Finance Research Letters 38 (2021) 101488

Contents lists available at ScienceDirect

Finance Research Letters

journal homepage: www.elsevier.com/locate/frl

Trade openness and economic growth quality of China: Empirical

analysis using ARDL model
Qunxi Kong a, Dan Peng a, Yehui Ni a, Xinyue Jiang a, Ziqi Wang b, 1, *
a
School of Industrial Development, Nanjing University of Finance & Economics, Nanjing, China
b
School of Business, Nanjing University, Nanjing, China

A R T I C L E I N F O A B S T R A C T

Keywords: This paper studies the influential relationship between Chinese opening up and economic growth
Trade openness quality in China under exchange rate fluctuation from 1994 to 2018. There is a long-term stable
Exchange rate fluctuation co-integration relationship between opening-up and the quality of economic growth. Trade
Economic growth quality
openness can significantly promote the quality of economic growth in both the short and long
Regional heterogeneity
Non-linear relationship
term. When the short-term fluctuation deviated from the long-term equilibrium, the quality of
economic growth can remain stable through automatic adjustment. The positive impact of trade
openness on quality of economic growth had a significant regional heterogeneity and non-linear
threshold characteristics.

1. Introduction and literature review

It is a relatively consistent view, regarding the relationship between trade openness and economic growth in the circle of foreign
economic theory, that trade openness can facilitate the promotion of economic growth quality. Neoclassical growth theory holds that
trade openness can promote capital formation and the enhancement of resource allocation efficiency, thus facilitating the improve­
ment of economic growth quality (Helpman, 1985; Rodrik, 1988). New growth theory posits that trade openness mainly elevates
economic growth quality through accelerating technical progress and boosting factor productivity (Romer, 1986; Robert, 1988). In
recent years, scholars have been studying the relationship between trade openness and economic growth. Idris (2016) has argued that
trade openness was one of the main drivers of economic growth. Keho (2017) has suggested that trade openness has a positive in­
fluence on economic growth in both the short and long term, and that there is a positive and powerful complementary relationship
between trade openness and capital formation in the promotion of economic growth. Huchet-Bourdon (2018) has stated that, in the

* Corresponding author.
E-mail address: 1012660908@qq.com (Z. Wang).
1
We would like to thank editors and anonymous referees for useful comments and suggestions. We acknowledge financial support from the
National Natural Science Foundation of China (No. 71303105): The Study of the Regional Innovation System Centered on Knowledge-intensive
Service Enterprises——based on the Perspective of Spatial Agglomeration, and National Natural Science Foundation of China (No. 71773047):
The Research of Human Capital Heterogeneity, Innovation and Producer Service Productivity: Influence and Approach, and National Social Science
Foundation (No. 19FJYB039): Research on the Relationship Between Outward Foreign Direct Investment and China’s Economic Growth in High
Quality Development Stage, and Jiangsu soft science program project (No. BR2018043): New Connotation of Science and Technology Finance:
Research on the Path, Mode and Mechanism of Financial Support for the High Quality Development and Upgrading of Science and Technology
Innovation and Entrepreneurship and the Construction of Modern Economic System of Service, and Jiangsu Practical Innovation Project (No.
SJKY19_1235): The Impact of Foreign Direct Investment on the Quality of China’s Economic Growth: Theoretical Mechanism and Realization Path.

https://doi.org/10.1016/j.frl.2020.101488
Received 5 November 2019; Received in revised form 17 February 2020; Accepted 5 March 2020
Available online 6 March 2020
1544-6123/© 2020 Elsevier Inc. All rights reserved.
Q. Kong et al. Finance Research Letters 38 (2021) 101488

long run, trade openness is conducive to economic growth.

Meanwhile, existing studies suggest that developing countries can better utilize the existing knowledge storage of developed
countries through trade. For example, Abid (2019) and Mensi et al. (2019) have found that an increase in trade openness can accelerate
technical progress of advantaged industries, thus promoting the improvement of economic quality. Additionally, trade openness
significantly influences capital deepening and promotes economic growth quality by changing the structure of the labor force. Wang
et al. (2019) have contended that trade openness improves capacity utilization and output levels significantly by expanding the market
size, as well as significantly influencing capital deepening and driving economic growth by changing the structure of the labor force.
According to Zhang et al.(2019a), opening up to the outside world has had a significant positive impact on China’s economic growth,
while regional opening up has significantly hindered China’s rapid economic growth. Zhang and Guo (2019) reported that there were
significant non-linear characteristics between trade openness and growth in the Chinese economy. The degree of trade openness had an
asymmetric effect on economic growth. The promoting effect of trade openness on economic growth decreased with the increase of the
degree of openness, but the two were always positively correlated. According to Wei et al. (2019) and Zhang et al (2019c), after 2008,
the influence of trade openness on economic growth showed a downward trend.
The common feature of the above studies on the linear relationship between trade openness and the quality of economic growth is
that all assume that there is a long-term stable relationship between the two. Moreover, these studies did not consider the cyclical
factors of economic operation, and did not describe the mechanisms of the two factors in the short term. Therefore, this paper first
builds an ARDL (Autoregressive-distributed Lag) model based on a unified analytical framework to study the long-term relationship
between trade openness and the quality of economic growth under exchange rate fluctuations. Second, this paper establishes an error
correction model to describe the short-term relationship between the two. Finally, this study uses a threshold model to explore the non-
linear relationship between trade openness and the quality of economic growth.

2. Methodology and data resources

2.1. Econometric methodology

2.1.1. ARDL model

The connection between trade openness, FDI and economic growth quality in China can be analyzed using the analytical framework
set up by Baharom et al (2008) and Zhang et al (2019b). Referring to Emery (1967), Balassa (1978) and Hendry (1987), and using the
degree of openness of trade as an indicator of opening up, the benchmark regression model is set as follows:
ECOt = f (FDIt , TOPt , REERt ) (1)

The explanatory variables in the model include foreign direct investment (FDIt), trade openness (TOPt) and real effective exchange rate
(REERt), and their lag values are expressed with FDIt − i, TOPt − i and REERt − i, respectively. The endogenous variable is economic
growth quality (ECOt), and its lag value is ECOt − i.
In consideration of the time lagging in the promotion of quality of economic growth, this study selected an ARDL model to study the
influence of the degree of trade openness on quality of economic growth. The ARDL model was calculated in two steps:
Step 1: The co-integration test ARDL model, which was used to test whether there was a long-term causal relationship between the
variables. The following model was established:
ΔlnECOt = β0 + β1 lnECOt− 1 + β2 lnFDIt− 1 + β3 lnTOPt− 1 + β4 lnREERt− 1
∑a ∑
b ∑
c
+ β5,i ΔlnECOt− i + β6,i ΔlnFDIt− i + β7,i ΔlnTOPt− i
i=1 i=0 i=0 (2)
∑
d
+ β8,i ΔlnREERt− i + ut
i=0

where Δ was the first-order differential operator, ut was the white noise and a, b, c, d were the maximum lag orders as determined by
AIC or BIC. Whether there was a long-term equilibrium relationship between horizontal variables was tested using F-statistic, and the
null hypothesis was that there was no long-term equilibrium relationship.
Step 2: The estimation ARDL model, which was used to analyze the long- and short-term relationships between the variables. The
long-term relationship can be estimated using the ARDL(p1,p2,p3,p4) model:
p1
∑ p2
∑ p3
∑ p4
∑
ΔlnECOt = γ0 + γ1 lnECOt− i + γ 2 lnFDIt− i + γ 3 lnTOPt− i + γ 4 lnREERt− i + εi (3)
i=1 i=0 i=0 i=0

While the short-term relationship can be estimated using the ARDL-ECM model. See Eq. (4) for the concrete model:
p1
∑ p2
∑ p3
∑ p4
∑
ΔlnECOt = γ0 + γ1 lnECOt− i + γ 2 lnFDIt− i + γ 3 lnTOPt− i + γ4 lnREERt−
(4)
i
i=1 i=0 i=0 i=0
+δ5 ECMt− 1 + μt

2
Q. Kong et al. Finance Research Letters 38 (2021) 101488

2.1.2. Threshold model

The single threshold regression model of this paper was designed according to the panel data threshold model of Hansen (1999):
lnECOij = α0 + α1 lnFDIij + α2 lnREERij + α3 lnTOPij ⋅ I(TOP < φ)
(5)
+α4 lnTOPij ⋅ I(TOP ≥ φ) + α5 Xit + εit

where i stood for the observed individual, t for the time, trade openness (TOP) for the threshold variable, ϕfor the threshold value and
Xit for the set of independent variables. The control variables selected in this paper were industrial structure, resource endowment,
innovation level, human capital level and government financial expenditure (Yuan Qian and Wu Lihua, 2019; Zhang et al., 2020a;
James and Aadland, 2011).

2.2. Variable description and data sources

Provincial-level China panel data during the period of 1994–2018 have been adopted as research samples for this study. Due to the
lack of some data from Tibet, 30 provinces (cities, regions) were ultimately chosen as the section samples. The variable data was taken
from China Statistical Yearbook, China Statistical Yearbook of Science and Technology, Statistical Bulletin of Chinese Foreign Direct In­
vestment, statistical yearbooks of each province and city, and the official database of the International Monetary Fund (IMF). Relevant
variables are defined as below:

Table 1
Economic growth quality index system.
Dimensions Sub-index Basic indicators Proxy variables

Economic growth Factor productivity Capital productivity GDP/Capital stock

efficiency Labor productivity GDP/Number of employees
Production organization TFP TFP
efficiency Technical efficiency EC
Technological changes TC
Market efficiency Industrialization rate Non-agricultural employment/total employment
Economic growth Industrial structure The second industry compares labor Second industry output ratio/Second industry
stability productivity employment ratio
The tertiary industry compares labor Tertiary industry output ratio/Tertiary industry
productivity employment ratio
Urban and rural structure Binary contrast coefficient Ratio of agricultural and non-agricultural labor
productivity
Binary Contrast Index ∣Non-agricultural output value minus the
proportion of labor∣
Investment and consumption Investment rate Gross capital formation/GDP
structure Consumption rate Final consumption expenditure/GDP
Financial structure Balance of deposits Deposit balance/GDP
Loan balance ratio Loan balance/GDP
Economic stability Economic volatility ∣Change in economic growth rate∣
Urban and rural income Rural Households Engel’s Coefficient Rural food expenditure/consumption
Urban Household Engel’s Coefficient Urban food expenditure/consumption
Employment fluctuations Urban registered unemployment rate Urban registered unemployment rate
Outcome allocation Workers’ share of compensation Workers’ compensation/GDP
∑
Theil index dis = 2i=1 ((pit /pt ) × ln(pit /pt /zit /zt ))
Economic growth Environmental pollution Wastewater discharge per unit of output(t/ Total industrial wastewater discharge/GDP
sustainability million)
Unit output solid waste emissions(t/ Industrial waste production/GDP
million)
Resource consumption Gross power consumption per unit of area Electricity consumption/GDP
(kWh/yuan)
Openness Public Service Total import and export/GDP Total import and export/GDP
OFDI/GDP OFDI/GDP
Public Service General Public Service Expenses General public service Expenditure/fiscal
expenditure
Public safety spending ratio Public safety expenditure/fiscal expenditure
Basic quality Railway mileage Railway mileage
Kilometers Kilometers
Technology and Innovation The proportion of science and technology Science and technology expenditure/fiscal
expenditure expenditure
Three types of patent approvals per year in Three types of patent approvals per year
each region

Note: Theil index is used to measure the income gap between regions. i represents towns and villages respectively;Ztare the total population at period
t;Ptare the total income at period t.

3
Q. Kong et al. Finance Research Letters 38 (2021) 101488

2.2.1. Economic growth quality(ECO)

Referring to Kong Qunxi et al. (2019). high-quality economic growth should be stable and continuous growth under efficient
growth mode; accordingly, multiple basic indexes are overall considered to construct the indexes of economic growth quality. The
specific evaluation system construction is shown in Table 1. Thirty-one basic indexes have been chosen, and dimensionless treatment
was conducted for each proxy variable. Then, dimensionality reduction treatment is carried out for each basic index through principal
component analysis. Finally, the comprehensive indexes of economic growth quality have been gained.

2.2.2. Foreign direct investment (FDI)

The investment effect of FDI often has time lag, and thus it is more scientific to use FDI stock data rather than FDI flow data.
Considering that no computation of depreciation will overestimate the value of FDI asset, calculation is required, and the formula is:
FDIt = It /Pt + (1 − δt )FDIt− 1 (6)

wherein, FDItis actual FDI stock for t years; FDIt − 1is actual FDI stock for t-1 years. Pt is price index of investment in fixed assets. It is
nominal FDI for t years converted with RMB. δt is depreciation rate of FDI asset for t years. This paper refers to the method of Hou et al.,
2005 and assumes that the depreciation rate of FDI asset is 11%.

2.2.3. Trade openness (TOP)

Under the precondition of considering basic conditions of China and data availability, trade interdependency (TI) has been selected
as the index to measure trade openness, and the computational formula is:
(IMPORTi + EXPORTi )
TIi = (7)
GDPi

wherein, IMPORTi and EXPORTi refer to the import and export amounts of a country, and GDPi is the gross domestic product of a
country.

2.2.4. Real exchange rate fluctuation (REER)

This paper calculated the exchange rate fluctuation using coefficient of variation and made a regression analysis on the variables.
To be specific, the coefficient of variation of the monthly data of real exchange rate in the same year was used for measuring. The
standard deviation was divided by the monthly average, to eliminate the influence of different dimensions of the real exchange rate
values of two sides. The following formula was established:
( )/ ( )
REERj,t = Std.Dev REERj,m Mean REERj,m (8)

3. The long-term and short-term equilibrium between opening-up and quality of economic growth

3.1. Preliminary investigation

In order to further study the correlation between opening up and the quality of economic growth, this paper chose two dimensions
of trade openness and foreign direct investment for specific research. Figs. 1 and 2 illustrate that both degree of trade openness and
outward direct investment were positively correlated to quality of economic growth, but, in the context of exchange rate fluctuation, if
quality of economic growth, degree of trade openness, FDI and real exchange rate were incorporated into the same analytical
framework, the influence relationship between opening-up and quality of economic growth must be further tested.

3.2. Results of unit root test and co-integration test

Firstly, an NP unit root test was carried out on the stationarity of multiple variable sequences. The results showed that the hori­
zontal series of natural logarithmic sequences of the variables involved in this paper were all non-stationary series, but became sta­
tionary series after first-order difference. Thus, I (1) was stationary and it was possible to estimate the ARDL model.
Secondly, this study performed a co-integration test on the relationship between four variables: lnECOt, lnFDIt, lnTOPt and lnREERt
using the Johansen test. The test results (Table 2) indicated that at the significant levels of 5% and 1%, there was a unique co-
integration relationship among lnECOt, lnFDIt, lnTOPt and lnREERt.

3.3. Estimation results of short-term and long-term ARDL model

Before estimating the long-term coefficient, it was necessary to determine the lag order of the model. According to the actual
statistics of sample data, the optimal lag order of each variable in the model was identified using Schwarz-Bayes Criterion (SBC).
Considering the length of sample data, in this paper, the maximum lag order of each variable was limited to 2, and finally ARDL
(1,2,2,2) was identified to be the most appropriate. The estimation results are shown in Table 3.
As Table 3 illustrates, the degree of trade openness grew by 1% and the quality of economic growth grew by about 70–80%,
indicating which meant that the degree of trade openness can promote the quality of economic growth in both long and short terms.

4
Q. Kong et al. Finance Research Letters 38 (2021) 101488

Fig 1. TOP and ECO.

Fig 2. FDI and ECO.

Moreover, the adjustment coefficient of the error correction terms was − 0.3265, indicating that when the short-term fluctuation
deviated from the long-term equilibrium, it could be pulled back to equilibrium with an adjustment intensity of 0.3265 and that
economic growth tended to be self-stabilizing.

5
Q. Kong et al. Finance Research Letters 38 (2021) 101488

Table 2
Co-integration test results.
Null hypothesis Trace statistic 5% critical value 1% critical value

0 co-integration vector 55.8260 47.85613** 54.6815***

1 co-integration vector 27.0124 29.79707 35.45817
2 co-integration vectors 9.8520 15.49471 19.93711
3 co-integration vectors 0.1220 3.841466 6.634897
Null hypothesis Maximum eigenvalue statistic 5% critical value 1% critical value
0 co-integration vector 48.8596 27.58434** 32.71527***
1 co-integration vector 22.9719 21.13162 25.86121
2 co-integration vectors 6.2669 14.2646 18.52001
3 co-integration vectors 0.0316 3.841466 6.634897

Note: collated by the author.

Table 3
The estimation results of the long-term and short-term model of ARDL.
Models Variables Coefficient Std. error T-test P-value

Long term model 1nTOP 0.7187** 0.3648 3.26 0.0450

1nREER 0.3048*** 0.3002 4.68 0.0269
C 0.6741*** 0.3280 2.97 0.0078
Short term model Δ1nTOP 0.8891** 0.2806 3.63 0.0114
Δ1nREER 0.4942*** 1.4528 3.47 0.0026
Δ1nFDI − 0.3377*** 0.5530 − 4.77 0.0023
ecm( − 1) − 0.3265*** 0.0432 − 3.76 0.0263

Note: (1) The subscript number in the variable represents the lag period, t-1 stands for lag phase 1 and t-2 stands for lag phase two. (2) ***,**, and *
2
indicate the level of significance of 1%, 5%, and 10%, respectively. (3) R = 0.9987, DW=2.0543.

4. Heterogeneity of opening-up and quality of economic growth

4.1. Regional heterogeneity analysis

Based on the above discussions, it clear that the studied variabls influenced each other in the long run. However, the quality of
economic growth in the current period was not only influenced by the degree of trade openness and other factors in the current period,
but also by the economic development level in the local area. In addition, since the data in this paper belongs to a short panel, Diff-
GMM has been adopted for the unbiasedness of sample statistics. In this section, samples were divided into three regions: eastern,
central and western2, to carry out a regional analysis. 2SLS and GMM methods were used to test and the results are shown in Table 4.
The results indicate that trade openness had a significant positive influence on the quality of economic growth in the eastern region,
but that this positive influence was not obvious in the central and western regions. One possible reason for this discrepancy is that the
coastal and eastern regions have well-established infrastructure and abundant physical resources and human resources, which can
offer a favorable external environment for trading, meaning that opening-up was mainly concentrated in the coastal and eastern region
and had a significant promotional effect on the quality of economic growth and formation of materials and capital in these regions.

4.2. Nonlinear characteristic analysis

In accordance with Tables 5 and 6, the relationship between trade openness and economic growth quality presents an “N-type”
characteristic trend on the whole, except for the central area (Xu Yingzhi and Guo Jin et al., 2014; Xu Jing and Meng Juan, 2015;
Zhang et al., 2020b). To be specific, when trade openness is low, opening up and economic growth quality are positively correlated;
when trade openness increases to the second threshold interval, opening up and economic growth quality are negatively correlated;
and when trade openness further rises, opening up and economic growth quality demonstrate a positive correlation once again. Hence,
increasing trade openness is a guaranteed method for improving China’s economic growth quality (Chen and Z, 2010). In addition, the
provinces and cities during the period of 1994-2018 are classified on the basis of different trade openness, as shown in Table 7. It can be
seen that from 1994 to 2018, the degree of trade openness of all cities in China increased steadily, the degree of trade openness of most
cities is getting higher and higher, and the quality of economic growth is also getting higher and higher. This proves the first

2
According to the division criteria published by the National Bureau of Statics, the eastern region included 11 provincial-level administrative
regions, that is, Beijing, Tianjin, Hebei, Liaoning, Shanghai, Jiangsu, Zhejiang, Fujian, Shandong, Guangdong and Hainan. The central region
included 8 provincial-level administrative regions, that is, Shanxi, Jilin, Heilongjiang, Anhui, Jiangxi, Henan, Hubei and Hunan. The western region
included 12 provincial-level administrative regions, that is, Sichuan, Chongqing, Guizhou, Yunnan, Tibet, Shaanxi, Gansu, Qinghai, Ningxia,
Xinjiang, Guangxi and Inner Mongolia. Due to the lack of partial data in Tibet, in this paper, finally, data from a total of 30 provinces (cities) were
selected as cross-sectional samples.

6
Q. Kong et al. Finance Research Letters 38 (2021) 101488

Table 4
Results of sub-regional regression.
Variables Nation East Mid West
2SLS GMM 2SLS GMM 2SLS GMM 2SLS GMM

1nECOt − 1 0.8915*** 0.8648***(4.45) 0.8896*** 0.7862*** 0.5035*** 0.6133** 0.7886** 0.7230**

(6.55) (5.14) (5.11) (3.57) (3.45) (2.04) (2.88)
1nFDI − 0.0026 0.0056 (0.45) − 0.0035** 0.0085 (1.34) 0.0019 (1.93) 0.0050 0.0024*** 0.0031***
(− 0.52) (− 2.42) (0.68) (2.87) (2.76)
1nFDIt − 1 − 0.0004 0.0014 (0.64) 0.0005 (1.52) 0.0077 (0.63) 0.0007 (0.44) 0.0059 − 0.0001 − 0.0004
(− 0.36) (0.87) (− 0.69) (− 0.32)
1nREER 1.6878*** 1.9300*** 4.5166*** 5.3731*** 2.3575 (1.94) 3.1709 0.8532*** 0.9571***
(2.83) (2.11) (4.14) (1.55) (1.29) (2.60) (2.15)
1nREERt 1.1450*** 2.2930*** 5.3594*** 4.0900*** 3.0819*** 2.2930 1.4103 (0.24) 2.1587 (2.45)
− 1 (2.74) (1.57) (2.42) (4.46) (2.19) (1.75)
1nTOP 0.4736*** 0.5511*** 0.6710*** 0.8981*** − 3.7469** − 2.5964 0.4736 (0.87) 0.5059 (0.70)
(2.39) (2.51) (4.14) (2.86) (− 1.24) (− 1.03)
1nFDIt − 1 0.8807 (0.78) 0.6826 (0.12) − 1.4736 − 0.8384 0.4103 (0.27) 0.5059 0.5989 (0.13) 0.6826 (0.62)
(− 0.56) (− 0.88) (0.79)
C 0.3335 (0.56) 0.4875 (0.36) 0.8782 (0.26) 0.5938 (0.52) 0.7821 (2.87) 0.9220 0.3992 (0.98) 0.2695 (0.75)
(1.48)
Wald test 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
AR(1) 0.4962 0.5054 0.6050 0.6926 0.7887 0.5051 0.8237 0.7171
AR(2) 0.6630 0.7037 0.0211 0.4848 0.7969 0.2134 0.9334 0.6099
N 403 403 170 170 112 112 121 121

Note: (1) The subscript number in the variable represents the lag period, t-1 stands for lag phase 1 and t-2 stands for lag phase two. (2) ***,**, and *
indicate the level of significance of 1%, 5%, and 10%, respectively. (3) AR (1) and AR (2) are Arellano and Bond (1991) tests for autocorrelation in
differences.

Table 5
Results of threshold effects regression.
Variables Nation East Mid West

1nFDI − 10.4484*** (− 10.53) − 11.6772*** (− 5.68) − 13.6158** (− 10.23) − 7.3842*** (− 6.27)

1nREER 0.8540*** (2.70) 0.9287*** (4.24) 0.7405 (1.31) 0.8868*** (2.82)
1n TOP 1 0.7492*** (2.79) 0.3693** (2.48) − 1.5513 (− 2.55) 2.1405** (2.48)
1n TOP 2 − 2.5111*** (− 4.21) − 2.1405*** (− 3.02) − 0.8654 (-0.44) − 0.5551 (− 1.72)
1n TOP 3 0.6892*** (2.62) 0.3711** (1.64) − 0.7597 (− 2.80) 1.0984*** (2.62)
C 2.9831*** (17.30) 5.8058*** (10.60) 2.1559*** (18.22) 1.7027***(7.66)
Control variables YES YES YES YES
R2 0.4534 0.7545 0.8532 0.6061
Prob > F 0.0000 0.0000 0.0000 0.0000
N 403 170 112 121

Note: ***,**, and * indicate the level of significance of 1%, 5%, and 10%, respectively.

Table 6
Sub-sample results with high, medium and low trade openness.
Variables Low trade openness Medium trade openness High trade openness
2SLS GMM 2SLS GMM 2SLS GMM

LnFDI − 11.7461*** − 12.6642*** − 12.1094*** − 9.6098*** − 12.5708*** − 13.2979***

(− 10.80) (-8.10) (− 5.99) (− 7.41) (− 11.44) (− 12.98)
LnREER 0.9222*** (2.21) 0.6849*** (2.73) 0.8588*** (3.50) 0.8922*** (2.53) 0.6425*** (2.93) 0.6749*** (2.62)
c 2.2403*** (17.41) 2.5703*** (11.56) 5.3600*** (15.39) 6.9419*** (14.32) 2.7528*** (16.45) 1.4141*** (10.15)
Controlvariables YES YES YES YES YES YES
R
2 0.6160 0.5569 0.7430 0.8577 0.8144 0.9145
Prob > F 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
N 372 372 132 132 96 96

Note: ***,**, and * indicate the level of significance of 1%, 5%, and 10%, respectively.

conclusion: trade openness can significantly improve the quality of economic growth.

5. Conclusions

An ARDL model and a threshold model have been applied in order to analyze the relationship between trade openness and

7
 Q. Kong et al.
Table 7
Sample spatial distribution of different trade openness.
Threshold 1994 2002 2010 2018
variable

Low trade Anhui, Jiangxi, Hunan, Guangxi, Hainan, Sichuan, Jiangxi, Hunan, Guangxi, Hainan, Sichuan, Jiangxi, Hunan, Guangxi, Hainan, Sichuan, Jiangxi, Guangxi, Hainan, Guizhou,
openness Guizhou, Yunnan, Gansu, Qinghai, Hubei, Guizhou, Yunnan, Gansu, Qinghai, Hubei, Guizhou, Yunnan, Gansu, Qinghai, Hubei, Yunnan, Gansu, Qinghai, Hubei, Xinjiang,
Xinjiang, Heilongjiang, Inner Mongolia Xinjiang, Heilongjiang, Inner Mongolia Xinjiang, Heilongjiang, Inner Mongolia Heilongjiang, Inner Mongolia
8

Medium trade Hebei, Shanxi, Jilin, Jiangsu, Zhejiang, Shandong, Hebei, Shanxi, Jilin, Jiangsu, Zhejiang, Hebei, Shanxi, Jilin, Shandong, Henan, Hebei, Shanxi, Jilin, Shandong, Henan,
openness Henan, Guangdong, Shaanxi, Ningxia, Liaoning, Shandong, Henan, Guangdong, Shaanxi, Guangdong, Shaanxi, Ningxia, Liaoning, Fujian, Guangdong, Shaanxi, Ningxia, Liaoning,
Fujian, Chongqing, Tianjin Ningxia, Liaoning, Fujian, Anhui Anhui Fujian, Anhui, Sichuan, Hunan
High trade Beijing, Shanghai Beijing, Shanghai, Chongqing, Tianjin Beijing, Shanghai, Chongqing, Tianjin, Zhejiang, Beijing, Shanghai, Chongqing, Tianjin,
openness Jiangsu Zhejiang, Jiangsu

Note: collated by the author.

Finance Research Letters 38 (2021) 101488

Q. Kong et al. Finance Research Letters 38 (2021) 101488

economic growth quality under exchange rate fluctuations. Based on the above empirical results, the following conclusions can be
drawn: Firstly, that there is a long-term stable co-integration relationship between degree of trade openness and quality of economic
growth. Moreover, trade openness significantly promotes the quality of economic growth in both short and long terms. Secondly, that
the positive effect of trade openness on economic growth quality is statistically significant, with obvious regional heterogeneity.
Thirdly, that an “N-type” relationship exists between trade openness and economic growth quality. The above conclusions show that
opening up is an effective means for improving the quality of China’s economic growth. This not only enriches the modeling and
discussion of China’s trade under the new growth theory, but also, to a certain extent, enhances the theory’s explanatory power.
Foreign trade will effectively improve the quality of economic growth by promoting capital formation, accelerating technological
progress, and increasing factor productivity. Therefore, the establishment of a comprehensive and open economic system is one of the
key measures necessary for promoting China’s economic growth in the high-quality development stage.

Author statement

I have made substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data
for the work. I have revised the work critically for important intellectual content; and I have approved the final version to be published.
I agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the
work are appropriately resolved.
All persons who have made substantial contributions to the work reported in the manuscript, including those who provided editing
and writing assistance but who are not authors, are named in the Acknowledgments section of the manuscript and have given their
written permission to be named.

Reference

Arellano, M., Bond, S., 1991. Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. Rev. Econ. Stud. 58,
277–297.
Abid, A., 2019. Economic policy uncertainty and exchange rates in emerging markets: short and long runs evidence. Financ. Res. Lett., 101378
Baharom, A., Habibullah, M. S., Royfaizal, R., 2008. The relationship between trade openness, foreign direct investment and growth: case of Malaysia.
Chen, J., Z., L., 2010. Technological innovation ability and the unbalanced regional distribution of FDI. Sci. Technol. Prog. Countermeas. 6, 26–30.
Wei, Does Ma, XS, Cao, Li, T, et al., 2019. Promote the regional economic growth? – An empirical analysis based on the panel data of countries along the belt and road.
Econ. Latit. Longit. 36, 64–71.
Emery, R.F., 1967. The relation of exports and economic growth. Kyklos 20, 470–486.
Hansen, B.E., 1999. Threshold effects in non-dynamic panels: Estimation, testing, and inference. J. Econ. 93, 345–368.
Helpman, E., Krugman, P.R., 1985. Market Structure and Foreign Trade: Increasing Returns, Imperfect Competition, and the International Economy. MIT press.
Hendry, D., 1987. Econometric methodology: a personal perspective. In: Bewley, T. (Ed.), Advances in Econometrics, 2. Cambridge University Press, Cambridge.
Hou, F., Fan, Y., Sun, G., 2005. Estimation of human capital stock in China. Bus. Rev. Nanjing Univ. 3, 27–54.
Huchet-Bourdon, M., Le Mouël, C., Vijil, M., 2018. The relationship between trade openness and economic growth: some new insights on the openness measurement
issue. World Econ. 41, 59–76.
Idris, J., Yusop, Z., Habibullah, M.S., 2016. Trade openness and economic growth: a causality test in panel perspective. Int. J. Bus. Soc. 17.
James, A., Aadland, D., 2011. The curse of natural resources: an empirical investigation of US counties. Resour. Energy Econ. 33, 440–453.
Keho, Y., 2017. The impact of trade openness on economic growth: the case of Cote d’Ivoire. Cogent Econ. Financ. 5, 1332820.
Kong, Q., Guo, R., Wang, Y., Sui, X., Zhou, S., 2019. Home-country environment and firms’ outward foreign direct investment decision: evidence from Chinese firms.
Econ. Model. 85, 390–399.
Robert, L., 1988. On the mechanics of economic development. J. Monet. Econ. 22 (1), 3–42.
Mensi, W., Shahzad, S.J.H., Hammoudeh, S., Hkiri, B., Yahyaee, Al, H., K., 2019. Long-run relationships between US financial credit markets and risk factors: evidence
from the quantile ARDL approach. Financ. Res. Lett. 29, 101–110.
Rodrik, D., 1988. Imperfect Competition, Scale Economies, and Trade Policy in Developing Countries, Trade Policy Issues and Empirical Analysis. University of
Chicago Press, pp. 109–144.
Romer, P.M., 1986. Increasing returns and long-run growth. J. Polit. Econ. 94, 1002–1037.
Wang, Y., Cao, X., Sui, X., Zhao, W., 2019. How do black swan events go global?-evidence from US reserves effects on TOCOM gold futures prices. Financ. Res. Lett.
31.
Xu, Y., Guo, J., Wang, J., 2014. Energy consumption, trade opening and economic growth. Financ. Trade Econ. 12, 99–110.
Xu, J., Meng, J., 2015. Trade opening, economic growth and human capital: research based on panel threshold model. World Econ. Stud. 6, 84–91, 128.
Yuan, Q., Wu, L., Zhang, P., 2019. Integration development of Yangtze River economic belt and R & D efficiency of high-tech industry. J. Quant. Tech. Econ. 4, 45–60.
Zhang, D., Du, P., Chen, Y., 2019a. Can designed financial systems drive out highly polluting firms? An evaluation of an experimental economic policy. Financ. Res.
Lett. 31.
Zhang, D., Du, W., Zhuge, L., Tong, Z., Freeman, R.B., 2019b. Do financial constraints curb firms’ efforts to control pollution? Evidence from Chinese manufacturing
firms. J. Clean. Prod. 215, 1052–1058.
Zhang, D., Guo, Y., Wang, Z., Chen, Y., 2019c. The impact of US monetary policy on Chinese enterprises’ R&D investment. Financ. Res. Lett.
Zhang, D., Guo, Y., 2019. Financing R&D in Chinese private firms: business associations or political connection? Econ. Model. 79, 247–261.
Zhang, D., Tong, Z., Li, Y., 2020a. The role of cash holding towards cleaner production in China’s manufacturing sectors: a financial constraint perspective. J. Clean.
Prod. 245, 118875.
Zhang, D., Zhuge, L., Freeman, R.B., 2020b. Firm dynamics of hi-tech start-ups: does innovation matter? China Econ. Rev. 59, 101370.