Nonparametric Cointegration Tests for Price Convergence within the Greater Bay Area of China

Abstract

The Chinese Government plans to create a world-class city cluster in the Guangdong-Hong Kong-Macau Greater Bay Area (GBA). The GBA is a very attractive destination for both national and international businesses to profit in trading and investment activities. To realize the anticipated profit potential, existence of purchasing power parity (PPP) is a necessary condition for a favorable prospect of closer economic cooperation and integration in the GBA city cluster. Research study that investigates the PPP relationships among the GBA cities is scarce. To fill this research gap, our study aims to examine validity of the PPP and the extent of goods market integration in the GBA. Since the functional form of the cointegrating relationship may not be exact or linear, we adopt the rank tests for analysis without prior knowledge and specification of the functional form. We also address the rank problems that occur in multivariate rank tests. Results of the rank tests confirm validity of the PPP relationships among the GBA cities with some empirical evidence of nonlinearity that clear the doubt concerning the potential barriers to goods market integration in the GBA. There are policy implications for the local and central authorities in justifying further integrated development programs.

Keywords:

• Greater Bay Area;
• price convergence;
• nonparametric rank tests;
• rank problem

JEL CLASSIFICATION:

• C10
• F41

Notes

1 The data for the comparison of the GBA with the world’s other major Bay Areas come from Yu (2021).

2 There are some current studies on the GBA such as Chen et al. (2020) and Choi et al. (2021), but not for PPP testing.

3 In Equation (5), the general functional form for Model B, is ${\mathrm{g}(\mathrm{p}}_{\mathrm{t}}$) = $\mathrm{f}\left(p_t^{*}\right)+$ ${\mathrm{u}}_{\mathrm{t}};$ for Model C, ${\mathrm{g}(\mathrm{e}}_{\mathrm{t}}$) = $\mathrm{f}\left(p_t-p_t^{*}\right)+$ ${\mathrm{u}}_{\mathrm{t}};$ and for Model D, ${\mathrm{g}({\mathrm{p}}_{\mathrm{t}}-\mathrm{ }\mathrm{e}}_{\mathrm{t}}$) = $\mathrm{f}\left(p_t^{*}\right)+$ ${\mathrm{u}}_{\mathrm{t}}.$

4 In Equation (10), the general functional form for Model A is ${\mathrm{g}(\mathrm{e}}_{\mathrm{t}}$) =$\mathrm{ }{\mathrm{f}}_1\left(p_t\right)+\mathrm{ }{\mathrm{f}}_2\left(p_t^{*}\right)$+ ${\mathrm{u}}_{\mathrm{t}}.$

5 Specifications of Equations (12) and (13) can be referred to Li (2006).

6 HKD is pegged to USD at the rate of USD 1 = HKD 7.80 and MOP is pegged to HKD at the rate of HKD 1 = MOP 1.03 under the Linked Exchange Rate systems of Hong Kong and Macau, respectively. RMB can move within a band every day against a basket of foreign currencies, including USD, under the managed float system. The details are found in the websites of the Hong Kong Monetary Authority (https://www.hkma.gov.hk/eng), the Monetary Authority of Macau (https://www.amcm.gov.mo/en) and the People’s Bank of China (http://www.pbc.gov.cn/english/130712/index.html).

7 We have also selected Shenzhen as the base region for cointegration analysis. The results (unreported) are qualitatively equivalent, and they are available upon request.

8 Haug and Basher (2011) obtained a similar contradictory result of the Johansen Trace test in our restricted Model C, and they abandoned the result.

9 From Table 5, when the negative value of b₂ rises to -0.5, the power would be greater than 50% in the case of a positive b₁. However, according to PPP, the absolute values of the parameters should be close to, although not necessarily equal to, unity.

10 In Equation (10), the functional form for Model A’ becomes ${\mathrm{g}(\mathrm{p}}_{\mathrm{t}}$) =$\mathrm{ }{\mathrm{f}}_1({\mathrm{e}}_{\mathrm{t}})+\mathrm{ }{\mathrm{f}}_2\left(p_t^{*}\right)+$ ${\mathrm{u}}_{\mathrm{t}}.$

11 Before the Shenzhen-Hong Kong Stock Connect was implemented in 2016, the Shanghai-Hong Kong Stock Connect was introduced in 2014. These two Stock Connects are performing similar functions for providing mutual stock market access between Mainland China and Hong Kong.

Additional information

Funding

The work was supported by the University Research Grant (URG) of the Hong Kong Shue Yan University (URG/19/01).