About the presentation:

This presentation offers a comprehensive and quantitative analysis of structural transformation in Asia-Pacific Least Developed Countries (AP LDCs) within both a global and developing countries context. This exercise sheds light on the critical challenge of conceptualising structural transformation and measuring it to conclude about the progress being made in AP LDCs. The presentation examines four measures of structural transformation, sectoral share (agriculture and manufacturing) of GDP, urbanisation and Economic Complexity Index (ECI) on country fundamentals and institutional factors. By conducting a comparative analysis which involved panel random effect estimation method, the study outlines that structural transformation in Asia-Pacific LDCs is not always in line with what has been experienced by today’s advanced and developing nations.

Question and Answer Session

Question: Why is the services' share of GDP not incorporated as a measure of structural transformation?

Response: In most countries, the services sectors are not developed and less progressive. Moreover, productivity growth in services is relatively lower in LDCs compared to many other countries. Although services are significantly important globally, this trend is not necessarily a good thing. It is much easier to trade physical items which tend to have higher returns to scale.

Question: Why is employment share not selected as a measure of structural transformation?

Response: The paper uses four of the most widely adopted measures of structural transformation, one of which, based on existing literature, emphasises output share rather than employment share. Moreover, data on employment by sector are also not uniformly available for all countries. There is much lower consistency in the construction of employment data across a large set of countries, which might lead to comparability issues.

Question: How urbanisation is linked to structural transformation?

Response: Urbanisation dynamics worked as a vehicle for economic transformation. Moreover, the relationship between the level of urbanisation and per capita income is positive. Further, productive cities are considered an engine of economic growth where high-value-added crucial sectors have replaced low-value-added agricultural sectors. Such a transition can be linked to defining structural transformation.

Question: What is the reason for including the square of GDP per capita?

Response: The paper uses GDP per capita and its squared term to address potential non-linearity issues. The relationship between the dependent and independent variable might not always be linear (straight line), but rather it might be curve-shaped. Therefore, to control this issue the paper takes the square of GDP PC.

Question: Why is random effect estimation instead of fixed effect estimation conducted for empirical purposes?

Response: For empirical analysis and to focus on our objective, the model includes country dummy variables along with other variables in the model. Fixed effect model controls for all time-invariant characteristics of all entities. For each group, the dummy variable takes 0 or 1. In this case, fixed effect implies within transformation and that dummy gets omitted. Therefore, the estimation procedure employed panel random effect method.

Comments and feedback:

Instead of panel random effect estimation procedure, there are other econometric approaches such as difference in difference (DID), least square dummy variable (LSDG) along with year fixed and regional fixed effect might be useful to make the empirical analysis more robust. Analysis of descriptive statistics would be more appropriate to understand and make valid comparisons or differences among country groups. In addition to comparing AP LDCs with global and developing countries, the analysis should also include comparisons with other least-developed countries (LDCs).