The growth experience of developing economies : a comparative study of Hong Kong, Japan, Korea, Singapore, and Taiwan

The present study is an empirical analysis of the growth experience of five fast growing Asian economies, viz. Hong Kong, Japan, Korea (the Republic of Korea or South Korea), Singapore, and Taiwan in the light of the existing growth and development theory. This study is largely inspired by the existence of an important gap between theories and facts in development economics. Theories and facts seem to have grown largely independently of each other in this field of study. Besides testing many of the existing empirical hypotheses in growth and development theory, attempts have also been made in this study to construct new hypotheses and reformulate some of the existing hypotheses. Notably, we have constructed new testable endogenous technical progress hypotheses, a simultaneous-equation model relating growth to domestic saving, and reformulate the conventional type of export-led growth models. The focus of the present study is on the morphology, causes, and effects of economic growth in the selected Asian economies. We begin with a presentation of the contours of economic growth so that we can establish some bases for our later empirical analysis. The trends of growth in output and factor inputs are first examined. We observe that during the period under study 1955-70, all selected economies experienced exceptionally high growth rates of above 7/» i n real output and above 4.70 in real income per capita. Indeed, during- the more recent sub- period of 1966-70, growth rates had been above 10% (Kong Kong was the exception owing to the 1966-67 political disturbances). Population growth in Japan was stable at the 1% level throughout the period under study. On the other hand, population growth in the four other economies was at first relatively fast but slowed down in the later sub-periods. At the same time, the growth of the labour force in these four economies was also very rapid, due largely to the rapid increases in the labour participation rate. Generally speaking, during the period 1955-60, population grew faster than the labour force bu^ vice versa for the period 1960- 70. This rapid increase in manpower resources played an important part in supporting the growth of the industrial sector. There was also a rapid growth in investment resulting in rapid growth in capital input for the economy. The high rates of growth in investment in these economies were due to the massive foreign capital inflows and/or the high saving rates. The rapid growth in output and factor inputs in these economies was accompanied by equally rapid structural changes in the sectoral shares of total product and employment. In general, we witness a drastic shift of resources from the primary sector to the secondary and the tertiary sectors. For the five economies as a group, our estimation results indicate that the per capita income level of US$1100 is an interesting dividing line between two different phases or levels of growth and development. Before such a per capita income level is attained, our results suggest drastic declines in the share of the primary sector and significant increases in the share of the secondary and the tertiary sectors. Beyond the US$1100 level, the rate of changes in sector shares slows down and the share of the secondary sector shows signs of slow decline. If the two city economies, Hong Kong and Singapore are taken by themselves, then even beyond the US$1100 per capita income level, the share of the secondary sector continues to rise while the share of the tertiary sector starts to decline slowly. The next important task in our study is to identify the various possible sources of economic growth and examine their relative importance. Specifically, we tried to identify such sources of growth as scale economies, factor substitution elasticities, rate of technical progress, and growth in factor inputs. For this task, we used the aggregate production function approach. We admit that there are various estimation and aggregation problems associated with this approach but we would like to use it in a constructive manner to generate useful empirical results. In view of the fact that the economic characteristics of different sectors in these economies (notably the agricultural and the non- agricultural sectors) can be very different, we estimated the production function of the agricultural and the manufacturing sectors separately in addition to the estimation of the whole economy. Specifying the Cobb-Douglas and CES production functions and using alternative estimation methods, we find that economies of scale were relatively unimportant in explaining growth in the economies under study, while technical progress is found to be a crucial factor. However, in the cases of Hong Kong, Singapore, and Korea, the estimated rates of technical progress (2-3% per annum) are relatively low, implying that there were other equally important factors contributing to growth. This result is in some ways different from the Solow-Denison findings for developed nations which indicated that technical progress was the single important factor contributing to growth. The part played by the growth in capital per worker in explaining productivity growth is less clear, as the regression coefficients are statistically significant in some cases but not in others. The estimated elas- ticities of substitution take a wide range of values. In using the profit-maximization equation, the estimated elasticities range from 1.156 for Hong Kong manufacturing to 0.169 for Korean Agriculture. In many cases, the estimates are high and statistically significant indicating the importance of the ease of factor sub- stitution in productivity growth. Using an alternative approach of the national income accounting method, we analysed the contribution to economic growth of factor inputs relative to total factor productivity (which is a somewhat broader term than technical progress). We find that changes in factor inputs accounted for about one-half of income growth in our group of economies for the period under study. It is of interest to note that this result lies somewhat between the existing findings for developed nations on the one hand and developing countries on the other. Confining ourselves to the contribution of factor inputs to growth, we find that for the economy as a whole the contribution of capital to growth is generally greater than labour. We also attempted to quantify the contribution to growth of reallocating resources from the agricultural to the non-agricultural sectors. It is found that the gains from reallocation was substantial in the cases of Japan. Korea, and Taiwan where the agricultural sector is relatively large. To take a more realistic view of technical progress, we assumed that technical progress is partly endogenously determined and constructed some testable hypotheses of endogenous technical progress. Our estimation results show that learning by doing was an important determinant of technical progress in Japan, and it also had some effects on technical progress in Singapore but relatively little effect in the cases of Hong Kong, Korea and Taiwan. It seems that the importance of learning by doing has something to do with the level of technology already attained. Furthermore, it is found that investment activities were an important determinant of technical progress in Japan but not in the four other economies under study. Further investigations reveal that in the cases of Hong Kong, Singapore and Taiwan, it is the importation of foreign technology that has been the principal determinant of technical progress. In the case of Korea, technical progress was largely related to foreign aid directed to the manufacturing sector, which is surely just another form of importation of foreign technology. In view of the fact that all economies under study have a relatively large foreign trade sector, we therefore examined the role of foreign trade and capital inflow as causes of rapid economic growth. A simultaneous-equation model relating exports to income growth through their effects on capital goods imports has been formulated. We are able to show that in three economies, viz. Hong Kong, Korea and Singapore, all our hypothesized relationships in the model can be established. The results for Taiwan are not completely satisfactory as our hypothesis of a close link between investment and capital goods import does not seem to be valid. However in view of the high explanatory power of exports on income in the case of Taiwan, we should perhaps take the view that capital goods import in Taiwan affects income growth via some deterninants other than investment. Our model largely fails to account for the growth experience of Japan for the period under study. Evidently, our model which assumes a foreign-resources constraint is not very suitable for Japan considering its capability of financing growth by its own capital goods supply. From the causes of growth, we turned to the possible effects of growth on the economic system. We built up simultaneous- equation models to investigate the interdependent relationships between income growth, saving, and capital inflow. Capital inflow is separated into private and official inflows, an attempt which turned out to be highly important as their effects on saving and growth can be quite different from each other. We find that both saving and income, and saving and capital inflow (particularly private inflow) are positively related to each other. Our results on the relationship between saving and capital inflow are contrary to many cross-section studies which almost invariably used the single-equation estimation methods. We also investigate the relationship between export earnings and domestic saving. It is found that in the cases of Hong Kong, Korea and Singapore, export earnings, especially earnings from manufacturing exports, constitute the major source of domestic saving, in the case of Taiwan, both the foreign-trade and domestic sectors are important in generating saving. For Japan, saving is mainly generated from earnings in the domestic sector. Lastly, we focused on a more recently raised issue in development economics: the relationship between economic growth and income distribution. It is found that with the exception of Japan, rapid economic growth was compatible with increases in income equality. Doubts are therefore cast on the established "inverted-U" hypotheses concerning the relationship between income inequality and stages of economic development. We find that a set of very complicated historical, social, economic, and political factors enters the determination of income distribution in the process of economic development. For instance, in our group of economies, income distribution in Hong Kong and Japan was mainly dominated by the economic forces of growth while historical factors were crucial in the cases of Korea and Taiwan. In sum, there cannot be any inexorable laws governing the relationships between equality and economic development. Cross-section studies with the hope of generating universal laws of changing distribution over time should perhaps be replaced by the more fruitful studies of individual countries. In the final Chapter, a summary of results on individual economies is presented, and this is followed by some concluding remarks on development policy. The probable favourable effects of the smooth working of the price mechanism on economic growth are put forth on the basis of the growth experience of the five economies under study.