Влияние инвестиций в строительную отрасль на региональную экономическую безопасность: пример малых и средних городов Китая

Introduction

The building industry is a vital sector of the national economy and serves a fundamental role [1] (Jiang et al., 2007). Since the 1980s, China's construction industry has transitioned from rapid growth to stable growth [2] (Jin et al., 2001). With the rapid advancement of economic globalisation and regional urbanisation, infrastructure, real estate, housing, and transit construction are also progressing swiftly. In 2021, the overall production value of China's construction industry reached 29,307.9 billion yuan, sustaining a year-on-year growth rate of 10%. At the same time, the related labor force involved in the construction industry exceeded 55 million. Compared with other industries, the construction industry has an increasing impact on the whole national economy in terms of scale, employment and capacity characteristics, and has a strong linkage value with other industries from the perspective of market [3] (Wei et al., 2004).

Object of research

In the first three quarters of 2023, the output value of all types of enterprises in China's construction industry increased by only 1.79 percent compared with last year [4] (Ye, 2007). The overall growth rate has been slowed down by the difficulties encountered by some construction enterprises in survival and development problems. This is bound to pose a threat to the economic environment. At the same time, the development of different regions in China is uneven. Economic foundation, natural resources, policies and infrastructure development degree and other reasons, the development of the construction industry in various regions will also show great differences and imbalances [5, 6] (Zheng et al., 2011; Zhou et al., 2007). This difference widens the difference in investment levels, which is not conducive to the effective allocation of resources and the overall growth of the construction industry [7] (Yang, 2009). In the long run, the overall competition level of the construction industry becomes low, which has a negative impact on the development of national economy and the security of regional economy.

Purpose of the research

This paper focuses on the development of construction in small and medium-sized cities in China, and tries to explore the relationship between the development of construction, especially investment, and regional economic security in prefecture-level cities. In the face of economic threats, this paper focuses on the influence effect of construction industry investment on regional economy and the influence effect of small and medium-sized cities in various regions. It clearly and comprehensively understands the position of construction industry in regional economic development and the influence degree of construction industry investment in small and medium-sized cities on regional economic growth, so as to achieve healthy and sustainable development in the risk of construction industry growth slowdown. It provides a reference for formulating relevant policies on regional economic security to deal with greater economic threats and challenges.

Methods of research

1. Literature research

Through sorting out the relevant research literature in the field of urban construction industry investment, this paper summarizes the regional performance of construction industry investment and development, and puts forward a new research perspective. Through sorting out the relevant research literature on construction industry and economic security development, the research content is constructed, that is, the impact of construction industry investment performance on regional economic security.

2. Qualitative analysis

Before the quantitative analysis, the relationship between investment in the construction industry and economic security is first clarified through qualitative analysis. The paper analyzes the investment performance of construction industry and the components of regional economic development, constructs the path of mutual influence between construction industry investment and economic development, and conducts qualitative analysis on the mutual influence between the two variables.

3. Quantitative analysis

Data representing the investment performance and economic development of the construction industry were selected, and unit root tests were performed on all data to analyze the cointegration relationship between the two variables. Then select the model suitable for the correlation analysis of the two variables, based on the statistical data, the construction industry investment performance and economic development of small and medium-sized cities for quantitative analysis. To provide a scientific and reasonable basis for the research on the impact of construction industry investment performance on economic development.

4. Comparative analysis

Aiming at the correlation results between the economic performance of construction industry investment and regional economic development in 132 small and medium-sized cities, this paper makes a comprehensive and systematic analysis of the regional differences and dynamic changes in the time axis of the development status of construction industry and economic security in each region from the two dimensions of cross section and time.

Literature review

1. Economic security

Academic research on economic security has evolved through periods of warfare, oil crises, and financial crises. Prior to the 1970s, economic security was far less prioritised in national security compared to political and military security. Following the second oil crisis in 1978, economic security was deemed the cornerstone of national security. Former US President Roosevelt asserted that the government must guarantee economic and personal security to attain genuine freedom [8] (Hudson, 1996). In 1992, Clinton prioritised economic security in the US election and incorporated it into his foreign policy objectives [9] (Guo, 1993). The investigation of economic security in the UK is delineated by welfare levels and the state's authority over resources, monies, and markets [10] (Buzan, 1991).

After the 1997 financial crisis, the research on economic security in various countries has become more independent and specific. Due to its wide scope, economic security mainly focuses on the discussion of: strategy, economic power, maintaining economic level and guaranteeing people's livelihood. This mainly includes mastering economic sovereignty, being free from domestic and foreign threats and infringement, and having strong economic competitiveness in energy, food, industry, finance and other fields [11] (Cai, 2022). With the rapid development of the world economy and the acceleration of the process of world economic integration, the degree and influence of international economic cooperation are gradually deepening. The process of international flow and allocation of production factors is also gradually accelerating. Whether national economic sovereignty has been eroded is the main discussion on the impact of globalization. Discussions and doubts about the ability to manage their own economies have led to reflections on the legitimacy of globalization. With the expansion of market scale and economic organization, the legitimacy of the state is threatened and challenged because the state cannot provide sufficient public goods for its citizens [12] (Cerny, 2003).

In response to this threat and challenge, economic regionalism has become the opposite of economic liberalism. A certain region or group of countries will form a unified security unit on the threatened issue, and become a mutual aid group of regional organizations through inter-regional coordination and cooperation, so as to maintain the economic security of a region. Therefore, the definition of regional economic security is more micro. It mainly emphasizes the economic development capacity of a region, usually involving industry, finance, resources, ecology and other issues.

2. Construction sector and investment

Most countries in the world classify industrial economic activities according to the International Standard Industrial Classification (ISIC) formulated by the United Nations. According to ISIC4.0, category F construction is divided into three new categories: house construction, civil engineering construction and special construction activities. The construction industry, which plays a role of accommodation and connection, provides a wide range of contact and space for human society by producing construction products such as houses and infrastructure, and is an important part of the national economy. At the same time, the construction industry has established business and connections with many upstream industries through reinforcing bar, cement, concrete, sand and other important and heavily consumed raw materials. In addition to the expectation that the investment in the construction industry will stimulate the development of other industries to stimulate the national economy, the construction industry also contributes to the growth of the national economy in terms of the capacity of the labor market [13] (Li et al., 2007). This is because the construction industry has a relatively low technical level for labor due to its labor-intensive characteristics, so it can absorb the surplus rural labor in the process of urbanization. In the downstream industry, the consumption driven by construction is also stimulating the growth of the national economy.

On the other hand, the misallocation of social resources will lead to the appearance of oversupply of infrastructure. This is that the excess economic scale will lead to the negative impact of the construction industry on the economy [14] (Wang et al., 2009). Due to the expansion of the construction industry, high-cost inputs appear to unevenly distribute the supply of capital and resources to other industries, thus affecting macroeconomic stability and even causing economic regression. Generally speaking, it is difficult to realize the normal operation of construction projects if only relying on the capital of construction companies because of the high cost of investment characteristics. Therefore, many construction enterprises need to rely on multi-party financing to obtain sufficient investment [15] (Liu, 2017). If the funds borrowed by construction enterprises do not bring sustained and stable economic benefits to the project, once a large risk occurs, it is bound to lead to the shutdown of the project or even the bankruptcy of the company, and cause irreversible losses to the capital situation of enterprises and individuals in the upstream and downstream of the industrial chain, so it will cause damage to many industries.

This paper is focusing on the economic function of the construction industry, trying to study the relationship between investment in the construction industry and economic security, and by introducing the total output value of the construction industry in different regions, explore the regional similarities and differences, and provide policy suggestions for the healthy and sustainable development of the construction industry.

Empirical examination

1. Data selection and processing

In order to analyze the impact of construction investment on regional GDP, according to the City Size Classification Standard (2014) issued by the Chinese government, this study classifies cities with permanent urban residents between 500,000 and 1 million as medium cities, and cities with a population less than 500,000 as small cities. In 2016, China had 656 administrative cities, comprising 4 municipalities directly governed by the Central Government, 291 prefecture-level cities, and 361 county-level administrative cities. Of the 29 prefecture-level cities, 178 are classified as small and medium-sized towns, representing over fifty percent. This article examines 132 prefecture-level cities, each with a district population not exceeding 1 million as of the end of 2019. Logarithms are applied concurrently to the data sequence to mitigate the effects of heteroscedasticity. All data analyses were conducted with Stata 18 software.

2. Description of variables

2.1. Variables of economic security

The methodologies for assessing regional economic security are in the exploratory phase, and a consensus has yet to be established. The approaches are primarily categorised into direct construction methods and indirect building methods. The American academic Haner introduced the Fuland Index to evaluate a nation's political, economic, and risk landscapes. In 2014, Hucker et al. introduced the Economic Security Index to evaluate data pertaining to the US national economy [16] (Hacker et al., 2014). This article utilises the conceptual frameworks of economic security proposed by Alasana Ouattara and R. Daniel Wadwa in the selection of metrics, emphasising the comprehension of national economic growth, the quality of life, and the safeguarding of wealth [17] (Gu et al., 2015). Consequently, the gross national product of a region was selected as the metric for economic security. This paper employs a direct construction method to assess the economic security level of small and medium-sized prefecture-level cities in China, utilising data on the trends in gross national product (GDP) and its growth rate for visual evaluation. This can more effectively illustrate the variables of regional economic security.

2.2. Investment factors in the construction sector

The advancement of the construction sector is intricately linked to the nation's overall economic condition. This study utilises the aggregate completed investment in residential development (in ten thousand yuan) and the completed investment in real estate development (in ten thousand yuan) of 132 prefecture-level cities from 2010 to 2023 to signify the performance of construction industry investment in this region. This data typically represents the fundamental investment landscape of the construction sector. To mitigate mistakes in data processing, the natural logarithm LNCI is applied to the data series of construction industry investment for this variable.

The natural logarithm ln (GDP_it) of the gross national product (GDP_it) for small and medium-sized cities with populations under one million in the ‘i’ region during the ‘t’ year serves as the dependent variable, while the natural logarithm ln (CI) of construction industry investment performance (CI) for these cities in the same region and year acts as the independent variable, thereby forming a regression equation:

ln(GDP_it) = α_i + β_i·ln( CI_it) + μ_it.

Therefore, the time span of this study is 14 years, with 132 small and medium-sized cities as cross-sectional units and a sample size of 1,716 observations. Table 1 shows the descriptive statistics of the main variables. Table 2 shows the descriptive statistics of the split variables.

Table 1

Descriptive statistics of main variables (level values)

Descriptive statistics of differenced variables

2.3. Panel data unit root test

Numerous economic variables can be depicted through time series. A series is termed stable if its mean and variance remain constant throughout time. To prevent pseudo-regression and guarantee the validity of the estimation findings, the stationarity of the panel series is assessed. This work included three prominent panel unit root test methodologies – Levin-Lin–Chu (LLC) test, Im–Pesaran–Shin (IPS) test, and Fisher–ADF test – due to the limited effectiveness of conventional time series unit root tests. Table 3 presents the findings of the three testing methodologies.

Table 3

Summary of unit root test results for level variables

For the variable GDP, the LLC test statistic is −7.0033, and the p value is 0.0000. However, the statistic of the IPS test is 3.4555 and the p-value is up to 0.9997, indicating that the hypothesis of unit root cannot be rejected for this series. Meanwhile, the statistic of Fisher test is 122.2659 and the p value is 1.0000, which also supports the hypothesis of non-stationarity.

For the variable CI, the LLC test statistic is −28.2620 with a p-value of 0.0000, and the Fisher test statistic is 454.3264 with a p-value of 0.0000. The test results show that the series is stationary. However, IPS test cannot calculate valid results due to insufficient time dimension. Due to the different assumptions of test methods, the inconsistency of test results is common in panel data analysis. Therefore, considering the contradiction of the test results, the stationarity test after first-order difference is adopted. Table 4 presents the test findings subsequent to the first-order difference.

Table 4

Unit root test results for first-differenced variables

The findings of the first-order difference test indicate that the first-order difference representation of the two variables robustly refutes the unit root hypothesis across all tests. The p-value of all test results is 0.0000, indicating that the series after difference is a stationary process. This result confirms that the original variable is a first-order single integration process, which meets the premise of cointegration relationship analysis.

2.4. Examination of cointegration relationships in panel data

Upon verifying that the variable constitutes a first-order unintegrated sequence via the unit root test, this study employed the panel cointegration test method provided by Westerlund to ascertain the existence of a long-term equilibrium relationship between GDP and CI. The test outcomes are presented in Table 5.

Table 5

Westerlund panel cointegration test results

The Westerlund test utilises an error correction model, wherein the Gt and Ga statistical tests exhibit at least one section cointegration relationship, while the Pt and Pa statistical tests demonstrate cointegration relationships across all sections. The test results show that the p-value of the four statistics is 1.000, which strongly supports the null hypothesis that there is no cointegration relationship. All four statistics are greater than the critical values at conventional significance levels. Therefore, through the cointegration test of the variables, it can be seen that there is no long-term equilibrium relationship between the regional gross product (GDP) and the investment performance (CI) of the construction industry in small and medium-sized cities. It provides an important basis for model setting.

From the perspective of economic theory, the impact of construction investment in small and medium-sized cities on economic growth may be mainly manifested as short-term effect. As an important part of regional financial investment, the fluctuation of construction industry is often closely related to macroeconomic cycle, policy regulation and other factors. Therefore, the use of the difference model to analyze the short-term impact has sufficient theoretical rationality.

2.5. Model configuration design

According to the unit root and cointegration relationship test results above, there is no long-term equilibrium relationship between the gross regional product (GDP) and the investment performance (CI) of the construction industry in small and medium-sized cities with a population less than 1 million in China. Under the condition that there is no cointegration relationship, the traditional level value regression model may produce spurious regression, which will lead to the deviation of statistical inference results. Therefore, this study uses the first-order difference model to test the short-term dynamic relationship among the variables.

2.5.1. Configuration of panel model

Formulate a first-order differential fixed effects model:

ΔGDP_it = α + βΔCI_it + μ_i + ε_it.

In the model, α is a constant intercept term that signifies the influence of variables other from the performance of construction sector investment on regional economic development. α_i denotes the fixed effect component in the model, illustrating the disparities among different locations. β_i denotes the coefficient of variation of the explanatory variable, which fluctuates across several cross-sections. The coefficient β_i of the explanatory variable signifies the effect of regional construction investment on regional economic growth. Given that the model is in a double logarithmic format, this coefficient should be interpreted as the elasticity coefficient of CI concerning regional economic growth. A bigger coefficient indicates a more significant contribution of CI to regional economic growth. μ_it denotes the stochastic disturbance component.

To ascertain the precise sort of panel data model in this study, the Hausman test was employed to analyse the correlation among variables. The findings of the Hausman test are presented in Table 6.

Table 6

Hausman test results

The Hausman test results suggest a test statistic of 120.05, with a p-value less than 0.0001, signifying a robust rejection of the null hypothesis that the random effects model is superior. From a pragmatic economic standpoint, when the disparities among individual member units in the data may be considered as variances in the regression coefficients, the fixed effects model is justifiable. Consequently, a substantial link exists between the regional gross domestic product (GDP) and the investment performance (CI) of the construction sector in small and medium-sized cities. The fixed effects approach can yield consistent and unbiased estimation outcomes.

2.5.2. Implementation of the fixed effects model

Table 7 presents the outcomes of the correlation analysis of the variables utilising the fixed effects model.

Table 7

Fixed effects estimation results

At the 1% level, the estimate results of the fixed-effect model indicate that the coefficient of the core explanatory variable ΔCI is 0.201, which is considerably positive (t=3.57, p=0.000). Also, the coefficient is significantly positive. That investment in the construction industry has a strong short-term promoting influence on economic growth is demonstrated by the fact that the regional GDP will expand by 0.201 units for every extra unit of investment in the construction industry. This demonstrates that investment in the construction industry has a considerable effect on economic growth.

By correcting for individual effects, the model's inside R2 value of 0.0088 indicates that short-term changes in construction industry investment can account for 0.88% of the total GDP change. This is the case even after allowing for individual factors. Even if this value is somewhat low, it is still within a suitable range according to the difference model. This is due to the fact that the short-term swings of economic variables are frequently influenced by a combination of causes occurring simultaneously.

R² is 0.4382, indicating that individual disparities among cities account for 43.82% of the variation in overall GDP. This unequivocally substantiates the imperative of utilising the fixed effects model. The model's total F-statistic is significant at the 1% level, demonstrating that the model specification possesses statistical significance.

2.6. Analysis of heterogeneity

In order to deeply explore the complex mechanism of regional construction industry investment performance on economic growth, this study uses two dimensions of cross section and time to construct a fixed impact variable coefficient model. The purpose is to reveal the heterogeneity and dynamic evolution characteristics of the effects of construction investment and economic growth in different cities and in different periods. It provides theoretical data support for regional economic security and construction industry economic development.

2.6.1. Section appraisal

To investigate the regional differences in the investment effect of the construction industry, this study constructed a cross-sectional variable coefficient model as follows:

ΔGDP_it = α_i + β_i·ΔCI_it + ε_it.

In this context, αi denotes the fixed effect component of the model, while β_i signifies the variation coefficient of the independent variable. Table 8 shows the estimation results of city-specific coefficients.

Table 8

Estimation results of city-specific coefficients

The results show that there are significant differences in the coefficients across cities, with the distribution ranging from -6.754 to 3.068. The coefficients are significantly positive for 26 cities and significantly negative for 3 cities at the 5% significance level. This confirms that there is strong regional heterogeneity in the economic growth effect of construction investment.

Among the cities with significant positive effects, several cities exhibit extremely high investment multipliers. The coefficients of Benxi City (2.677), Chaoyang City (2.144), Jixi City (2.023) and Hebi City (2.066) all exceed 2.0, which means that for every 1 unit increase in construction investment in these cities, the regional GDP will increase by more than 2 units in the short term. In addition, Tieling city (1.184), Fuxin city (0.959) and Dandong city (0.769) also show strong investment driving effect.

On the contrary, Luliang City (−5.145) and Nanping City (−0.623) show a significant negative effect. The strong negative effect in Luliang City is of particular concern, indicating that construction investment in this region may have a serious problem of resource misallocation or have an obvious crowding-out effect on other industries.

From the perspective of geographical distribution, the cities with high efficiency are mainly concentrated in the northeast region and some resource-based cities in central China, while the cities with negative effect are scattered in different regions, which indicates that the local industrial structure, resource endowment and policy environment jointly affect the heterogeneity of construction investment.

2.6.2. Time estimation

To further investigate the temporal variations of the investment effect in the construction industry, this study estimates a time-varying coefficient model:

ΔGDP_it = α_i + β_t·ΔCI_it + ε_it.

In this context, αi denotes the fixed effect component of the model, whereas βt signifies the variation coefficient of the independent variable. Table 9 illustrates the dynamic trajectory of the economic growth impact of investment in the building sector.

Table 9

Dynamic evolution of construction investment effects on economic growth

The analysis shows that the investment effect of construction industry shows obvious policy sensitivity and cycle dependence. As the starting point of the observation period, 2011 (0.548) shows a mild positive effect; From 2015 to 2016, the coefficient increased significantly from 0.395 to 0.759, reflecting the effect of the policy of stabilizing growth at that time; The coefficient reached an all-time peak of 1.640 in 2021, highlighting the key role of the construction industry in the process of economic reconstruction after the pandemic.

On the contrary, both 2020 (−1.017) and 2023 (−0.456) show a significant negative effect. The strong negative effect in 2020 clearly reflects the severe impact of COVID-19, while the continued negative effect in 2023 May indicate the pressure of structural adjustment in the construction industry in the post-epidemic era.

It is worth noting that the coefficient is statistically insignificant during 2012–2014 and 2017–2019, indicating that the economic growth effect of construction investment in these periods is relatively weak or unstable.

Table 10 summarizes the distribution characteristics of the coefficients of the cross-section and time dimensions. The standard deviation of the urban cross-section coefficient is 1.624, which is much higher than the 0.698 of the time series coefficient, indicating that regional differences are the main source of heterogeneity in construction investment. In addition, 65.2% of the cities have positive coefficients, but only 19.7% of the cities have statistically significant coefficients, which further confirms the conditionality and complexity of the investment effect of the construction industry.

Table 10

The distribution of the coefficients of the cross-section and time dimensions

Conclusion and policy recommendations

Based on the construction industry investment and regional GDP of small and medium-sized cities with a population of less than 1 million in China, this study examines the impact, heterogeneity and dynamic evolution characteristics of regional construction industry investment and economic development through unit root test, cointegration test, benchmark regression and fixed impact variable coefficient model. This chapter aims to summarize the conclusions of empirical research, put forward corresponding policy suggestions and point out future research directions. Although the fixed effect model and the variable coefficient model control the impact of individual heterogeneity, there may be a bidirectional causality between construction investment and economic growth, that is, economic growth will in turn lead to more construction investment. Although the difference model alleviates this problem to a certain extent, future research can try to find effective instrumental variables or use natural experiments and other methods to identify the causal relationship more strictly. To provide theoretical and data basis for future construction industry investment and regional economic security.

The empirical analysis results of this study are mainly as follows:

(1) The investment development of the construction industry is highly related to the economic growth of small and medium-sized cities, which has a significant role in promoting regional economic development in the short term.

According to the estimation results of the benchmark fixed effect model, the coefficient of the investment performance of the construction industry (ΔCI), the core explanatory variable of the study, is 0. 201 and is significant at the 1% level. This means that if the construction investment in small and medium-sized cities does not increase by one unit, the GNP of the region will increase by 0.201 units. However, the within R² of the model is 0.0088, which means that after controlling the individual fixed effects, the short-term fluctuation of construction investment can only have an impact of 0.88% on the GDP. In other words, although there is a correlation between construction investment and regional economic growth in small and medium-sized cities, economic development is affected by a variety of factors, and construction investment only plays a partial role.

The cross-sectional estimation results of the model show that there are significant differences in the coefficients of the 132 selected small and medium-sized cities, and the distribution range is between -6.754 and 3.068. The coefficients are significantly positive for 26 cities and significantly negative for 3 cities at the 5% significance level. This huge difference may be closely related to the industrial structure, resource reserve, urbanization stage and investment efficiency of different cities. The difference Between individuals (R²) is 43.82%, which strongly confirms that the individual characteristics of the city are the key to affect the construction industry economy.

The time estimation results of the model show that the investment coefficient reaches the peak of 0.759 and 1.649 in the policy transition period of 2015–2016 and the post-pandemic recovery period of 2021, respectively. However, in the period of COVID-19 impact in 2020 and structural adjustment in 2023, the coefficient is significantly negative. This shows that external shocks and internal structures have weakened the driving effect of the construction industry on the economy.

Therefore, from the perspective of region and time, the investment development of construction industry is highly related to the economic growth of small and medium-sized cities, and it has a significant role in promoting regional economic development in the short term.

(2) The impact of construction industry expansion on regional economic growth significantly differs across various regions, with a pronounced effect observed in northeast China.

From the estimation results of the model, it can be seen that there are great differences in the contribution of construction industry development to regional economic development in 132 small and medium-sized cities in China. Benxi (2.677), Chaoyang (2.144), Jixi (2.023) and other northeastern regions with strong investment multiplier effect show positive results. However, some cities in Southeast China, such as Luliang (−5.145) and Nanping (−0.623), show a negative effect. It can be seen that there is a big gap in the development of the construction industry in different small and medium-sized urban areas in China. From the perspective of the process of regional economic development, due to the significant differences in natural conditions, transportation facilities, ideology and culture, the original economic foundation and original capital of each region are also different. Therefore, the great difference in the speed of economic development in various regions leads to the unbalanced development of the construction industry.

This paper proposes the following policy recommendations to improve investment efficiency and foster coordinated regional development, based on the framework of the empirical analysis:

The government offers pertinent industrial support and advantageous measures to regions with underdeveloped construction sectors. Suitable support measures must be implemented for firms in regions with comparatively sluggish construction sector growth to maximise the guiding and incentivising effects of fiscal investment. The municipal administrations of small to medium-sized cities exhibiting relatively high effect coefficients should maintain focus and extend policy support. This aims to promote the collective growth of many regions through the diffusion effect and attain coordination and interaction among them. In cities with a comparatively low effect coefficient, the government ought to augment infrastructure investment to attract greater capital and technology resources, thus fostering comprehensive development of the regional economy and society.

Digital transformation of the construction sector in small and medium-sized cities within the context of the digital economy. Regions with a substantial influence on regional economic growth should concentrate on advancing intelligent construction and building industrialisation. Incorporate BIM technology, big data, and other artificial intelligence innovations with local competitive industries to enhance regional economic growth. For the areas with weak and negative effects, the Internet platform of construction industry should be used to optimize resource allocation and control inefficient investment. At the same time, the digital level will be incorporated into the industry evaluation system, so that the construction industry can develop to the quality and efficiency, and finally form a new pattern of sustainable development coordinated with regional economy.