranking after the United States ($476.5 billion). From the perspective of its share of GDP, China is only 1.97%, lower than South Korea (4.32%), Japan (3.50%), the United States (2.84%), Germany (2.84%), Canada (2.34%), France (2.25), ranking seventh in the world in 2018 (as shown in Fig. 4).

Fig. 4

The Proportion of R&D investment in GDP (%) of major countries and regions in 2018. Source: World Bank WDI database.

4.4 Increasing efficiency of the resource allocation

China’s resource allocation efficiency has further improved in recent years. In terms of logistics performance and the time required for companies to start a business, China’s logistics performance index has steadily increased, and the time has steadily declined. China’s logistics performance index in 2016 is 3.66, compared with 3.32 in 2007 (as shown in Fig. 5).

Fig. 5

China’s logistics performance index: 2007–2016. Source: World Bank WDI database.

Meanwhile, the time required for Chinese companies to start a business has dropped rapidly from 32.40 days in 2013 to 8.60 days in 2018 (as shown in Fig. 6). In the process of fully promoting the supply-side structural reform, China will further strengthen and coordinate resource allocation, give full play to the decisive role of the market, establish a differential allocation mechanism for resource, and rationally allocate public services resources. In the future, China will usher in a new round of reform dividends, innovation dividends, and institutional dividends, which will promote industrial transformation and upgrading, and optimize the efficiency of resource allocation, and then form a labor productivity dividend to promote further economic growth.

Fig. 6

The time required for Chinese companies to start a business: 2013–2018. Source: World Bank WDI database.

4.5 Accelerating process of urbanization

In the past 40 years of transformation and start-up, development process of China has been hastening. urbanization rate of China has been increasing by rate of 17.90%in year 1978 to rate of 59.15%in year 2018, with an average yearly growing rate of 1.03%(as shown in Fig. 7). Although it is higher than the world average level (55.27%), there is still a big gap compared with Japan (91.62%), the United States (82.26%), South Korea (81.46%), the European Union (76.99%) and other developed countries. Compared with the average urbanization rate of more than 80%in high-income countries. China’s urbanization process still has a lot of room for development. There is more space for development in the process of urbanization in China.

Fig. 7

China’s urbanization rate: 1978–2018. Source: World Bank WDI database.

In addition, the proportion of the urban population in China with a population of more than 1 million in 2018 is 27.89%. Comparing with high-income countries such as Japan (64.63%), South Korea (50.15%), and the United States (46.26%), it is still relatively low. The acceleration of the urbanization process will not only bring the expansion of the total consumption, infrastructure, housing construction but also further expand the demand for public service facilities such as education, medical care, insurance. It will provide a broad development space for the subsequent entry of various types of capital. It can be foreseen that the accelerating urbanization process will bring unlimited development potential and broad development opportunities to China.

5. Research on the countermeasures to improve potential economical growing rate of China in the new normal: from the perspective of AI

It shows that AI can stimulate economic growth potential in the following way. first, AI greatly improves labor productivity by changing working methods; second, it can even replace most of the labor force and become a brand-new production factor; third, it can drive the upgrading of the industrial structure and promote innovation.

5.1 The strategic layout of main nations of world in the field of AI

AI has been the principal technology of the latest generation of technical revolt in recent years, especially the advent of the 5 G era, which has accelerated the development and application of AI, profoundly affecting the global human production and life, as well as the political, economic and cultural development of various countries. In order to become accustomed to the latest scientific and technical revolt, major nations in the worldwide have formulated the development strategies and plans of AI to promote their own economic growth. However, due to different market measure, information resources, technical aspects, rules and protocols, industrialized endowment, the planned layout of nations in the field of AI is diverse, which is responsible for the basis and circumstances for universal cooperation.

5.1.1 United states

In recent years, due to the rising unemployment rate and falling output, the potential economic growth rate in the United States has been generally declining. According to the US Congressional Budget Office, it is 2.60%in 2007. However, it falls to 2.10%in 2017. The U.S. government proposed a series of controversial reforms to boost economic development momentum in response to the continuous downturn in the domestic economy.

The United States officially raised AI to the national strategic level in October 2016, focusing on the Internet, chips, operating systems and other computer hardware and software, financial fields, military and energy and other fields, in order to maintain its global technological leadership. In the overall AI planning of the United States, it seeks to explore the expected impact of AI-driven automation on the economy, study the opportunities and challenges that AI brings to social employment, and then propose corresponding plans and measures to deal with the relevant impacts. In addition, the United States is also the first country in history to prioritize AI, autonomous, and unmanned systems in its budget.

These stimulus policies have promoted domestic economic growth to a certain extent in the short term, and the United States economy has entered a period of sustained moderate and steady recovery. In 2018, the nominal economic growth rate of the United States reaches 5.18%, and the real economic growth rate reaches 2.86%, which is close to the target of 3%set by the United States Government(as shown in Fig. 8). However, due to the labor supply problem caused by the aging of the population, structural reasons such as the slowdown in labor productivity growth and the continued low level of consumption, coupled with the uncertainty caused by large-scale trade wars with countries such as China and Mexico (Jason, 2017), the potential economic growth rate of the United States will still remain between 1.80%and 2.00%in the short term, according to the forecast of the Federal Reserve.

Fig. 8

U.S. real economic growth rate: 1990–2018. Source: World Bank WDI database.

5.1.2 Japan

Since the 1990 s, due to the aging of the population, the rapid decline in capital stock growth, the slow increase in human capital, and the negative growth in total factor productivity, Japan’s potential economic growth has generally shown a rapid decline. In recent years, it has been maintained at a low level of around 0.50%(Liu and Fan, 2019).

The Japanese government and the business community attach great importance to the development of AI, not only taking the IOT, AI and robots as the core of the fourth industrial revolution, but also establishing a relatively complete R&D promotion mechanism at the national level. Japan’s AI strategy advocates the integration of AI technology with various fields, and implements applications in industries such as industry, agriculture, medicine, logistics, and intelligent transportation. Japan hopes that by vigorously developing AI, it will maintain and expand its technological advantages in the fields of automobiles and robots, and gradually solve social problems such as population aging, labor shortage, medical care, and pension.

In general, Japan’s real economic growth has been retreating since the financial crisis. Although it returns to the growth trend again in 2009, it has been plagued by problems such as the serious aging of the population, the high per capita debt, the increase of export risk caused by the trade friction with the United States and South Korea, and the lack of personal consumption growth. It will remain at a low level in the short term (as shown in Fig. 9).

Fig. 9

Japan’s real economic growth rate: 1990–2018. Source: World Bank WDI database.

5.1.3 South korea

The South Korean government attaches great importance to the development of AI, vigorously supports the AI industry and related companies, and focuses on the fields of IOT, cloud, big data, and speech recognition. Various policies issued by the government pay more attention to the cultivation of talents and the cultivation of AI enterprises, which are highly targeted. South Korea’s GDP growth rate has been increasing rapidlly. The average growth rate of South Korea’s GDP remains at a relatively high level of 3.43%during 2010–2017. In May 2018, the South Korean government formulated the “AI Development Strategy”, focusing on increasing investment in talents, technology and infrastructure. In terms of AI applications, South Korea pays attention to the practical application of AI technology in the fields of finance, medical care, smart cities, and transportation. Although it has been declining slightly since 2018, it still remains above 2.5%(as shown in Fig. 10).

Fig. 10

South Korea’s real economic growth rate: 1990–2018. Source: World Bank WDI database.

However, due to the decrease in investment and labor force, the decline in the export value-added, the weak domestic demand and the slow development of new industries, the average potential economic growth rate of South Korea has dropped from 4.70%in 1998–2007 to 3.80%in 2008–2012. It can be seen that the deterioration of the financial growth rate has already produced a heavy burden on the South Korean economy. The South Korea Economic Research Institute predicts that it will fall to 2.50%in 2019–2022, and will even fall to 2.30%in 2023–2030.

5.1.4 India

As one of the fastest-growing major economies in the world, India’s economic growth has been eye-catching. Since the 1990 s, India has vigorously promoted reforms and launched a series of related measures to stimulate economic development, and it has been running on a fast track in the economic sphere. India has vigorously promoted technological innovation and development, and has made great efforts in the field of AI. With the rapid development of mobile Internet technology and software technology, India’s economy has entered the fast lane. India’s actual average economic growth rate is 5.77%during 1990–1999. However, it reaches 8.50%in 2010, exceeding China for the first time. In recent years, India’s real economic growth rate has declined slightly, but it still reaches a high value of about 7.00%(as shown in Fig. 11).

Fig. 11

India’s real economic growth rate: 1990–2018. Source: World Bank WDI database.

The Indian government has launched a series of related initiatives aimed at promoting economic growth potential in recent years. India focuses on the development of cloud computing, 5 G, machine learning, big data and other technologies. At the same time, India emphasizes the practicality of AI, focusing on the practical changes brought about by strengthening AI in the areas such as health care, agriculture, education, smart cities, infrastructure and intelligent transportation. In May 2018, India released the “National AI Strategy”, which aims to achieve the goal of “AI for all”. Against this background, India has strengthened investment in scientific research, encouraged skills training, and accelerated the application of AI throughout the industry chain. The introduction of these related measures has achieved certain results. India’s economic growth has been remarkable in recent years. India’s real economic growth rate reaches 7.55%during 2014–2017. It has been slightly reduced to 6.98%in 2018, but since 2014 it has surpassed China again, and its development strength can’t be underestimated. It will remain between 8.00%and 10.00%in the future.

5.2 AI: boosting potential economical growing rate of China in the new normal

At present, there are still many uncertainties in the international situation. The global epidemic of the new coronavirus pneumonia has not been effectively controlled, global trade remains sluggish, and world economic growth remains weak. For China, it is difficult to recover effectively in the short period. The service industry represented by tourism, catering, accommodation and entertainment, as well as the manufacturing industry, have been greatly affected. In addition, the impact of the epidemic on the employment quantity and quality of SMEs in China is still not to be underestimated, and it is highlighted that the problems of returning migrant workers and the college graduates’employment still exist seriously.

With the recent aggregation of many transformative technologies, all kinds of markets are coming towards the inside of new era. Artificial Intelligence (AI)I is expected to overcomesomatic constraints such as assets and effort, and exposed new homes of growth and value. Recent analysis of Accenture shows that by 2035, AI has the broad prospective to improve potential economical growing rate of China by 1.6 percentage points. As a brand-new manufacturing factor, AI will promote financial progress in at least three significantfacets. First, it can generate a new sort of simulated labor force, which is called “smart automation”. Second, AI can increment and improve the skill and proficiencies of existing employment and physical assets. Third, similarto other techniques used earlier, AI can also motivate innovation. Eventually, when numerous markets not only employ AI to transform construction methods, but use it to dynamically exposed up to new improvement space, AI will unquestionably encourage structural revolution comprehensively and intensely.

Promoting the conversion of old and new kinetic energy, and accelerating the cultivation of new driving forces for development. Countermeasures and suggestions are as follows: a) taking digital economy as the core, and deeply implementing the strategy of “integration of informatization and industrialization”; b) expanding R&D investment of the whole society and gradually increasing the contribution rate of scientific and technological progress; c)accelerating the construction of innovation capacity in key common technology, leading technology, disruptive technology, and creating a new engine for economic growth; d)vigorously supporting new technologies, new models, new formats, new industries, and cultivating the “economic growth pole”;e) implementing major science and technology projects, accelerating the advancement of infrastructure construction in frontier fields such as information science, materials science, and life sciences, and focusing on a new round of technological revolutions such as big data, new generation AI, 5 G, IOT, cloud computing, robotics, and blockchain.

In terms of potential economic growth, the significant impact of AI is not to replace existing labor and capital, but to empower it and make it more effectively used. Therefore, the governments at all levels should optimize the investment structure, enhance the endogenous power of economic growth, and reactivate the potential of China’s economic growth and revitalize the industry. For example, they should not only increase effective investment to make up for the shortcomings in the field of public services, especially in the fields of health care, culture and education, but also strengthen the construction of key projects, focusing on large projects, platforms and industries, and increase effective investment in new digital economic industries such as big data, new generation AI, 5 G, IOT, cloud computing, robotics, and blockchain.

Optimizing the industrial structure and improving the modern industrial system. Accelerating the construction of smart manufacturing, based on modern industrial clusters, further developing key development areas such as agglomeration economy, platform economy, sharing economy, and smart economy, and making the biological economy, green and low-carbon economy bigger and stronger. Accelerating technological innovation, especially the research and development and application of new materials, high-end equipment, key core components and other “jam neck” technologies.

As a new production factor, AI will bring huge development space for China’s potential economic growth. At present, China’s demographic dividend is declining, which is undoubtedly a huge challenge to the future Chinese economy. However, with the help of AI, China will once again benefit from the huge population base and effectively solve the problem of labor shortage. AI can not only improve labor output, but also improve the efficiency of human capital. Therefore, the governments at all levels should continuously increasing investment in education, and raise the level of human capital. Taking the improvement of human capital as the focus of supply-side reform, the total factor productivity should be constantly promoted by relying on the growth factors such as education and knowledge. They have to increase investment in education to promote connotation development, characteristic development of education, as well as strengthen the accumulation of human capital.

Footnotes

Acknowledgments

This research was funded by the Chinese National Funding of Social Sciences (Grant No.17CJL008), Zhejiang Provincial Natural Science Foundation of China (Grant No. LY17G030001), Program for the Cultivation of Excellent Youth Scholars in Social Science, Hangzhou, China (Grant No. 2018RCZX25) and Scientific Research Fund for Teachers of Zhejiang University City College (Grant No. J-20004).

References

Ang

, Bekaert

and Wei

, Do macro variables, asset markets, or surveys forecast inflation better? Journal of Monetary Economics, Elsevier 54(4) (2007), 1163–1212.

B.X.

, Zhao

and Feng

Z.X.

, 70 Years of Economic Growth in New China: Trends, Cycles and Structural Characteristics, Management World (9) (2019), 16–29.

Bobadilla

, Ortega

, Gutiérrez

and Alonso

, Classification-based Deep Neural Network Architecture for Collaborative Filtering Recommender Systems, International Journal of Interactive Multimedia and Artificial Intelligence 6(Special Issue on Soft Computing) (2020), 68–77.

Buragohain

and Mahanta

, A novel approach for ANFIS modelling based on full factorial design, Appl Soft Comput 8 (2008), 609–625.

Furceri

and Mourougane

, The Effect of Financial Crises on Potential Output: New Empirical Evidence from OECD Countries, Journal of Macroeconomics 34(3) (2012), 822–832.

Cai

and Lu

, Population Change and Resulting Slowdown in Potential GDP Growth in China, China & World Economy 21(2) (2013), 1–14.

Cai

, How to Tackle the Slowdown of Potential Growth Rate in China? China Finance & Economic Review 3(1) (2015), 2.

Yuan

, Structural Accelerating and Structural Decelerating in the Long Run Growth: An Exposition, Economic Research Journal (3) (2012), 127–140.

Jason

, Potential Growth Rate and Policy of the US Economy, New Finance Review (3) (2017), 25–36.

10.

Ibrahim

, Kamel

, Rashad

, Nasrat

and Jurado

, Performance Enhancement of Wind Farms Using Tuned SSSC Based on Artificial Neural Network, International Journal of Interactive Multimedia and Artificial Intelligence 5(Regular Issue) (2019), 118–124.

11.

Fernald

J.G.

, Productivity and Potential Output Before, During, and After the Great Recession, NBER Macroeconomics Annual 29(1) (2015), 1–51.

12.

Ahmadi

, Ghoushchi

S.J.

, Taghizadeh

and Sharifi

, Presentation of a new hybrid approach for forecasting economic growth using artificial intelligence approaches, Neural Computing and Applications 31(12) (2019), 8661–8680.

13.

Mardani

, Van Fan

, Nilashi

, Hooker

R.E.

, Ozkul

and Streimikiene

and Loganathan

, A two-stage methodology based on ensemble Adaptive Neuro-Fuzzy Inference System to predict carbon dioxide emissions, J Clean Prod 231 (2019), 446–461. [CrossRef]

14.

Aghion

, Jones

B.F.

and Jones

C.I.

, Artificial Intelligence and Economic Growth (9) (2017), NBER Working Papers.

15.

Halmai

and Vásáry

, Crisis and Economic Growth in the EU: Medium and Long–term Trends, Acta Oeconomica 61(4) (2012), 411–431.

16.

Chen

Q.L.

, Xu

and Zhou

, AI’s Effects on Economic Growth in Aging Society: Induced Innovation and Labor Supplemental Substitution, China Economist (5) (2019), 54–66.

17.

Abdulov

, Artificial Intelligence as an Important Factor of Sustainable and Crisis-Free Economic Growth, Procedia Computer Science (169) (2020), 468–472.

18.

Santis

R.D.

, Esposito

and Masi

, Structural Determinants of Potential Output Growth in Europe and the Role of Fiscal Policy, International Economics and Economic Policy 16(4) (2019), 565–591.

19.

Research Group of Macroeconomic Research Institute of China National Development and Reform Commission, Research on China’s Economic Growth Potential and Economic Growth Prospects from the 12th Five-Year Plan Period to 2030, Economic Perspectives, (3) (2008), 19–25.

20.

ThiNgan

, Ali

, Fujita

, Giang

N.L.

, Manogaran

and Priyan

M.K.

, A New Representation of Intuitionistic Fuzzy Systems and Their Applications in Critical Decision Making, IEEE Intelligent Systems.

21.

Rodríguez

, Corredor

and Suárez

, Genetic Operators Applied to Symmetric Cryptography, International Journal of Interactive Multimedia and Artificial Intelligence 5(Regular Issue) (2019), 39–49.

22.

Roopa

C.K.

and Harish

B.S.

, Automated ECG Analysis for Localizing Thrombus in Culprit Artery Using Rule Based Information Fuzzy Network, International Journal of Interactive Multimedia and Artificial Intelligence 6(Special Issue on Soft Computing) (2020), 16–25.

23.

Saleem

and Khattak

M.I.

, Deep Neural Networks for Speech Enhancement in Complex-Noisy Environments, International Journal of Interactive Multimediaand Artificial Intelligence 6(Special Issue on Soft Computing) (2020), 84–90.

24.

Mazilescu

, Artificial Intelligence as a New Factor of Global Economic Growth, Economics & Applied Informatics (3) (2018), 124–133.

25.

Liu

and Fan

, China’s Development Is Still in an Important Period of Strategic Opportunity: China’s Potential growth rate and Growth Leap, Management World (13) (2019), 13–23.

26.

Guo

and Ren

B.P.

, Structural Change, Factor Output Elasticity and China’s Potential growth rate, The Journal of Quantitative & Technical Economics (9) (2014), 72–84.

27.

Kim

W.J.

and Hammodeh

, Impacts of Global and Domestic Shocks on Inflation and Economic Growth for Actual and Potential GCC Member Countries, International Review of Economics & Finance (27) (2013), 298–317.

28.

and Cai

, From Demographic Dividends to Reform Dividends: A Simulation Based on China’s Potential Growth Rate, The Journal of World Economy (13) (2016), 3–23.

29.

and Yin

J.F.

, High Saving Rate, High Investment Rate and Chinese Economic Growth During Labour Transition, Economic Research Journal (2) (2005), 4–15.

30.

and Zhang

X.J.

, The New Normal: The Logic and Perspective of Economic Development, Economic Research Journal (5) (2015), 4–19.