Measuring neighbouring effects on poverty in regional economies

ABSTRACT

Using the concepts of poverty trap and neighbouring effect, this paper explores whether the economies of low-income regions and high-income regions respond differently to national economic fluctuations. It identifies an asymmetric pattern of regional income change in low-income regions of Korea due to a relatively small population size and weak urbanization economies. These low-income regions could increase their income levels by 3.75% if they succeeded in developing a symmetric relation to national economic fluctuations. Spatial proximity between high- and low-income regions can generate greater backwash (negative) effects on the economic growth of low-income regions than spread (positive) effects.

KEYWORDS:

• Regional income disparity
• asymmetric regional income change
• neighbouring effect

JEL CLASSIFICATION:

• O18
• R11
• R58

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ Regional resilience is defined as the adaptive ability to resist and renew an economic growth path in response to economic shock (Cowell 2013; Martin 2012).

² On the national level, Kose, Otrok, and Prasad (2008) and Fidrmuc and Korhonen (2010) analyse business cycle synchronization and convergence among nations.

³ Generally, in Europe, poverty is defined as the percentage of individuals living in a household whose equivalent income is below the poverty threshold (equal to 60% of the national median equivalent income), while in the US the poverty line depends on family size and the age of the household members (Betti et al. 2012).

⁴ Income inequality and poverty have been measured by the James-Stein income estimator of Fay and Herriot (1979), the multi-dimensional estimation of Alkire and Foster (2011), and the systematic reported survey approach of Piacentine (2014).

⁵ Kang, Kugler, and Kugler (2011) argue that capital infusion from foreign or domestic sources and the saving rate are critical to recovery, based on the dynamics of capital accumulation in the neoclassical economic growth model.

⁶ Those are 1) an increase in α under a fixed δ, 2) a decrease in δ under a fixed α, 3) an increase in α with a decrease in δ, and 4) fixed α and δ.

⁷ The source of annual economic data are Statistics Korea (2018), and the total number of observations is 2688 (244 cities × 12 years).

⁸ The regional income model focuses on high-income and low-income regions, as defined above. HH regions are regions where the regional income level (GRP) and regional income growth rate were higher than the national average (3.25 billion USD and 0.55, respectively) from 1998 to 2009. LL regions are regions where regional income and income growth rate were lower than the national average.

⁹ Especially, Stimson, Stough, and Roberts (2006) mention that regional economic growth depends on physical capital, human capital, infrastructure, R&D and knowledge-spillover, public finance, spatial structure, population growth, and agglomeration economies from industrial structures and region [city] size in the neoclassical growth model.

¹⁰ The concentration of the manufacturing sector is measured by the location quotient of manufacturing employees.

¹¹ The index of industrial diversity in each region is calculated from D_i = 1 – Σ |(L_i,j/L_i) – (L_j/L)|, where i, j represent, respectively, region and industrial sector. L represents total employees in the nation.

¹² If the regional income growth rate is lower than the national average, the value is set to 1.

¹³ Lee, Kim, and Sim (2014) argue that the wage rate is more elastic in response to contract type than to skill intensity, and average wage rate could be one proxy variable for measuring the quality of labour input.

¹⁴ The neighbouring effect has received much attention because it provides a way of understanding why poverty traps exist. For example, if initially a community was exclusively comprised of poor members, it would remain poor over long time periods because capital could not be accumulated through the low savings caused by the vicious spiral among the poor members (Durlauf 2004; Wu, He, and Webster 2010).

¹⁵ The average ratio of capital stock of low-income region to high-income region is 7.29%.

¹⁶ Birnie and Hitchens (1998) explain that the growth of the manufacturing sector led to the convergence of Gross Domestic Product per capita among the peripheral countries in the European Union.

¹⁷ The sectoral variety might be able to avoid economic shock and minimize negative effects on the regional economic measures such as output and employment (Hill, Wial, and Wolman 2008; Simmie and Martin 2010). For example, Bailey and Berkeley (2014) consider the West Midlands in the UK an example of vulnerability to recession because it lacked industrial diversity and had poor economic growth, a relatively low skill level, and a low level of innovation compared with other UK regions.

Additional information

Funding

This work was supported by the National Research Foundation of Korea grant number [NRF-2017S1A3A2066771].