![Sample our Economics, Finance,Business & Industry journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5931/TOC-Subject-Sample-2021-Economics-Finance-Business-Industry.jpg"})
ABSTRACT
The newly coined term ‘middle-income trap’ has been widely used in recent years by policymakers to refer to those middle-income economies that seem to be stuck in the middle-income range of the income distribution. This has been done despite that there is no accepted definition of the term in the literature. In this paper, we study historical transitions across income groups to see whether there is any evidence that supports the claim that some middle-income economies do not advance. Overall, the data refute this proposition and, as a consequence, we reject the existence of a middle-income trap as a generalized phenomenon. Instead, we argue that what distinguishes economies in their transition from middle to high income is the speed of these transitions, fast versus slow, a standard growth question. We find that, historically, those economies that graduated from lower-middle income ($2000 in 1990 purchasing power parity [PPP] $) to upper-middle income ($7250 in 1990 PPP $) did it in about 55 years. Likewise, we find that, historically, it took 15 years for an economy to graduate from upper-middle income to high income (above $11,750 in 1990 PPP $). Our analysis implies that, as of 2013, there were 10 (out of 39) lower-middle-income economies and 4 (out of 15) upper-middle-income economies that were experiencing ‘slow transitions’ (i.e., above 55 and 15 years, respectively). The historical evidence presented in this paper indicates that economies move up across income groups. The analysis of a large sample of economies over many decades also indicates that many economies that today are high income spent many decades traversing the middle-income segment.
Acknowledgements
We thank Arnelyn Abdon for collaboration on the previous version of this paper. We acknowledge comments from the participants at an ADB's ERCD Seminar. The usual disclaimer applies. This paper reflects the views of the authors and not those of the Asian Development Bank, its Executive Directors, or those of the countries that they represent.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1. In an earlier version of the paper, Eichengreen, Park, and Shin (Citation2013) report the existence of only a single node around $15,000–$16,000 at which slowdowns occur.
2. In this paper, we update and extend the results presented in a previous working paper by Felipe, Abdon and Kumar (Citation2012). Specifically, we (i) extend the data coverage until 2013, (ii) revise the income classification of economies to smooth out the fluctuations in the income categories, and (iii) revise the criteria used in the earlier paper to determine whether an economy is ‘trapped’ or not.
6. These economies are as follows: (i) populations below 1 million people in 2012: Bahrain, Comoros, Cape Verde, Djibouti, Equatorial Guinea, Sao Tome and Principe, and Seychelles. Bahrain's population is more than 1 million today but was excluded as its population has exceeded 1 million since 2007 only. Pacific islands are not included in the Maddison data. Also, all these islands, except Papua New Guinea, have very small populations; (ii) economies of the former Soviet Union (15), the former Yugoslavia (5), and the former Czechoslovakia (2), for which data is not complete for 1950–2008; and (iii) Cuba, Democratic Republic of Korea, Puerto Rico, Somalia, and West Bank and Gaza whose GDP per capita estimates are not reported by the IMF database. In addition, we continue to leave out Trinidad and Tobago which was dropped from the data used for the previous version of this paper.
9. For the 124 economies with consistent data since 1950, we calculated annual growth rates. This resulted in 7,812 (124*63) annual growth rates. Of these 7,812 growth rates, 75 were higher than 20% (positive or negative). Most of these 75 observations are either resource-rich economies or economies in Sub-Saharan Africa. The other cases are Afghanistan, Albania, and Bulgaria. Three observations that stand out are the Republic of Korea's growth rate in 1953, and Cambodia's growth rates in 1973 and 2004. We take the Maddison data as it is for all these observations, except for Cambodia in 2004. The growth rate for Cambodia, based on the original Maddison date, in 2004 is estimated at 41.1%, which seemed implausible. For Cambodia, from 1990 to 2010, we use data from the revision of Maddison's data set under the “New Maddison Project Database” available at http://www.ggdc.net/maddison/maddison-project/data.htm. We do not update any other historical data, i.e., use the original Maddison data set.
10. ‘The process of setting per capita income thresholds started with finding a stable relationship between a summary measure of wellbeing such as poverty incidence and infant mortality on the one hand and economic variables including per capita GNI estimated based on the Bank's Atlas method on the other. Based on such a relationship and the annual availability of Bank's resources, the original per capita income thresholds were established.’ Source: World Bank (http://econ.worldbank.org/WBSITE/EXTERNAL/DATASTATISTICS/0,,contentMDK:20487070∼menuPK:64133156∼pagePK:64133150∼piPK:64133175∼theSitePK:239419,00.html).
11. The year the original threshold was established is not explicitly identified in the World Bank website (see previous footnote).
13. To decide the range of t0, t1, and t2, income categories of the economies in the Maddison data for 1990 were identified according to the World Bank's income classification in 1990. The mean and standard deviation of GDP per capita (as reported in the Maddison data) for each of the four income groups was then obtained. To obtain the bounds within which to vary t0, t1, and t2, the mean plus one standard deviation (rounded off) of GDP per capita of each group was used as the lower bound for each group. The mean plus one standard deviation for the low income, lower-middle income, upper-middle income, and high income are $1,542; $5,011; $9,104; and $19,642, respectively. This gives $1,500 as the lower bound for t0, $5,000 as the lower bound for t1, and $9,000 as the lower bound for t2. The upper bounds of each group are $250 below the lower bound of the next threshold with the exception of the upper bound for t2 which is assumed to be $20,000 based on the mean plus one standard deviation of GDP per capita of upper-middle-income group in 1990 of $19,642.
14. The polychoric correlation provides a measure of the degree of agreement between two raters (here, the World Bank and the present study) over a continuous variable (income) that has been transformed into ordered levels (several income levels). Ekstrom (Citation2010) argues that the polychoric correlation is a better measure of the association of the underlying continuous variables if the ordinal variables arise from groupings of values into categories.
15. The use of these constant thresholds is, in principle, equivalent to what the World Bank does. As discussed above, the World Bank's thresholds are inflation-adjusted and, therefore, remain constant in real terms.
16. For example, Angola was classified as lower-middle income and Egypt as low income in 1990 under the World Bank classification. The GDP per capita of Angola in the same year, according to Maddison's estimates in 1990 PPP $, was $868, and that of Egypt was $2,523. This makes Angola a low-income economy and Egypt a lower-middle-income economy in 1990 based on the thresholds defined in this paper.
17. Scatter plots showing the income categories before and after the adjustments are available upon request from the authors.
18. Only the United Arab Emirates (UAE) has remained high income for the entire period 1950–2013. After taking into account adjustments to the income groups (Appendix A), Kuwait fell back into the upper-middle-income category in 1981 and regained high-income status in 2005. Qatar fell upper-middle income in 1985 and regained high-income status in 2005. Though Kuwait, Qatar, and UAE had higher per capita incomes than all Western economies in 1950, today (as measured in 1990 PPP $) most Western economies have higher per capita incomes.
19. Note that many of these economies were in fact colonies during the 1950s and 1960s.
20. Some economies transitioned from low income to middle income during 1950–1980, and others transitioned from middle income to high income, over the same period. The net increase in the number of economies in the middle-income group is 13 (54–41).
21. We do so because data is complete for all economies after 1950, but not before 1950.
22. In the case of economies for which data begins in 1950, the number of years spent in any income group might be an underestimate. Israel, for example, may have turned lower-middle income before 1950, and therefore 19 years as lower-middle income may be an underestimate.
23. As noted above, the paper does not use for the analysis the 22 economies of the former Soviet Union, former Yugoslavia, and former Czechoslovakia. However, the income classifications based on the income thresholds identified are provided in
24. These results imply that Malaysia should graduate and become a high-income economy in 2014–2015.
25. The reported median in is not the sum of the median of the transitions from LM into UM, and then from UM into H, shown earlier in and . Rather, this is the median number of years that it took the 30 economies in our data set that transitioned from LM into H.
26. It is important to note that our criteria will have to be revised as more countries transit the middle-income segment.
27. Felipe et al. (Citation2014) find that all of today's rich nonoil economies enjoyed at least 18% manufacturing employment shares in the past, and often did so before becoming rich. High manufacturing output shares are not as important.
Additional information
Notes on contributors
Jesus Felipe
Jesus Felipe is Advisor in the Asian Development Bank's Economic Research and Regional Cooperation Department. He has been with ADB since 1996, and he is the Managing Editor of the Asian Development Review. His research interests spread across areas such as long-run growth in Asia, the dynamics of structural transformation, industrial policy, inclusive growth and full employment, the impact of technology on employment, productivity, technological progress, the functional distribution of income, business cycles, and the path of profit rates.
Utsav Kumar
Utsav Kumar is an economist in the Asian Development Bank's Economic Research and Regional Cooperation Department. His current research interests include growth and structural transformation, development challenges of small island countries, and Indian economy with publications in Cambridge Journal of Economics, Japan and the World Economy, Journal of Comparative Economics, Journal of Monetary Economics, and Structural Change and Economic Dynamics.
Reynold Galope
Reynold V. Galope is an assistant professor in the College of Community Studies and Public Affairs at Metropolitan State University, Saint Paul, Minnesota. He has a PhD in Public Policy and has published papers in Economic Development Quarterly, the Journal of Technology Management and Innovation, and the American Review of Politics.