Middle-Income Transitions: Trap or Myth?

Felipe, Jesus; Kumar, Utsav; Galope, Reynold Working Paper Middle-Income Transitions: Trap or Myth? ADB Economics Working Paper Series, No. 421 Provided in Cooperation with: Asian Development Bank (ADB), Manila Suggested Citation: Felipe, Jesus; Kumar, Utsav; Galope, Reynold (2014) : Middle-Income Transitions: Trap or Myth?, ADB Economics Working Paper Series, No. 421, Asian Development Bank (ADB), Manila, https://hdl.handle.net/11540/4217 This Version is available at: https://hdl.handle.net/10419/128535 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/3.0/igo/ ASIAN DEVELOPMENT BANK AsiAn Development BAnk 6 ADB Avenue, Mandaluyong City 1550 Metro Manila, Philippines www.adb.org Middle-Income Transitions Trap or Myth? The historical evidence presented in this paper indicates that economies move up across income groups. Historically, it has taken a “typical” economy 55 years to graduate from lower-middle income to uppermiddle income. Likewise, we find that, historically, it has taken 15 years for an economy to graduate from upper-middle income to high income. Our analysis implies that as of 2013, there were 10 (out of 39) lowermiddle-income economies and 4 (out of 15) upper-middle-income economies that were experiencing slow transitions, i.e., above 55 and 15 years, respectively. About the Asian Development Bank ADB’s vision is an Asia and Pacific region free of poverty. Its mission is to help its developing member countries reduce poverty and improve the quality of life of their people. Despite the region’s many successes, it remains home to approximately two-thirds of the world’s poor: 1.6 billion people who live on less than $2 a day, with 733 million struggling on less than $1.25 a day. ADB is committed to reducing poverty through inclusive economic growth, environmentally sustainable growth, and regional integration. Based in Manila, ADB is owned by 67 members, including 48 from the region. Its main instruments for helping its developing member countries are policy dialogue, loans, equity investments, guarantees, grants, and technical assistance. MIDDle-IncoMe TrAnsITIons: TrAp or MyTh? Jesus Felipe, Utsav Kumar, and Reynold Galope adb economics working paper series no. 421 november 2014 ADB Economics Working Paper Series Middle-Income Transitions: Trap or Myth? Jesus Felipe, Utsav Kumar, and Reynold Galope No. 421 | 2014 Jesus Felipe ([email protected]) is Advisor and Utsav Kumar ([email protected]) is Economist in the Economics and Research Department of the Asian Development Bank. Reynold Galope ([email protected]) is visiting Assistant Professor at Georgia State University. We thank Arnelyn Abdon for collaboration in the previous version of this paper. We acknowledge comments from the participants at an ADB’s Economics and Research Department seminar. ASIAN DEVELOPMENT BANK Asian Development Bank 6 ADB Avenue, Mandaluyong City 1550 Metro Manila, Philippines www.adb.org © 2014 by Asian Development Bank November 2014 ISSN 2313-6537 (Print), 2313-6545 (e-ISSN) Publication Stock No. WPS146986-3 The views expressed in this paper are those of the author and do not necessarily reflect the views and policies of the Asian Development Bank (ADB) or its Board of Governors or the governments they represent. ADB does not guarantee the accuracy of the data included in this publication and accepts no responsibility for any consequence of their use. By making any designation of or reference to a particular territory or geographic area, or by using the term “country” in this document, ADB does not intend to make any judgments as to the legal or other status of any territory or area. Note: In this publication, “$” refers to US dollars. The ADB Economics Working Paper Series is a forum for stimulating discussion and eliciting feedback on ongoing and recently completed research and policy studies undertaken by the Asian Development Bank (ADB) staff, consultants, or resource persons. The series deals with key economic and development problems, particularly those facing the Asia and Pacific region; as well as conceptual, analytical, or methodological issues relating to project/program economic analysis, and statistical data and measurement. The series aims to enhance the knowledge on Asia’s development and policy challenges; strengthen analytical rigor and quality of ADB’s country partnership strategies, and its subregional and country operations; and improve the quality and availability of statistical data and development indicators for monitoring development effectiveness. The ADB Economics Working Paper Series is a quick-disseminating, informal publication whose titles could subsequently be revised for publication as articles in professional journals or chapters in books. The series is maintained by the Economics and Research Department. CONTENTS TABLES AND FIGURES iv ABSTRACT v I. INTRODUCTION 1 II. DATA AND MIDDLE-INCOME THRESHOLDS 3 A. Data 3 B. Methodology: Identifying Income Cut-offs 4 C. Distribution of Economies by Income Classes 6 III. DEFINING MIDDLE-INCOME TRANSITIONS 9 IV. MIDDLE-INCOME TRANSITIONS TODAY 16 V. CONCLUSIONS: WHAT ARE WE TO MAKE OF THE TERM MIDDLE-INCOME TRAP? 20 APPENDIX 21 REFERENCES 27 TABLES AND FIGURES TABLES 1 Change in the Distribution of Economies by Income Categories, 1950--2013 8 2 Economies That Have Always Been Low Income during 1950–2013 8 3 Economies That Became Lower-Middle Income before 1950 and Graduated to Upper-Middle Income 10 4 Economies That Became Lower-Middle Income after 1950 and Graduated to Upper-Middle Income 11 5 Threshold Number of Years to Separate Fast from Slow Transitions during Lower-Middle Income to Upper-Middle Income 13 6 Economies That Became Upper-Middle Income before 1950 and Graduated to High Income 13 7 Economies That Became Upper-Middle Income after 1950 and Graduated to High Income 14 8 Threshold Number of Years to Separate Fast from Slow Transitions during Upper-Middle Income to High Income 15 9 Economies Undergoing Slow Transitions from Lower-Middle Income to Upper-Middle Income as of 2013 16 10 Economies Undergoing Slow Transitions from Upper-Middle Income to High Income as of 2013 17 11 Economies Not Undergoing Slow Transitions from Lower-Middle Income to Upper-Middle Income as of 2013 18 12 Economies Not Undergoing Slow Transitions from Upper-Middle Income to High Income as of 2013 19 13 Total Number of Years Taken to Graduate from Lower-Middle-Income Status to High Income 20 FIGURES 1 Number of Years Elapsed since the Economy Reached $3,000 GDP per Capita Income 1 2 Distribution of Economies by Income Categories, 1950–2013 7 3 Year an Economy Turned Lower-Middle Income and Number of Years It Spent as Lower-Middle Income 12 4 Year an Economy Turned Upper-Middle Income and Number of Years It Spent as Upper-Middle Income 15 ABSTRACT During the last few years, the newly coined term middle-income trap has been widely used by policymakers to refer to the middle-income economies that seem to be stuck in the middle-income range. However, 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 economies do not advance. Overall, the data rejects this proposition. Instead, we argue that what distinguishes economies in their transition from middle to high income is fast versus slow transitions. We find that, historically, it has taken a “typical” economy 55 years to graduate from lowermiddle income ($2,000 in 1990 purchasing power parity [PPP] $) to upper-middle income ($7,250 in 1990 PPP $). Likewise, we find that, historically, it has taken 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 that 4 (out of 15) upper-middleincome 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. Analyzing a large sample of economies over many decades, indicates that experiences are wide, including many economies that today are high income that spent many decades traversing the middle-income segment. Keywords: middle-income trap, middle-income transition JEL Classification: O10, O40 I. INTRODUCTION Since Gill and Kharas (2007) introduced the term “middle-income trap” in the lexicon, the number of papers mentioning this supposedly observable phenomenon, as well as indirect references to it, has increased significantly. In many developing economies, policy discussions center on it and governments even speak of drawing plans to avoid it. The problem is that despite the numerous references to it, the idea of a middle-income trap is rather vague, not to mention that the term is not part of development-growth literature. The idea derives from the observation that some economies that managed to cross the low-income category into the middle-income category have not yet made it into the high-income category; while some others have made it. The former economies have been referred to as being stuck in the “middleincome trap.” Economies like Malaysia, Thailand, Brazil, or the Philippines have been said to be in the trap. And many others, including the People’s Republic of China (PRC), have been warned that they might fall into it soon. One way to see graphically what those who refer to this phenomenon could possibly mean is shown in Figure 1, which plots in the horizontal axis the number of years elapsed since economies reached $3,000 (measured in 2005 purchasing power parity [PPP] $), against income per capita in the vertical axis. While the Republic of Korea has progressed significantly, the other economies in Figure 1 seem to be “stuck.” Figure 1: Number of Years Elapsed since the Economy Reached $3,000 GDP per Capita Income GDP = gross domestic product, PPP = purchasing power parity, PRC = People’s Republic of China. Notes: GDP per capita is calculated as the ratio of GDP to population. GDP in PPP is from the Penn World Tables database, version 8.0. The series used is the output-side real GDP at chained PPP, in 2005 $. Population is from the World Bank’s World Development Indicators online database. We take the ratio. The PRC reached $3,000 in 1996, Indonesia in 1991, the Republic of Korea in 1974, Malaysia in 1961, the Philippines in 1989, and Thailand in 1980. Horizontal axis shows the time elapsed since attaining $3000 (in 2005 PPP $). References to this term can be classified into three groups. First are the references to the fact that the transition from low into middle income is a major leap in the quest to become high income but without explicitly mentioning the term (e.g., Spence 2011, chapter 16); or explicit references to the 0 5,000 10,000 15,000 20,000 25,000 30,000 1 6 11 16 21 26 31 36 41 46 51 GDP per capita, 2005 PPP $ PRC Indonesia Korea, Republic of Malaysia Philippines Thailand 2 | ADB Economics Working Paper Series No. 421 term but with strong qualifiers, e.g., “…if such trap indeed exists” (World Bank 2010). Second, are the believers in the idea of a middle-income trap (Gill and Kharas 2007, Ohno 2009, Kharas and Kohli 2011). They define it as a situation where economies are unable to compete with low-income, lowwage economies in manufactures and unable to compete with advanced economies in high-skilled innovation. Third is the recent literature on growth slowdowns, which refers to rapidly growing economies stagnating at middle income and failing to graduate to high income (Eichengreen, Park, and, Shin 2011, 2013; Aiyar et al. 2013). Eichengreen, Park, and Shin (2013) argue that growth in middleincome economies decelerates in steps—once around $10,000–$11,000 (in 2005 PPP $) and then again around $15,000–$16,000.1 Aiyar et al. (2013) show that middle-income economies are more likely to experience growth slowdowns than either low-income economies or high-income economies. Hence, they argue, the former economies are “trapped.” In this paper, we update the results of Felipe, Abdon, and Kumar (2012), where we introduced a definition of the middle-income trap.2 To the best of our knowledge, it was the first paper to do so. We argue, first, that none of the papers referred to above has provided a definition of the middleincome trap, much less a theoretical treatment of the phenomenon. The idea of being unable to compete (the second strand of literature referred to above) is almost a tautology since, under such view, all middle-income economies are, by definition, trapped. We believe that some authors have tried to draw a parallel with a well-established concept in the development literature, namely Nelson’s (1956) notion of “low-level equilibrium trap.” The parallel, however, is rather unfortunate because of the lack of a theory that explains what the middle-income trap is. The third view does not provide a definition either. This group is concerned with growth slowdowns, which is not the same as being “trapped.” A growth slowdown at best means a bump in the path to high-income status but does not mean that an economy experiencing a growth slowdown will be unable to reach high income. It only implies that it might take it slightly longer. At best, this strand of the literature provides a characterization of the economies that have not reached high-income status, and identified some factors that may be behind the growth slowdown and the inability to transition to high income. Second, we propose to study transitions across income groups to see if there is evidence that economies do not advance, i.e., that they are stuck. This is fundamental for any sound discussion of the likelihood of the alleged phenomenon of the middle-income trap, as well as for policy debates. The criterion we propose (note that we avoid using the term middle-income trap) is based on a thorough analysis of the historical transitions of a large number of economies across income categories. Based on this historical experience, we determine the number of years that economies have typically spent in the middle-income segment. Together with the income thresholds for each income category, this allows us to calculate the growth rate (of per capita gross domestic product [GDP]) that economies would need to achieve to cross the middle-income segment in this typical (more precisely, the median) number of years. The logical consequence of our argument is that some economies cross the middle-income segment faster than others simply because they (the former) grow faster. This helps bring the discussion back to the familiar turf of growth theory and its central question, namely, why some economies grow faster than others, a question that we do not tackle in this paper. The rest of the paper is structured as follows. In Section II we explain how we construct the data set that we use and show how we obtain the income per capita cutoffs in order to identify the 1 In an earlier version of the paper, Eichengreen, Park, and Shin (2012) report the existence of only a single node around $15,000–$16,000 at which slowdowns occur. 2 In this paper, we (i) extend the data coverage to 2013, (ii) revise the income classification of economies to smoothen 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. Middle-Income Transitions: Trap or Myth? | 9 Figure 2 also shows that there was a sharp increase in the number of high-income economies from the late 1960s to 1980, and from the late 1980s to 2013. The former period overlaps with what Maddison (1982) referred to as the “Golden Age” (1950–1973), when productivity accelerated considerably. During this period, several non-European economies, particularly East Asian (the Republic of Korea; Singapore; and Taipei,China) and Latin American (Argentina and Chile) reached high-income status. The number of economies that reached the high-income threshold increased to 10 in 1970, 21 in 1980, and 33 in 2013 (Table 1). In the next section, we use the income thresholds derived here to examine historical transitions from one income category into the next one. In doing so, we separate “slow” from “fast” transitions. We think that what distinguishes economies in their quest to reach high-income status is the speed of these transitions. As the discussion above indicates, historically, economies do advance. It is this distinction that matters most and helps bring the debate back to the familiar territory of growth theory, i.e., why do some economies grow faster than others? Based on the benchmarks that we develop in the next section to distinguish slow from fast transitions, we identify the economies that as of 2013 can be considered to have slow transitions from LM to UM, and those with slow transitions from UM to H. We also identify the economies that as of 2013 cannot be considered to have slow transitions. III. DEFINING MIDDLE-INCOME TRANSITIONS Our analysis of middle-income transitions (MIT) is based on the historical experience of economies that reached high income and the time it took them to do so. Given the lack of a theory of how long it takes for an economy to traverse from LM to UM and from UM to H, we adopt a simple procedure that consists in determining the threshold number of years that an economy has to be in one of the middleincome groups so that, beyond it, one can say that it is relatively slow to graduate. This number of years is determined by examining the historical experience of the economies that graduated from lowermiddle income to upper-middle income, and from upper-middle income to high income. We take the median number of years that it took these economies to transition as our benchmark to separate fast from slow transitions. Consequently, we will say that an economy is slow in graduating from the lower- /upper-middle-income group today if it has been in that group longer than the median, based on historical transitions. This method entails an unavoidable element of subjectivity, and therefore, one has to be careful in taking the threshold number of years literally. It is only a guide. We examine both the lower-middle-income and upper-middle-income transitions separately. The transition from lower-middle income into upper-middle income We first determine the number of years that economies remained in the lower-middle-income group before they graduated to upper-middle income. To do so, we separate data before and after 1950 because we have a complete time series for 124 economies starting in 1950. From the list of 124 economies, a total of 45 economies have graduated from lower-middle income into upper-middle income. We divide them into two groups: (i) the 36 economies that became lower-middle income before or in 1950 and then graduated to upper-middle income (Table 3); and (ii) the 9 economies that became lower-middle income after 1950 and then graduated to upper-middle income (Table 4). This allows us to compare recent transitions with those that took place earlier. The tables give the year these economies attained lower-middle-income status; the year they attained upper-middle-income status; the number of years they were lower-middle income; and their average income per capita growth rate during their transition from lower-middle income to upper-middle income. 10 | ADB Economics Working Paper Series No. 421 The time spent as lower-middle income for economies in Table 3 (economies that became lower-middle income before 1950) ranges from 19 years for Israel to 128 for the Netherlands. The latter was the first economy to become lower-middle income (in 1827, over 100 years earlier than Japan) but spent 128 years, until 1955, in this category. Maddison (1982, p.4) pointed out that the acceleration of productivity growth happened during what he referred to as the “capitalist era” that began in 1820. The Netherlands, being the economic leader during the 1700s, was the richest economy during that time until the UK overtook it in the second half of the 19th century. Japan (a latecomer with respect to the advanced Western economies) spent 35 years as a lower-middle-income economy. Table 3: Economies That Became Lower-Middle Income before 1950 and Graduated to Upper-Middle Income Economy Region Year the Economy Turned LM Year the Economy Turned UM Years as LM Average Growth Rate, LM to UM (%) Australia AP 1851 1950 99 1.2 Hong Kong, China* AP 1950 1976 26 5 Japan AP 1933 1968 35 3.9 New Zealand** AP 1860 1949 80 1.4 Singapore* AP 1950 1978 28 4.6 Austria Europe 1876 1964 88 1.5 Belgium Europe 1854 1961 107 1.2 Denmark Europe 1872 1953 81 1.6 Finland Europe 1922 1964 42 3.1 France Europe 1874 1960 86 1.4 Germany Europe 1874 1960 86 1.5 Greece*** Europe 1924 1972 38 2.7 Hungary**** Europe 1925 2001 73 1.6 Ireland***** Europe 1913 1975 55 1.6 Italy Europe 1906 1963 57 2.3 Netherlands Europe 1827 1955 128 1 Norway Europe 1907 1961 54 2.5 Poland Europe 1950 2000 50 2.2 Portugal Europe 1947 1978 31 4.2 Spain Europe 1913 1973 60 2.2 Sweden Europe 1896 1954 58 2.2 Switzerland Europe 1868 1945 77 1.8 United Kingdom Europe 1845 1953 108 1.2 Argentina****** LAC 1980 1970 71 1.5 Chile LAC 1891 1992 101 1.3 continued on next page Middle-Income Transitions: Trap or Myth? | 11 Table 3 continued Economy Region Year the Economy Turned LM Year the Economy Turned UM Years as LM Average Growth Rate, LM to UM (%) Colombia LAC 1946 2013 67 1.9 Mexico LAC 1942 2004 62 2.1 Panama LAC 1945 2011 66 2 Uruguay LAC 1870 1994 124 1 Venezuela LAC 1925 1948 23 5.7 Israel* MENA 1950 1969 19 5.5 Saudi Arabia* MENA 1950 1970 20 6.3 Syrian Arab Republic* MENA 1950 1996 46 2.5 Canada North America 1881 1950 69 1.9 United States******* North America 1860 1941 72 1.7 Mauritius* SSA 1950 1991 41 2.8 AP = Asia and the Pacific, GDP = gross domestic product, LAC = Latin America and Caribbean, LM = lower-middle income, MENA = Middle East and North Africa, SSA = Sub-Saharan Africa, UM = upper-middle income. * For Israel, Mauritius, and Saudi Arabia, 1950 is the first year for which data is available in the Maddison database. For Hong Kong, China; Singapore; and Syrian Arab Republic, data for 1914–1949 is not available. All three economies were low income in 1913, the earliest year before 1950 for which data is available. ** New Zealand turned LM in 1860. However, there is no data on GDP per capita for 1861–1870. These 10 years are not counted as part of the time New Zealand was LM from 1860 to 1948. *** For 10 years from 1941 to 1950, Greece was a low-income economy. We do not consider this period a jump for purposes of adjustment. Therefore, in calculating the time Greece was LM from 1924 to 1971, these 10 years are excluded. **** Hungary turned LM in 1925. However, there is no data on GDP per capita for 1943–1945. These 3 years are not counted as part of the time Hungary was LM from 1925 to 2000. ***** Ireland turned LM in 1913. However, there is no data on GDP per capita for 1914–1920. These 7 years are not counted as part of the time Ireland was LM from 1913 to 1974. ****** Argentina turned LM in 1890. However, there is no data on GDP per capita for 1891–1899. These 9 years are not counted as part of the time Argentina was LM from 1890 to 1969. ******* The United States turned LM in 1860. However, there is no data on GDP per capita for 1861–1869. These 9 years are not counted as part of the time the US was LM from 1860 to 1940. Source: Authors. On the other hand, the time spent as lower-middle income for the nine economies that became lower-middle income after 1950 (Table 4) ranges from 17 years for the PRC to 50 years and above for Bulgaria, Costa Rica, and Turkey. This is significantly lower than the time spent as lower-middle income by most economies that had crossed the lower-middle-income threshold before 1950 (Table 3). Table 4: Economies That Became Lower-Middle Income after 1950 and Graduated to Upper-Middle Income Economy Region Year the Economy Turned LM Year the Economy Turned UM Years as LM Average Growth Rate, LM to UM (%) PRC AP 1992 2009 17 7.5 Malaysia AP 1969 1996 27 5.1 Korea, Republic of AP 1969 1988 19 7.2 Taipei,China AP 1967 1986 19 7.0 Thailand AP 1976 2004 28 4.7 Bulgaria Europe 1953 2006 53 2.5 Turkey Europe 1955 2005 50 2.6 Costa Rica LAC 1952 2006 54 2.4 Oman MENA 1968 2004 36 2.4 AP = Asia and the Pacific, LAC = Latin America and Caribbean, LM = lower-middle income, MENA = Middle East and North Africa, PRC = People’s Republic of China, UM = upper-middle income. Source: Authors. 12 | ADB Economics Working Paper Series No. 421 Figure 3 uses the information in Tables 3 and 4 for the 45 economies that made the transition from lower-middle income into upper-middle income (See Appendix Table 1 for the codes of each economy). It shows the regression line between the year an economy entered the lower-middleincome group and the number of years it spent in that group, before graduating into upper-middle income. Clearly, recent transitions have been significantly faster than those in the past. Figure 3 shows a statistically significant and negative relationship between the two variables with a slope of –0.6. Uruguay, Chile, or Hungary are well above the regression line. Hungary, for example, became a lowermiddle-income economy in 1925, the same year as Venezuela. However, while it took the former 73 years to cross the lower-middle-income range, Venezuela did it in just 23. Figure 3: Year an Economy Turned Lower-Middle Income and Number of Years It Spent as Lower-Middle Income LM = lower-middle income, N = Sample size, R-sq = R-squared. Notes: The line shown is obtained from the regression of the number of years in LM on year the economy turned LM. The regression result is shown in the figure. Both the constant and the coefficient on “year turned LM” are statistically significant at the 1% level of significance. See Appendix Table 1 for the codes of each economy. Source: Authors. The idea of a middle-income trap was conceived relatively recently by analyzing recent development experiences, not those of the 19th century or earlier ones. The median number of years that the economies in Table 3 spent as lower-middle income is 64 years, while the median of the economies in Table 4 is 28 years (Table 5). The latter is clearly driven by the fast transition of the five East and Southeast Asian economies, significantly faster than those of the other economies in Table 3. Only a few economies that made the transition before 1950 (Table 3) match the experience of these Asian economies (e.g., Israel, Portugal, and Venezuela). Table 5 shows that the median and mean of the time taken to traverse from LM to UM post-1950 is influenced by the experience of East and Southeast Asian economies; without the latter, the median increases to 52 years. Since the fast transitions seen post-1950 do not seem to be the norm, we combine all 45 economies that made the transition from lower-middle income into upper-middle income. The median number of years spent in ARG AUS AUT BEL BGR CAN CHL COL CRI DEN FIN FRAGER GRC HUN IRE ISR ITA MUS MEX NET NZL NOR OMN PAN POL POR SAU SPA SWE SWI SYR TUR UKG USA URY VEN PRC HKG JPN MAL KOR SIN TAP THA 20 40 60 80 100 120 Years in LM 1800 1850 1900 1950 2000 Year turned LM Years in LM = 1215 - 0.60(Year turned LM) R-sq = 0.75, N = 45 Middle-Income Transitions: Trap or Myth? | 13 the lower-middle-income group has been 55 years (Table 5). We use this as a guide to separate slow from fast transitions. We will say that an economy is undergoing a slow transition in 2013 if it has spent over 55 years as a lower-middle-income economy, from the year it became an LM economy. With the guide of 55 years at hand, we can estimate the growth rate of per capita GDP that is necessary to transit from $2,000 to $7,250 in 55 years or less. This is 2.37% (or higher) per annum. Under this criterion, many advanced economies today went through slow transitions, although this did not prevent them from becoming high income. Table 5: Threshold Number of Years to Separate Fast from Slow Transitions during Lower-Middle Income to Upper-Middle Income Set of Economies Economies That Became LM after 1950 and Then Became UM Economies That Became LM in or before 1950 and Then Became UM All Economies Median Mean Median Mean Median Mean With East and Southeast Asia 28 34 64 65 55 58 Without East and Southeast Asia 52 48 67 68 62 66 LM = lower-middle income, UM = upper-middle income. Note: Entry in each cell is the number of years. Source: Authors. The transition from upper-middle income into high income In the second stage, we determine the number of years that economies remained in the upper-middleincome range before moving into high income. There are 30 economies that transitioned from uppermiddle income into high income (recall that Kuwait, Qatar, the United Arab Emirates were high income in 1950). These are again split into two groups: (i) the 5 economies that made the transition from upper-middle income into high income before 1950 (Table 6); and (ii) the 25 economies that made the transition from upper-middle income into high income after 1950 (Table 7). Looking at the list of economies in Table 6, the number of years spent as upper-middle income ranges from 14 years for Switzerland to 23 years for New Zealand. On the other hand, the time spent as upper-middle income for the economies in Table 7 ranges from a decade or less for the Asian economies to 41 years for Argentina. The difference between the maximum number of years spent as upper-middle-income economy before graduating to high income between these two groups is smaller than in the case of transition from lower-middle income to upper-middle income (compare with Table 3 and Table 4). Note that more than half of the economies in Table 7 are European, and five are Asian. Table 6: Economies That Became Upper-Middle Income before 1950 and Graduated to High Income Economy Region Year the Economy Turned UM Year the Economy Turned H Years as UM Average Growth Rate (%) Australia AP 1950 1970 20 2.4 New Zealand AP 1949 1972 23 2.0 Switzerland Europe 1945 1959 14 3.1 Canada North America 1950 1969 19 2.6 United States North America 1941 1962 21 1.8 AP = Asia and the Pacific, H = high income, UM = upper-middle income. Source: Authors. 14 | ADB Economics Working Paper Series No. 421 Table 7: Economies That Became Upper-Middle Income after 1950 and Graduated to High Income Economy Region Year the Economy Turned UM Year the Economy Turned H Years as UM Average Growth Rate (%) Hong Kong, China AP 1976 1983 7 5.9 Japan AP 1968 1977 9 4.7 Korea, Republic of AP 1988 1995 7 6.5 Singapore AP 1978 1988 10 5.1 Taipei,China AP 1986 1993 7 6.9 Austria Europe 1964 1976 12 4.1 Belgium Europe 1961 1973 12 4.4 Denmark Europe 1953 1968 15 3.3 Finland Europe 1964 1979 15 3.6 France Europe 1960 1971 11 4.4 Germany Europe 1960 1973 13 3.4 Greece Europe 1972 2000 28 1.8 Ireland Europe 1975 1990 15 3.2 Italy Europe 1963 1978 15 3.4 Netherlands Europe 1955 1970 15 3.3 Norway Europe 1961 1975 14 3.5 Portugal Europe 1978 1996 18 2.8 Spain Europe 1973 1990 17 2.7 Sweden Europe 1954 1968 14 3.6 United Kingdom Europe 1953 1973 20 2.5 Argentina LAC 1970 2011 41 1.2 Chile LAC 1992 2005 13 3.7 Uruguay LAC 1994 2012 18 2.6 Israel MENA 1969 1986 17 2.6 Mauritius SSA 1991 2003 12 4.0 AP = Asia and the Pacific, H = high income, LAC = Latin America and Caribbean, MENA = Middle East and North Africa, SSA = Sub-Saharan Africa, UM = upper-middle income. Source: Authors. As above, we use the information on the 30 economies in Tables 6 and 7, and regress the year an economy entered the upper-middle income and the number of years it spent in that group, before graduating into high income. Recent transitions have also been faster than those in the past. Figure 4 shows a statistically significant and negative relationship between the two variables with a slope of – 0.11. Argentina and Greece appear to be well above the regression line. Middle-Income Transitions: Trap or Myth? | 15 Figure 4: Year an Economy Turned Upper-Middle Income and Number of Years It Spent as Upper-Middle Income N = Sample size, R-sq =R-squared, UM = Upper-middle income. Note: The line shown is obtained from the regression of the number of years in UM on the year the economy turned UM. The regression result is shown in the figure. The constant and the coefficient on “year turned UM” are statistically significant at the 5% and 10% level of significance respectively. See Appendix Table 1 for the codes of each economy. Source: Authors. Table 8 provides the summary information to separate fast transitions from UM to H from the slow ones. The median of all 30 economies is 15 years. Therefore, we will say that an economy is undergoing a slow transition as of 2013 if it has spent over 15 years as an upper-middle-income economy, from the time it became a UM economy. With this at hand, we can estimate the growth rate of per capita GDP that is necessary to avoid a slow transition from UM to H, that is, to transit from $7,250 to $11,750 in 15 years or less. This is, at least, 3.27% per annum. In this case, and as noted above, only Greece and especially Argentina spent a very long time as upper-middle-income economies. Table 8: Threshold Number of Years to Separate Fast from Slow Transitions during Upper-Middle Income to High Income Set of Economies Economies That Became UM after 1950 and Then Became H Economies That Became UM in or before 1950 and Then Became H All Economies Median Mean Median Mean Median Mean With East and Southeast Asia 14 15 15 16 Without East and Southeast Asia 15 17 20 19 15 17 H = high income, UM = upper-middle income. Note: Entry in each cell is the number of years. Source: Authors. ARG AUS AUT BEL CAN CHL DEN FIN FRA GER GRC IRE ISR ITA MUS NET NZL NOR POR SPA SWE SWI UKG USA URY HKG JPN KOR SIN TAP 10 20 30 40 Years in UM 1940 1960 1980 2000 Year turned UM Years in UM = 240 - 0.11(Year turned UM) R-sq = 0.06, N = 30 16 | ADB Economics Working Paper Series No. 421 IV. MIDDLE-INCOME TRANSITIONS TODAY The definitions in section III of slow/fast transitions from LM to UM and from UM to H allow us to identify the economies that in 2013 are in what we refer to as slow or fast transitions. Based on our income classification, there were 54 middle-income economies in 2013, 39 lower-middle income and 15 upper-middle income.21 Table 9 shows the 10 economies that as of 2013 were making the transition from LM to UM relatively slowly, compared to the historical benchmark of 55 years identified above. This means that they have been in this income segment for over 55 years, or, stated in terms of growth rate, their GDP per capita growth rate since the year they became lower-middle income, has been below 2.37% per annum. All 10 economies belong to Latin America, Middle East and North Africa, and Sub-Saharan Africa. Guatemala is the economy that has been a lower-middle-income economy the longest, 78 years. The table also provides, just for reference, an estimate of the time it will take them to become upper-middle-income economies by assuming they continue growing at the same rate as during 2003–2013. Table 9: Economies Undergoing Slow Transitions from Lower-Middle Income to Upper-Middle Income as of 2013 Economy Region 2013 GDP per Capita (1990 PPP $) Year the Economy Turned LM Years as LM until 2013 Average Growth Rate (%) since Turning LM <2.37%* Year to Reach $7,250** Brazil LAC 6,917 1958 56 2.14 2 Ecuador LAC 4,498 1954 60 1.27 17 Guatemala LAC 4,627 1936 78 0.90 30 Jamaica LAC 3,406 1955 59 0.89 -*** Peru LAC 6,385 1946 68 1.72 3 Jordan MENA 6,339 1956 58 1.89 4 Lebanon**** MENA 5,091 1950 64 1.16 11 Gabon**** SSA 4,428 1950 64 0.55 29 Namibia**** SSA 5,286 1950 64 1.41 9 South Africa**** SSA 5,328 1950 64 1.17 12 GDP = gross domestic product, LAC = Latin America and Caribbean, LM = lower-middle income, MENA = Middle East and North Africa, PPP = purchasing power parity, SSA = Sub-Saharan Africa, UM = upper-middle income. *This column shows the average GDP per capita growth rate from the time an economy turned lower-middle income until 2013. Thus, growth rates are calculated over different durations for each economy. These durations are shown in the fifth column. **Number of years to reach $7,250 is calculated as [ln(7250/gdppc2013) / ln(1 + grGDPpc2003-2013)] where gdppc2013 is the GDP per capita in 1990 PPP $ in 2013 and grGDPpc2003-2013 is the average growth rate in GDP per capita during 2003–2013. ***Jamaica’s GDP per capita growth during 2003–2013 was negative. As a result, the expected time taken to reach $7,250 cannot be calculated based on the ten-year growth rate. ****Data for these economies is only available since 1950. It is possible that they may have been in LM for a longer time. Source: Authors. Table 10 shows the four economies that as of 2013 were experiencing slow transitions from UM to H based on the historical benchmark identified above. Venezuela has been the longest, 66 years (however, its transition of the lower-middle-income segment was very fast, 23 years). The table also provides the estimated growth rate to become high income under the assumption that GDP per capita 21 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 Appendix Table 2. Middle-Income Transitions: Trap or Myth? | 17 growth is the same as during 2003–2013. Results indicate that Malaysia should graduate and become a high-income economy in 2014. Table 10: Economies Undergoing Slow Transitions from Upper-Middle Income to High Income as of 2013 Economy Region 2013 GDP per Capita (1990 PPP $) Year Economy Turned LM Years as LM Year Economy Turned UM Years as UM until 2013 Average Growth Rate (%) since Turning UM <3.27%* Years to Reach $11,750** Malaysia AP 11,654 1969 27 1996 18 2.29 <1 Venezuela LAC 10,414 1925 23 1948 66 0.52 3 Saudi Arabia*** MENA 10,090 1950 20 1970 35**** 0.80 5 Syrian Arab Republic*** MENA 8,947 1950 46 1996 18 1.00 17 AP = Asia and the Pacific, GDP = gross domestic product, H = high income, LAC = Latin America and Caribbean, LM = lower-middle income, MENA = Middle East and North Africa, PPP = purchasing power parity, UM = upper-middle income. * This column shows the average GDP per capita growth rate from the time an economy turned upper-middle income until 2013. Thus, growth rates are calculated over different durations for each economy. These durations are shown in the seventh column. ** Number of years to reach $11,750 is calculated as [ln(11750/gdppc2013) / ln(1 + grGDPpc2003-2013)] where gdppc2013 is the GDP per capita in 1990 PPP $ in 2013 and grGDPpc2003-2013 is the average growth rate in GDP per capita during 2003–2013. *** Data for these economies is only available since 1950. It is possible that they may have been in LM for a longer time. **** Saudi Arabia was a high-income economy from 1974 to 1982. These 9 years are not considered as a jump and therefore not adjusted. In calculating the time Saudi Arabia was UM since 1970 these 9 years have been excluded. Source: Authors. Tables 11 and 12 show the 29 lower-middle income and the 11 upper-middle-income economies that, as of 2013, were not having slow transitions and may be able to make it to the next income category in fewer years than the historical benchmark identified for transition from LM to UM (55 years) and from UM to H (15 years). Given the number of years they have been lower-middleincome economies and their recent growth performance, we can speculate about the economies that are at risk of making a slow transition from LM to UM. Table 11 shows that there are economies whose growth rates during 2003–2013 (shown in previous to last column) were below those required to reach $7,250 (shown in last column) within the number of years remaining before falling into a slow transition (shown in third column from last), e.g., Libya, Romania, Algeria, El Salvador, or Swaziland. If these economies want to transition into upper-middle income within the historical median of 55 years, they should implement policies to accelerate growth. 18 | ADB Economics Working Paper Series No. 421 Table 11: Economies Not Undergoing Slow Transitions from Lower-Middle Income to Upper-Middle Income as of 2013 Economy Region 2013 GDP per Capita (1990 PPP $) Year Economy Turned LM Years in LM until 2013 Years before Falling into a Slow Transition* Average Growth Rate (%) 2003– 2013 Average GDP per Capita Growth to Reach $7,250** Cambodia AP 2,969 2006 847 6.1 2.0 India AP 3,834 2002 12 43 6 1.5 Indonesia AP 5,548 1986 28 27 4.5 1.0 Lao PDR AP 2,220 2012 2 53 5.3 2.3 Myanmar AP 4,323 2004 10 45 8.5 1.2 Pakistan AP 2,386 2005 9 46 2.5 2.4 Philippines AP 3,429 1975 39 16 3.3 4.8 Sri Lanka AP 6,431 1983 31 24 5.5 0.5 Viet Nam AP 3,711 2002 12 43 5.6 1.6 Albania Europe 4,695 1970 44 11 4.0 4.0 Romania Europe 4,810 1962 52 3 3.2 14.7 Bolivia LAC 3,408 1968 46 9 2.8 8.7 Dominican Republic LAC 5,153 1973 41 14 3.2 2.5 El Salvador LAC 2,972 1964 50 5 0.8 19.5 Honduras LAC 2,357 2004 10 45 1.9 2.5 Paraguay LAC 3,789 1973 41 14 2.5 4.7 Algeria MENA 3,682 1972 42 13 1.6 5.4 Egypt MENA 3,935 1980 34 21 2.6 3.0 Iran MENA 7,153 1959 55 0 6.0 <1.0 Libya MENA 2,162 1962 52 3 -1.2 49.7 Morocco MENA 4,041 1977 37 18 3.3 3.3 Tunisia MENA 6,451 1972 42 13 2.7 0.9 Yemen, Republic MENA 2,501 1976 38 17 -0.6 6.5 Botswana SSA 5,155 1983 31 24 1.5 1.4 Congo, Republic SSA 2,502 1979 35 20 2.2 5.5 Ghana SSA 2,222 2012 2 53 5.1 2.3 Lesotho SSA 2,470 2009 5 50 4.5 2.2 Mozambique SSA 2,699 2007 7 48 5.1 2.1 Swaziland SSA 3,027 1970 44 11 1.0 8.3 AP = Asia and the Pacific, GDP = gross domestic product, LAC = Latin America and Caribbean, Lao PDR = Lao People’s Democratic Republic, LM = lower-middle income, MENA = Middle East and North Africa, PPP = purchasing power parity, SSA = Sub-Saharan Africa, UM = uppermiddle income. * Calculated as 55-number of years in LM until 2013 as shown in the fifth column. **Calculated as ((7,250/gdppc2013)^(1/(55-years in LM until 2013))-1)*100 where gdppc2013 is the GDP per capita in 1990 PPP $ in 2013 and years in LM until 2013 is as shown in the fifth column. Source: Authors’ calculations. Table 12 shows that the average growth rate needed to avoid a slow transition assuming the average growth seen during 2003–2013 prevails, shows that economies, such as Costa Rica, Hungary, Mexico, Oman, and Turkey, may experience a slow transition from UM to H. Thailand and Bulgaria may avoid the slow transition and the rest are likely to make it from UM to H in accordance to historical experience. Appendix: Adjustments to Income Classifications | 25 Appendix Table 2: Income Classification of Economies of the Former Soviet Union, Former Yugoslavia, and Former Czechoslovakia Income Classification in 2013 Number of Years in Economy L LM UM H Former Soviet Union Armenia H - 14 9 1 Azerbaijan UM - 17 7 - Belarus H - 13 5 6 Estonia H - - 11 13 Georgia LM -24 - - Kazakhstan H - 13 7 4 Kyrgyz Republic LM -24 - - Latvia H - 10 5 9 Lithuania UM -12 12 - Moldova LM - 24 - - Russian Federation UM -15 9 - Tajikistan L 24 - - - Turkmenistan LM -24 - - Ukraine LM - 24 - - Uzbekistan UM -23 1 - Former Yugoslavia Bosnia and Herzegovina LM -24 - - Croatia UM - 14 10 - Macedonia, Former Yugoslav Republic o f LM -24 - - Serbia and Montenegro LM - 24 - - Slovenia H - - 9 15 Former Czechoslovakia Czech Republic H - - 16 8 Slovak Republic H - 6 11 7 L = low income, LM = lower-middle income, UM = upper-middle income, H = high income. Notes: The following adjustments to income classifications were made in the case of the above listed 22 economies: (i) Azerbaijan’s income classification for the years 1995–1997 was adjusted from L to LM; (ii) Bosnia and Herzegovina’s income classification for the years 1993 and 1994 was adjusted from L to LM, and for the year 2008 was adjusted from UM to LM; (iii) Croatia’s income classification for the year 1990 was adjusted from UM to LM; (iv) Georgia’s income classification for the year 1990 was adjusted from UM to LM; (v) Kazakhstan’s income classification for the year 1990 was adjusted from UM to LM; (vi) the Kyrgyz Republic’s income classification for the years 1994–1996 was adjusted from L to LM; (vii) Latvia’s income classification for the years 1990 and 1991 was adjusted from UM to LM; (viii) Lithuania’s income classification for the years 1990 and 1991 was adjusted from UM to LM; (ix) the Russian Federation’s income classification for the years 1990 and 1991 was adjusted from UM to LM; (x) Slovak Republic’s income classification for the year 1990 was adjusted from UM to LM; (xi) Tajikistan’s income classification for the years 1990 and 1991 was adjusted from LM to L; and (xii) Turkmenistan’s income classification for the years 1997 and 1998 was adjusted from L to LM. Source: Authors. REFERENCES Aiyar, S., R. Duval, D. Puy, Y. Wu, and L. Zhang. 2013. Growth Slowdowns and the Middle-Income Trap. International Monetary Fund Working Paper. No. WP/13/71. Washington, DC: International Monetary Fund. Eichengreen, B., D. Park, and K. Shin. 2011. When Fast Growing Economies Slow Down: International Evidence and Implications for China. Asian Economic Papers. 11 (1). pp. 42–87. ———. 2013. Growth Slowdowns Redux: New Evidence on the Middle-income Trap. National Bureau of Economic Research Working Paper. No. 18673. Cambridge, MA: National Bureau of Economic Research. Ekstrom, J. 2010. On the Relation Between the Polychoric Correlation Coefficient and Spearman's Rank Correlation Coefficient. UCLA Statistics Preprint #629. Felipe, J., A. Abdon, and U. Kumar. 2012. Tracking the Middle-Income Trap: What is It, Who is In It, and Why? Working Paper. 715 (April). New York: Levy Economics Institute of Bard College. Gill, I. and H. Kharas. 2007. An East Asian Renaissance. Washington, DC: The World Bank. Kharas, H. and H. Kohli. 2011. What is the Middle Income Trap, Why do Countries Fall into It, and How it can be Avoided? Global Journal of Emerging Markets. 3 (3). pp. 281–289. Kolenikov, S. and G. Angeles. 2009. Socioeconomic Status Measurement with Discrete Proxy Variable: Is Principal Component Analysis a Reliable Answer? Review of Income and Wealth. 55 (1). pp. 128–165. Maddison, A. 1982. Phases of Capitalist Development. New York: Oxford University Press. Maddison, A. 2010. Historical Statistics of the World Economy: 1-2008 AD. http://www.ggdc.net/ MADDISON/oriindex.htm (accessed on 5 July 2011). Nelson, R. 1956. A Theory of the Low-Level Equilibrium Trap in Underdeveloped Economies. American Economic Review. 46 (5). pp. 894–908. Ohno, K. 2009. Avoiding the Middle Income Trap: Renovating Industrial Policy Formulation in Vietnam. ASEAN Economic Bulletin. 26 (1). pp. 25–43. Olsson, U. 1979. Maximum Likelihood Estimation of the Polychoric Correlation. Psychometrika. 44. pp. 443–460. Spence, M. 2011. The Next Convergence. The Future of Economic Growth in a Multispeed World. New York: Farrar, Straus and Giroux. World Bank. 2010. Research for Development: A World Bank Perspective on Future Directions for Research. World Bank Policy Research Working Paper. No. 5437. Washington, DC.  ADB recognizes China as the People’s Republic of China. ASIAN DEVELOPMENT BANK AsiAn Development BAnk 6 ADB Avenue, Mandaluyong City 1550 Metro Manila, Philippines www.adb.org Middle-Income Transitions Trap or Myth? The historical evidence presented in this paper indicates that economies move up across income groups. Historically, it has taken a “typical” economy 55 years to graduate from lower-middle income to uppermiddle income. Likewise, we find that, historically, it has taken 15 years for an economy to graduate from upper-middle income to high income. Our analysis implies that as of 2013, there were 10 (out of 39) lowermiddle-income economies and 4 (out of 15) upper-middle-income economies that were experiencing slow transitions, i.e., above 55 and 15 years, respectively. About the Asian Development Bank ADB’s vision is an Asia and Pacific region free of poverty. Its mission is to help its developing member countries reduce poverty and improve the quality of life of their people. Despite the region’s many successes, it remains home to approximately two-thirds of the world’s poor: 1.6 billion people who live on less than $2 a day, with 733 million struggling on less than $1.25 a day. ADB is committed to reducing poverty through inclusive economic growth, environmentally sustainable growth, and regional integration. Based in Manila, ADB is owned by 67 members, including 48 from the region. Its main instruments for helping its developing member countries are policy dialogue, loans, equity investments, guarantees, grants, and technical assistance. MIDDle-IncoMe TrAnsITIons: TrAp or MyTh? Jesus Felipe, Utsav Kumar, and Reynold Galope adb economics working paper series no. 421 november 2014