Do Real Interest Rates Influence Economic Growth in China? An Econometric Analysis (1980–2024)

An Econometric Analysis of the Impact of Real Interest Rates on GDP Growth in China (1980–2024)

Results and Discussion

Section A: Distribution and Dependence Analysis

1. Dataset Description

The data utilized in the analysis will include 45 annual observations of China between the year 1980 and 2024, which will display a long-term consideration of the macroeconomic indicators of China. There are four important variables of it: the calendar year, it is represented as an integer; real interest rate (%) is the cost of money with inflation factor, that is, it will be the cost of borrowing which indicates the monetary policy position; GDP (current US), the market value of all commodities and services generated in the economy and log(GDP) which is added so that it can normalize distribution of GDP values and on the skewness there is minimizing of the skewness of GDP variables to be used in statistical modeling. The data have been mostly sourced by sources that are supplied by the users but ensured by cross-verification with some authorities’ sources such as World Bank Open Data and publications by the People Bank China, making the data reliable and accurate. The data will provide strong analyses of distributional aspects, correlation patterns and possible relationships between interest rates and the economic output during a time period as nominal and transformed measures of GDP have been included. This model is especially appropriate in time-series econometric research, where it will be possible to examine the dynamic relationship between monetary policy indicators and macroeconomic growth in China and obtain evidence on how fluctuations in interest rates can contribute to the long-term economic performance (Eniola, 2024).

2. Descriptive Statistics

Table 1: Detailed Descriptive Statistics for Real Interest Rate (%) and log(GDP)

Source: Stata output (summarize, detail)

Real interest rate is between -8.01 and 7.38 with a small amount of skewness (Left skewness) of -0.73 which displays more to get the moderately high and low rates, which is in line with periods of policy credit expansion. The variable of log(GDP) has a right skewness (0.08) and low kurtosis, which is an indication that the growth is relatively stable throughout the periods of the sample.

3. Summary Statistics (Mean, SD, Range)

Table 2: Summary Statistics (Mean, SD, Variance, Min, Max)

The small mean interest rate in real terms (c. 2) and the large dispersion imply that the policy environment fluctuated between tightening and relaxing stages. The long-term growth in the economy is seen in the form of the pronounced positive GDP growth (mean log =28.3).

Figure 1: Plot of univariate empirical distributions

Figure 2:  Copula estimation results

Figure 3: Scatter plot of the integral transformed data

4. Correlation and Covariance

Table 3: Covariance and Correlation Matrix

The Tau (τb = 0.0596, p = 0.5704) value by Kendall is another test supporting a statistic insignificant correlation between interest rate and GDP. It would mean that GDP growth and interest rates vary independently of each other in the same year, which is in line with the research that indicates slow policy transmission effects (Chen and Liu, 2020; Bank of England, 2021).

5. Normality Tests

Normality was tested using Shapiro–Wilk and Skewness–Kurtosis (sktest) tests.

Table 4: Normality Tests for Real Interest Rate and log(GDP)

The real interest rate distribution approximates normality, while log(GDP) significantly deviates, indicating a need for transformation in parametric modeling (Patton, 2021).

Section B: Regression Analysis

1.     Regression Analysis

A simple linear regression model was estimated:

Table 5: OLS Regression Results for log(GDP) on Real Interest Rate

Model statistics: F(1,43) = 0.25, p = .619, R² = .0058, Adj R² = −.0173, Root MSE = 1.6032, AIC = 172.14, BIC = 175.75.

The slope coefficient (= 0.038) shows that an increase in the real interest rate by one per cent leads to an insignificant rise of 0.038 AD in the log(GDP), which is statistically insignificant at p = 0.619. Therefore, the contemporaneous effect of movement in the short run rates on GDP is very limited.

2.     Regression Diagnostics

Table 6: Diagnostic Tests for Model Adequacy

Despite the non-normal residual, homoskedasticity and specification exists. The tails of the distribution are heavy and this could be brought about by macroeconomic shocks that could not be modelled. However, rigorousness tests indicate that the model is not mis specified.

Section C: Discussion

The correlation between the actual interest rate of the Chinese and GDP of 1980-2024 was evaluated. The descriptive statistics (Table 1) shows that, the descriptive statistics of the real interest rate is an average of 2.13 with -8.01 to 7.38 range and skewness is an average of right. The trend shows that relaxed and contractive actions of the policy pursued are driven by ancient monetary interventions in China (Jonas, 2025). The average of the logarithm of the GDP stands at 28.26, which shows that the economy experiences the positive growth in the course of analysis. The positive skewness (0.08) is very minor and takes the form that despite the overall stability of the growth there are cases that had high-growth years which are the ones that offered the long-term direction.

Incorporating correlation and dependence tests, it implies that the relationship between real interest rates and the GDP are not very strong ( 0 = 0.0596, p = 0.5704), which presupposes that the changes in the interest rates are not necessarily aligned with the corresponding alteration of the GDP. The covariance analysis (Table 3) validates this observation as these variables have a slight co- movement. The findings are correlated with the earlier research that outlines the monetary policy in China as the one with high time-lags, thus decreases its influence on the results of the macroeconomic activity in the short-run (Ramdane et al., 2025).

Based on the tests of normalcy (Table 4), the real interest rate is near normally distributed, however, the log (GDP) is strongly reported not to be normally distributed. Non-normality GDP means that there are really extreme values and there are temporary high growth rates that may be caused by economic shocks, policy interference and structural changes (Eniola, 2024). Such deviations point to the necessity to apply potent statistical tools whenever estimating macroeconomic time series, not to get the bias estimations.

According to Table 5 outcomes of the linear regression, the real interest rate is positively, yet, insignificantly and unimportantly, affecting the log (GDP) =( =0.038, p =0.619). The value of the R 2 shows only 0.58 which justifies the fluctuation in the GDP. The point that follows this observation is that the alteration in the real interest rates in the short run creates no significant impacts to an economic output. This is in line with the existing literature that argues that economic growth in China is highly founded on investment growth, productivity and export-oriented growth as compared to monetary adjustments in the short-term (Ramdane et al., 2025).

The validity of the model is demonstrated with the help of the diagnostic test (Table 6), where the significant omitted variable bias (Ramsey RESET, p = 0.402) and the presence of multicollinearity (VIF = 1.00) have not been observed. The BreuschPagan (p = 0.583) gives the homoskedasticity support. These residues are not however normally distributed by indicating the presence of outliers or non-linearity in the GDP growth. All of this highlights the shortcomings of naive OLS models to model complex economic systems in the emerging economies (Harb, 2017).

The paper depicts that short-run fluctuations in interest rates do not inform the GDP in the Chinese context. The policymakers will have to consider the long-term structural factor, fiscal policy and external economic conditions in estimating the growth prospects. It is also possible to support the latter idea by the fact that there is a weak correlation between monetary policy transmission in China on the one hand and the presence of several intervening variables on the other hand, including the banking regulation, access to credit, and dynamics of international trade (Jonas, 2025).

References

Eniola, A. B. (2024). The Impact of Government Spending (2008-2024) And The 14th Five-Year Plan (2021-2025) On the Economic Growth of The People’s Republic of China. https://uniselinus.education/sites/default/files/2025-03/ADEBAYO%20BUSOLA%20ENIOLA.pdf

Harb, N. M. A. (2017). A Study on the non-linearity hypothesis between various macroeconomic variables and economic growth in developing countries (Doctoral dissertation, University of Leicester). https://figshare.le.ac.uk/articles/thesis/A_Study_on_the_Non-Linearity_Hypothesis_between_Various_Macroeconomic_Variables_and_Economic_Growth_in_Developing_Countries/10216853

Jonas, A. (2025). Determinants of China’s Gdp Growth: An Empirical Analysis of Macroeconomic Variable. Management, 15(1), 20-35. https://elibrary.ru/item.asp?id=80459422

Ramdane, M., Hamid, R., & Ahmed, T. (2025). Analyzing the Dynamic Relationship between Money Supply (M2), Nominal Interest Rate, and Economic Growth in Algeria: An Empirical Investigation Using the ARDL Approach (2000–2024).  Doi.org/10.56352/sei/8.12.11