Human Resource Landscape in Mainland ASEAN

193 Table 3: Japan’s ODA and Myanmar’s Growth and Trade. 3SLS Estimates 2001-2015 Variables Growth Trade/GDP Const 11.812** 60.091** Trade/GDP -0.021 FDI/GDP -0.000 Services/GDP -.001 ODA/GDP 0.0003 0.015** Economic Plan 2002 -36.537** Economic Plan 2003 1.366** Economic Plan 2005 10.450* Pre-GFC 2008 -3.075** GFC 2009 68.505* Euros Crisis 2010 -91.914** Post-Euro Crisis 2011 -3.763** Post-Euro Crisis 2012 -1510.33 FESR 2013 1.633** 1511.64 Domestic Demand -5.755** Trade Partner Demand 6.772** Real Exchange Rates 0.129 Terms of Trade 0.369 RSQ 0.941 0.692 DW 2.373 2.448 DF p-value 0.131 0.196

194 Table 4: Japan’s ODA and Vietnam’s Growth and Trade. 3SLS Estimates. 2001-2015 Variables Growth Trade/GDP Const 7.158** 17.433 Trade/GDP 0.037* FDI/GDP 0.014 Services/GDP -0.001 ODA/GDP 0.004 -0.113** Iraq War 2003 10.639** 5-Yr Plan/Strategy 2005 -0.519 -0.256 5-Yr Plan/Strategy 2007 9.408** GFC 2009 -0.971 -56.012** End of 5-Yr Plan 2010 48.860** Euro Crisis 2012 -0.885* -10.474** Post-Euro Crisis 2013 -1.041 Post-Euro Recovery 2014 0.634 Domestic Demand -0.121 Trade Partner Demand -4.764** Real Exchange Rates -2.421** Terms of Trade -0.734** RSQ 0.757 0.993 DW 1.967 2.820 DF p-value 0.076 0.398

195 Table 5: Friedman-Kydland Modelling Performance. Statistical Characteristics of Estimated Models of Growth and Trade, 2001-2015, Theil’s-MSE Decomposition Cambodia Laos Growth Trade Growth Trade Data Model Data Model Data Model Data Model Mean 7.769 7.769 3.470 3.470 7.335 7.335 0.441 0.441 Stand Dev 2.914 2.856 7.346 7.456 0.862 0.808 13.045 11.355 RSQ 0.829 0.842 0.895 0.406 MSE 1.391 8.457 0.073 102.981 Mean Error 0.000 0.000 0.000 0.000 Bias 0.000 0.000 0.000 0.000 Variance 0.002 0.002 0.036 0.026 Covariance 0.998 0.998 0.964 0.974 Note: For Tables 5-6, RSQ=R-squared, MSE=mean-squared error, Bias+Variance+Covariance=1 (See Pindyck and Rubinfeld 1998). Table 6: Friedman-Kydland Modelling Performance. Statistical Characteristics of Estimated Models of Growth and Trade, 2001-2015, Theil’s-MSE Decomposition Myanmar Vietnam Growth Trade Growth Trade Data Model Data Model Data Model Data Model Mean 10.509 10.508 -1.935 -1.936 6.504 6.504 4.053 4.053 Stand Dev 2.600 2.537 15.319 15.160 0.803 0.709 7.586 7.569 RSQ 0.941 0.692 0.757 0.993 MSE 0.373 73.029 0.146 0.364 Mean Error 0.000 0.000 0.000 0.000 Bias 0.000 0.000 0.000 0.000 Variance 0.010 0.000 0.058 0.001 Covariance 0.990 1.000 0.942 0.999 Note: For Tables 5-6, RSQ=R-squared, MSE=mean-squared error, Bias+Variance+Covariance=1 (See Pindyck and Rubinfeld 1998).

196 5. The Long-Run Economic Impact of Japan’s ODA (With the Interaction between Demand-Supply) The econometric estimation of the demand-oriented model above has shown rather weak relationships of growth, trade with Japan’s ODA. Our study would add the supply-oriented equation to the system. We hypothesize that all ODA from DAC to CLMV would be a source of resources to be added on to domestic saving. It is well- known that CLMV had been insufficient as a source of ‘investment or gross fixed capital formation.' The capital stock comprises both hard infrastructure like road and transportation system, electricity generation, etc., are necessary economic development at the early stage of food production and export and industrialization to get out of poverty enclave. The social infrastructure like education and training, health as well as welfare programs such as low-income housing, etc., are sufficient for public employment, welfare improvement, and equitable income distribution. Japan's ODA would be helpful to start up CLMV's economic development. Recently, we have observed Japan's ODA as compared with the Overall ODA (including Japan's ODA) flow to Vietnam substantively in ODA’s and trend as compared with Cambodia, Laos, and Myanmar during 1990-2012. In 2013, Japan had provided debt-relief for Myanmar. The trend of Japan’s ODA has become significant for Myanmar after Myanmar has turned to the path of ‘Democracy.' We have estimated the parameters of the system of equations below shown in Table 7-10. Each economy model is different from one another. The case of Vietnam, Japan’s ODA has played a crucial role in capital formation to be the basis for her rapid economic growth and development. Vietnam would still need ODA including Japan’s ODA for her capitalization of physical as well as human capital necessitated for her economic development to medium-high income country. The Myanmar model has shown a similar result to Vietnam. As later comer, Myanmar would need a continuous flow of Japan’s ODA (and other donors) to build up of necessary infrastructure such as transportation, electricity generation, human capability, etc. for her economic development. Japan's ODA including debt relief would be significant to improve the growth potential, opening up trade and capital formation simultaneously in the coming decades, should the resources flow continue in coming decades. The case of Cambodia and Laos PDR has outstanding notes. Even though these two neighbors have approximately small economic size, but there are fundamental differences in their policy planning. Laos was more cautious on trade openness than Cambodia. The more a country opened for trade, risk exposure through the current account and balance of payment’s transmission mechanism will be mounting. The negative sign of the coefficient of trade openness with lag (t-3) on economic growth determination confirms this. Thus, ODA would not bring about such volatility to country profile in the long-run. The Laos government has planned to exploit the hydroelectricity generation from her abundant natural water resource of the Khong international river. The private-public investment was a road map using the Thai companies. The electricity output is planned to supply to the ASEAN grid system. With sufficient future stream of income, Laos may have to balance between the foreign flow of ODA and own income-saving generation. The equations system has depicted clearly that ODA,

197 although positively related to the capital formation would need some gestation period of adjustment not instantaneous. Japan's ODA has a positive relationship with capital formation in Cambodian. The capital-labor ratio would have positive repercussions on economic growth. The openness to trade would finally strengthen the Cambodian economy. The supply-side estimation has further validated our findings. Japan's ODA as part of the overall ODA has contributed to growth via trade openness and capital accumulation. The econometric estimation has produced significant statistical results in CLMV. The findings have remedied the weak impact of ODA (with Japan's ODA as part of it) to growth and trade found by the demand model above. Thus, ODA impact is measured from both demand and supply side simultaneously. In other words, the impact of ODA should not be assessed only from supply-side as most of the literature has done, neither from the demand-side (expenditure side of the SNA). The ODA has interacted with growth and development via physical as well as human capital built up. The realization of the ODA impact would take a long-run than we have perceived. The demand-side model is impressive as it takes into account the economic crisis on trade openness and growth. The supply-side model which is longer-run bypassed such crisis factors but exhibit the capital accumulation process. The demand-supply has interacted in the medium to long run and produce heuristic results.

198 Table 7: Impact of Japan's ODA on Cambodia Growth Determination Equation: LOG(GDPR_CAMBODIA) = C(1) +C(2)*OPEN_RATIO_CAMBODiA(-2) +C(3)*CAPR_CAMBODIA(-1) /LABOR_CAMBODIA(-1) Trade Openness Equation: OPEN_RATIO_CAMBODIA = C(4)+C(5)*(GDPR_CAMBODIA(-1)) +C(6)*SERVICE_RATIO_CAMBODIA +C(7)*(FDI_GFCF_CAMBODIA(-1) ) Capital Formation Equation: CAPR_CAMBODIA/LABOR_CAMBODIA = C(8) + C(9) *ODA_JP_CAMBODIA + C(10)*ODA_JP_CAMBODIA(-1) + C(11)*(GDPR_CAMBODIA(-1)) Growth Prob. Trade Prob. Capital Prob. Openness Formation Constant 7.656204 *** -0.76127 *** -326.267 *** OPEN_RATIO_CAMBODIA(-2) 0.796546 *** CAPR_CAMBODIA(- 0.00028 *** 1)/LABOR_CAMBODIA(-1) (GDPR_CAMBODIA(-1)) 9.09E-05 *** SERVICE_RATIO_CAMBODIA 4.951631 *** (FDI_GFCF_CAMBODIA( 0.004374 *** ODA_JP_CAMBODIA 1.92533 *** ODA_JP_CAMBODIA (-1) 1.720482 *** (GDPR_CAMBODIA(-1)) 0.173496 *** R-squared 0.964791 0.949688 0.987434 Adjusted R-squared 0.95775 0.93711 0.982722 S.E. of regression 0.063123 0.116969 46.71513 Durbin-Watson stat 1.552473 0.922357 1.500368 Probability of Rejection; 0=*** sig at 99%; <0.05 ** sig at 95% ;<0.10 * sig at 90% Note: OPEN_RATIO= import+ export/GDP ratio; CAPR/Labor= real capital-labor ratio; service=service trade; FDI_GFCF= foreign direct investment-gross fixed capital formation ratio; ODA_JP=Official Development Assistance from Japan.

199 Table 8: Impact of Japan's ODA on Laos Growth Determination Equation: LOG(GDPR_LAOS ) = C(1)+ C(2)*OPEN_RATIO_LAOS(-3) +C(3)*CAPR_LAOS(-1) /LABOR_LAOS(-1) Trade Openess Determination Equation: OPEN_RATIO_LAOS = C(4) +C(5)*GDPR_LAOS(-1 ) +C(6) *TT_LAOS Formation of Physical and Human Capital Equation: CAPR_LAOS/LABOR_LAOS = C(7) +C(8)*ODA_JP_LAOS(-2) /FDI_GFCF_LAOS(-2) +C(9)*GDPR_LAOS(-1) +C(10)*D07 Growth Prob. Trade Prob. Capital Prob. Openness Formation Constant 7.518539 *** -0.95281 *** -646.5402 *** OPEN_RATIO_LAOS(-3) -0.16935 *** CAPR_LAOS(-1) /LABOR_LAOS(-1) 0.000422 *** GDPR_LAOS(-1 ) 0.000252 *** TT_LAOS 0.008125 *** ODA_JP_LAOS(-2) /FDI_GFCF_LAOS(-2) 4.996595 *** GDPR_LAOS(-1) 0.628177 *** Dummy 2007 583.9047 *** R-squared 0.992765 0.94133 0.975273 Adjusted R-squared 0.991318 0.932304 0.96703 S.E. of regression 0.026909 0.083136 133.6281 Durbin-Watson stat 2.566495 1.912385 2.570667 Probability of Rejection; 0=*** sig at 99%; <0.05 ** sig at 95% ;<0.10 * sig at 90% Note: OPEN_RATIO= import+ export/GDP ratio; CAPR/Labor= real capital-labor ratio; service=service trade; FDI_GFCF= foreign direct investment-gross fixed capital formation ratio; ODA_JP=Official Development Assistance from Japan; TT= term of trade.

200 Table 9: Impact of Japan's ODA on Myanmar Growth Determination LOG(GDPR_MYANMAR) = C(1)+C(2)*OPEN_RATIO_MYANMAR +C(3)*CAPR_MYANMAR/LABOR_MYANMAR+C(4)*D03 Trade Openness Determination OPEN_RATIO_MYANMAR = C(5)+C(6)*LOG(GDPR_MYANMAR) +C(7)*D03 Formation of Physical and Human Capital CAPR_MYANMAR/LABOR_MYANMAR = C(11)+C(12) *LOG(GDPR_MYANMAR(-1))+C(13)*ODA_JP_MYANMAR /ODA_ALL_MYANMAR +C(10)*D97 Growth Prob. Trade Prob. Capital Prob. Openness Formation Constant 7.598316 *** -3.71105 *** -6.13626 *** OPEN_RATIO_MYANMAR 1.593138 *** CAPR_MYANMAR/LABOR_MYANMAR 0.453826 *** Post Banking crisis 2003 0.839714 ** LOG(GDPR_MYANMAR) 0.498452 *** Post Banking crisis_ 2003 -0.483 *** ODA_JP_MYANMAR /ODA_ALL_MYANMAR 0.859567 *** LOG(GDPR_MYANMAR(-1)) 0.682316 *** Post Asian Financial Crisis_ 1997 Adjustment 0.215377 ** R-squared 0.956601 0.783444 0.794146 Adjusted R-squared 0.948464 0.762819 0.755549 S.E. of regression 0.136624 0.084741 0.099643 Durbin-Watson stat 2.006009 2.078119 1.132634 Probability of Rejection; 0=*** sig at 99%; <0.05 ** sig at 95% ;<0.10 * sig at 90% Note: OPEN_RATIO= import+ export/GDP ratio; CAPR/Labor= real capital-labor ratio; service=service trade; FDI_GFCF= foreign direct investment-gross fixed capital formation ratio; ODA_JP=Official Development Assistance from Japan; ODA_ALL= overall ODA; Post Banking crisis-2003=dummy variable; Post Asian Financial Crisis_1997 Adjustment= dummy variable.

201 Table 10: Impact of Japan's ODA on Vietnam Growth Determination LOG(GDPR_VIETNAM) = C(1)+C(2)*OPEN_RATIO_VIETNAM +C(3)*SERVICEIN_VIETNAM+C(4)*CAPR_VIETNAM/LABOR_VIETNAM Trade Openess Determination OPEN_RATIO_VIETNAM = C(5)+C(6)*LOG(GDPR_VIETNAM) +C(7)*D04 Formation of Physical and Human Capital CAPR_VIETNAM/LABOR_VIETNAM = C(9)+C(10) *ODA_JP_VIETNAM(-1)/ODA_ALL_VIETNAM(-1)+C(11) *LOG(GDPR_VIETNAM(-1))+C(12)*FDI_GFCF_VIETNAM(-1) Growth Prob. Trade Prob. Capital Prob. Openness Formation Constant 10.1003 *** -35.1319 *** -40233.5 *** OPEN_RATIO_VIETNAM 0.164476 *** SERVICEIN_VIETNAM -6.45E-05 *** CAPR_VIETNAM/LABOR_VIETNAM 0.000352 *** LOG(GDPR_VIETNAM) 3.357548 *** Post_Iraqwar2004 -0.49183 *** ODA_JP_VIETNAM(-1)/ODA_ALL_VIETNAM(-1) 1188.618 *** LOG(GDPR_VIETNAM(-1)) 3903.383 *** FDI_GFCF_VIETNAM(-1) 12.62788 *** R-squared 0.992982 0.964713 0.988251 Adjusted R-squared 0.990643 0.959284 0.984334 S.E. of regression 0.024446 0.171809 123.6193 Durbin-Watson stat 1.741336 1.784235 0.931786 Probability of Rejection Probability of Rejection; 0=*** sig at 99%; <0.05 ** sig at 95% ;<0.10 * sig at 90% Note: OPEN_RATIO= import+ export/GDP ratio; CAPR/Labor= real capital-labor ratio; service=service trade; FDI_GFCF= foreign direct investment-gross fixed capital formation ratio; ODA_JP=Official Development Assistance from Japan; ODA_ALL = overall ODA; Post_Iraqwar2004= dummy variable

202 Noted that in econometric estimation, we have deployed the ratio of Japan's ODA-Overall ODA in the system of equations. The estimated coefficient is positive for all relevant country. It would be misled to interpret that Japan's ODA produces ‘positive' while the overall ODA, on the contrary, produces a ‘negative' sign. Since Japan's ODA is part of the overall ODA, thus, the coefficient would imply an increasing ‘net positive value' if and only if Japan's ODA increases over time as a significant donor. 6. Conclusions and Policy Impaclications The CLMV countries have, in recent years, integrated into the global economy, achieving high growth and living standard and liberalizing trade. Being the emerging members of the ASEAN with underexplored abundant natural and human resources, the CLMV have also received large ODA flow from OECD-DAC. Notably, Japan’s ODA has assisted their development programs’ goal, in meeting the United Nations sustainable development objectives. The paper has developed an econometric model consistent with the RTA framework to study the impact of Japan's ODA not only on growth but also on the trade of the CLMV countries. Empirical findings with policy relevance to Japan’s ODA, RTA policy, and domestic reform and crisis management were obtained and analyzed for policy recommendations and compared with previous related studies for robustness. The findings provide useful short-run demand-side evidence where Japan’s ODA has a weak impact on growth and trade. Since the impact of ODA is a long-run supply side, we have added equation on the determination of capital formation, i.e., capital-labor ratio. In the previous sections, we have described the empirical findings from our simultaneous multi-equation model of growth-trade-capital formation Japan's ODA specifically under the economic integration (expenditure and production side) framework to study the impact of Japan’s ODA on the CLMV countries. We have avoided discussion of the donor and development policies and domestic reforms in recipient countries. (See further detail in Mikishima 2010, Pan 2014, WB 2017, and the ADB 2017) The relevance of various regional and global (external) crises have had simultaneously affected growth, trade and capital formation in these countries. The impact of Japan ODA may be better explainable when the demand side is allowed to interact with the supply side. That is to say growth and trade openness in the short-run interact with capital formation in the longer-run. Japan's ODA has significantly induced capital formation, both physical and human capital. Capital structure has produced GDP growth and trade openness. Economic growth and trade openness have indirectly started the next round of simultaneous relationship. Some policy implications of our findings can be derived for the CLMV

203 as a whole or a single country separately, but their main points can be summarized as follows. First, they point out the effectiveness of Japan’s ODA in promoting capital accumulation which simultaneously determined economic growth and trade in the CLMV via different channels. Second, they point out the importance of the different economic structure, governance, suitable human resources, opening-up reform policies and the contagion of regional and global crises in each of the CLMV countries. Our study has taken into account proper country-specific factors and has produced meaningful outcomes. Third, the liberalization of commodity trade, also the FDI and services trade has promoted growth. This liberalization among the CLMV has produced mixed results during the 2001-2015 period. Japan’s ODA which has supported the hard and soft infrastructure projects, as well as the social infrastructure like education, health, and welfare in the CLMV, has long-term effects on growth. The policy implications are as follows. First, the study would recommend the continuation of Japan’s ODA flow to the CLMV. Second, the ODA's flow which helped accumulate the capital of both the physical and human capital should be sufficient for growth in the long-run. Third, Japan’s ODA policy has to consider differentiated characteristics between each recipient country as endowments as well as the country's development planning policy is different. Further study may be as follows. The effectiveness of Japan’s ODA is assessed by econometric estimation with the historical data in our study. An in-depth analysis of a demand-supply interaction in the medium to long-term by a computable general equilibrium model may be a good attempt.

204 Reference (1)
ASEAN (2017), \"ASEAN Free Trade Area,\" http://asean.org/asean- economic-community/asean-free-trade-area-afta-council/. Accessed 2 March 2017. (2)
ADB (2017), “ADB Focus Areas”, https://www.adb.org/focus-areas. Accessed 2 March 2017. (3)
Easterly, W. (2006), The Whiteman Man’s Burden: Why the West’s Efforts to Aid the Rest Have Done So much Ill and So Little Good?, New York: Penguin Press. (4)
Easterly, W. (2007), “Was Development Assistance A Mistake?” American Economic Association Meeting, Chicago, 5-7 January 2007. (5)
EIC (2016), Japan’s ODA Paves the way for Further Economic Ties with the CLMV, Bangkok Post, 7 March 2016. (6)
Frankel, J. A. and D. Romer (1999), “Does Trade Cause Growth?” American Economic Review, 89: 379-399. (7)
IMF (2007), http://www.imf.org/external/pubs/ft/weo/2004/01/index.htm, 2007. (8)
Kitti Limskul
and Nattapong Puttanapong (2017), “Asia’s Social Infrastructure Demand Estimate: The Case of Thailand,\" presentation at the ADB-JICA Joint Side Event at the 50th Annual Meeting of the ADB, Yokohama, Japan. Electronic publishing by the JICA-RI in 2018. (9)
Kydland, F. E. (2006), “Quantitative Aggregate Economics” American Economic Review, 96(5): 1373-1383. (10)
Mikishima, M. and M. Yokoyama (2008), “Japan’s ODA to Mekong River Basin Countries,\" Jetro. (11)
Mikishima, M. (2010), “Japan’s ODA to Mekong River Basin Countries: With Case Studies” in Mikishima, M. (ed) Japan and Korea with the Mekong River Basin Countries, Kagami, BRC Research Report No. 3, Bangkok Research Center, Bangkok, Thailand. http://www.ide.go.jp/English/Publish/Download/Brc/pdf/03_chapter9.pdf (12)
OECD (2017), “International Development Statistics”, http://www.oecd.org/dac/stats/idsonline.htm. Accessed 10 Feb 2017.

205 (13)
Pan, N. ( 2014) , “ Japanese ODA to Asian countries: An empirical study of Myanmar (14)
Compared with Laos, Myanmar, and Vietnam. Accessed 10 Jan 2017. https://www.mof.go.jp/pri/international_exchange/visiting_scholar_program/ ws2014_d.pdf. (15)
Rajan, R. and A. Subramanian ( 2007) , “ Does Aid Affect Governance?” American Economic Association Meeting, Chicago, 5-7 January 2007. (16)
Sala- I- Martin, X. ( 1991) , “ Comment” in Blanchard, O. J. and S. Fischer (eds.), NBER Macroeconomic Annual 1991, Cambridge, MA: MIT Press, 368- 378. (17)
Tran, V. H. ( 1992) , “ Modelling Output Growth: A New Approach” Economics Letters, 38: 279-284. (18)
Tran, V. H. ( 2002a) , “Korea, China and Japan’ s Trade with the World and Its Impact on New Asian Regionalism ASEAN+3” Paper presented at the Joint Conference of the Association of the Korean Economic Studies, Korea Development Institute, and Research Centre for International Economics, Yonsei University, Seoul, 21-22 July 2002. (19)
Tran, V. H. (2002c), “New Asian Regionalism and ASEAN+3 Free Trade Agreement: Theoretical and Empirical Foundation, Policy Challenges and Growth Prospects” Chulalongkorn Journal of Economics (Thailand), 14(3): 366-384. (20)
Tran, V. H. ( 2004) , “ Korea’ s Trade, Growth of Trade and the World Economy in Post- crisis ASEAN+ 3 Free Trade Agreement: An Econometric and Policy Analysis”, Journal of the Korean Economy, 5(2): 73-108. (21)
Tran, V. H. (2005), “Modelling the Impact of China’s WTO Membership on Its Investment and Growth: A New Flexible Keynesian Approach” ( with Comment), in G. S. Heiduk and Karyiu Wong (eds), WTO and World Trade, New York: Physica-Verlag. (22)
Tran, V. H. (2007a), “Foreign Aid and Development in Thailand: Causality and Political Economy,\" Thammasat Economic Journal ( Thailand) , 27( 4) : 127-154. (23)
Tran, V. H. ( 2007b) , “ Causal Empirics and Political Economy of Official Development Assistance and Development in Asia: The Case of Vietnam,\" Journal of Korean Economy, 8(1): 91-119.

206 (24)
Tran, V. H. and K Limskul (2013), “Economic Impact of CO2 Emissions on Thailand's Growth and Climate Change Mitigation Policy: A Modelling Analysis”, Economic Modelling, 33: 351-358. (25)
Tran, V. H. and K Limskul (2017), \"Japan's ODA in Globally Integrated CLMV Economies: Aid Effectiveness and Political Economy,\" Korea and the World Economy, Vol.18, No.3 ISBN 2234-2346, p.299-327 (26)
UNCTAD (2017), “Data Centre”, http://unctadstat.unctad.org/EN/. Accessed 10 Feb 2017. (27)
USDA-ERS (2017), “International Macroeconomic Data Set”, https://www.ers.usda.gov/data-products/international-macroeconomic-data- set/. Accessed 10 Feb 2017 (28)
WB (2017), “What WE do”, http://www.worldbank.org/en/about/what-we- do. Accessed 2 March 2017 (29)
WTO (2017), “Regional Trade Agreements”, https://www.wto.org/english/tratop_e/region_e/region_e.htm. Accessed 2 March 2017.

207 8 Emigrant’s Remittances, Dutch Disease and Capital Accumulation in Mekong Countries72 Hiroyuki Taguchi73 and Ni Lar74 1.
Introduction The latecomer’s economies in Mekong region, namely, Cambodia, Lao PDR, Myanmar and Vietnam have recorded high economic growth in the recent decades. In fact, Lao PDR, Myanmar and Vietnam ( thereafter, CLMV) already graded up their economies towards the “Middle Income” stage in 2010, 2014 and 2009, respectively, according to the “ Income Classifications” by the World Bank. 75 For the upcoming decades, these economies are expected to sustain their economic growth to avoid the “Middle Income Trap”.76 One of the characteristics of these Mekong four economies is that they still have a lot of emigrant workers outside of their countries and depends on the revenues from emigrant’ s remittances77 in their economies. Regarding the economic impacts of the remittances, the literature tells us that they have contributed to poverty alleviation and income equalization from microeconomic aspects. On the other hands, the studies from macroeconomic perspectives often argue that the remittance revenues have led to the “ Dutch Disease” . Then, for the CLMV economies who pursue sustainable growth for the upcoming decades, it could be of great significance to diagnose the macroeconomic impacts of their remittances, for instance, whether or not their dependence on the remittances accompanies the risk of the Dutch Disease. In case that some symptom of the disease was identified, the CLMV should have a strategy to remedy it to ensure their sustainable growth. 72 This paper was presented at the 3rd ASEAN Community Forum at Chulalongkorn University, Session on “Human Resource Landscape in Mainland ASEAN” on August 3, 2017. 73 Saitama University, Japan 74 Saitama University, Japan 75 See the website of https://datahelpdesk.worldbank.org/knowledgebase/articles/378834-how-does- the-world-bank-classify-countries. 76 The concept of “Middle Income Trap” was initially proposed by Gill and Kharas (2007). 77 The remittances in this paper denote “personal remittances”, which is defined as personal transfers and compensation of employees by the six edition of Balance of Payments Manual issued by the International Monetary Fund (IMF).

208 This article examines macroeconomic impacts of emigrant remittances with a focus on the CLMV economies by using a vector auto-regression (VAR) estimation as an analytical framework, and also represents some recommendations on how to mobilize the remittances for productive use, for instance, for human resource development. The rest of the paper is structured as follows. Section 2 illustrates the current situations on the number of emigrant workers and the size of the remittance revenues in the CLMV countries. Section 3 describes the literature review on micro- and macro-economic impacts of remittances, in particular, with a focus on their Dutch Disease effects. Section 4 represents a theoretical framework for analyzing the Dutch Disease effect and the capital accumulation effects of remittances in small open economies. Section 5 conducts empirical studies of remittance impacts in the CLMV economies, containing the descriptions of data for key variables, methodologies for a VAR estimation, and estimation outcomes with its interpretation. Section 6 proposes some policy recommendation on how to mobilize the remittances for productive use. The last section summarizes and concludes. 2.
Emigrant Workers and Their Remittances in CLMV This section displays the current situations on the number of emigrant workers and the size of the remittance revenues in the CLMV countries. First of all, we observe the emigrant flows of the CLMV by their origin and destination in Table 1. The size of emigrants as a percentage of population in 2015 is larger in the total CLMV at 4. 7 percent than in the world average at 3. 3 percent. As an individual country, Lao PDF has the largest ratio, 19.8 percent, among the CLMV. According to “ migration and remittances factbook 2016” published by the World Bank, the CLMV are classified as “ Top 10 emigration countries” in the category of “East Asia and Pacific”.78 When we see their origin and destination in Table 1 again, the destination clearly differs between a group of CLM and Vietnam: the CLM emigrants concentrate on Thailand by around 70 percent, whereas Vietnamese on U.S. and the other develop countries. The remittance flows are reflected by the emigrant flows. Table 2 indicates that the remittances as a percentage of GDP are larger in the total CLMV at 4.9 percent than in the world average at 0. 7 percent in 2015. The remittance growth rate from 2000 to 2015 in the total CLMV by 9. 1 times also exceeds that in the world average by 4. 6 times. Since the remittance growth is higher than the GDP growth, the remittance- GDP ratios have an increasing trend in the total CLMV as well as in the 78 See the website: https://siteresources.worldbank.org/INTPROSPECTS/Resources/334934- 1199807908806/4549025-1450455807487/Factbookpart1.pdf.

209 world. As an individual country among CLMV, Vietnam has the largest remittance- GDP ratio, 6.8 percent, in 2015, and Lao PDR has the highest growth of remittances, 141 times, from 2000 to 2015. When we compare the level of remittances with those of the other foreign currency gains such as net inflows of official development assistance ( ODA) and foreign direct investment (FDI), and natural resources rents (NRR)79 as a percentage of GDP in Table 3, we can see that the remittances are at comparable levels with the other foreign incomes. In particular, the remittances are the top in the list of foreign incomes in Vietnam. 3.
Literature Review and Contribution Regarding the literature of empirical studies on the economic impacts of emigrant’ s remittances, the microeconomic aspects such as poverty alleviation and household incomes have been centered in their researches. From this perspectives, the favorable effects of remittances for the recipient developing economies were identified on poverty ( e. g. , Adams and Page, 2005; Acosta, et al. , 2008; Lokshin, et al. 2010), on income distribution (Barham and Boucher, 1998; Adams and Cuecuecha, 2010) , on school attendance ( e. g. , Cox- Edwards and Ureta, 2003; Gorlich, et al. , 2007) , on financial development ( e. g. , Aggarwal, et al. , 2006; Giuliano and Ruiz- Arranz, 2009; Chowdhury, 2011), and on entrepreneurship of microenterprises (e.g., Woodruff and Zenteno, 2001; Yang, 2005). For the macroeconomic viewpoint, the arguments have been focused on whether remittances would cause the Dutch Disease. The theoretical framework of analyzing the Dutch Disease effect of “capital inflows” in small open economies was usually represented by the Salter-Swan-Corden-Dornbusch model, which was initially demonstrated by Corden and Neary ( 1982) . This model has also been applied to examine the economic impacts of emigrant’ s remittances, since they constitute a major component as an origin of capital inflows. There have been, however, relatively few empirical studies of remittance impacts by using the Dutch Disease model. Acosta, et al. ( 2009) examined the remittance effects in El Salvador by applying a dynamic stochastic general equilibrium model with a Bayesian technique, based on the Dutch Disease concept. They extended the original Dutch Disease model by incorporating an additional transmission mechanism of remittances through labor supply: remittances increase the reservation wage of recipients and, thus, cause a decline in labor supply; a shrinking labor supply accompanies a higher wage that, in 79 The natural resources rents are defined as the sum of oil rents, natural gas rents, coal rents (hard and soft), mineral rents, and forest rents by World Development Indicators of the World Bank.

210 turn, leads to higher production costs and a further contraction of the tradable sector. Through their empirical analyses, they identified the existence of the Dutch Disease effects of remittances: the reallocation of labor away from tradable sectors toward non-tradable sectors. However, Bourdet and Falck (2007) argued, from the viewpoint of different time- horizon, that in the longer- term, emigrants’ remittances could boost capital accumulation through their domestic saving and investment, thereby resulting in an increase in the production of both tradables and non-tradables. They found, from the case study of Cape Verde, that the Dutch Disease effects of remittances was not so large, and suggested that growth- and export- oriented policies could contribute to limiting the Dutch Disease effect. As a comprehensive empirical study, Lartey, et al. (2012) applied the Dutch Disease model to the remittance assessment for 109 developing and transition countries for the period 1990- 2003. They identified their Ditch Disease effects that favored the nontradable sector sacrificing tradable goods production, accompanied with real exchange rate appreciation. Fayad ( 2011) , on the other hand, sampled 27 countries in the 1980s and 28 countries in the 1990s, and identified a transmission channel through which remittances were conductive to the relative growth of exporting industries in the manufacturing sector of recipient economies, contrary to what the standard Dutch Disease theory implied. In this way, even among the limited studies on the application of the Dutch Disease model to emittance evaluation, the theoretical message have remained unsettled in the time horizons between short and long term, and the empirical evidence has also been inconclusive. In this context, this study contributes to enriching evidence on remittance impacts by targeting those of remittance- dependent economies, the CLMV, who pursue sustainable growth by mobilizing the remittances for productive use. Another contribution of this study is to adopt a VAR estimation for the remittance analysis, which enables us to avoid the endogeneity problem of remittance variable. 4.
Theoretical Framework of the Dutch Disease This section describes the theoretical framework for analyzing the Dutch Disease effect of “ capital inflows” in small open economies by the Salter- Swan- Corden- Dornbusch model. We first introduce the basic framework that is composed of “ spending effect” and “ resource movement effect” based on Corden and Neary (1982). Then we add “capital accumulation effect” from the longer-term perspective by following Bourdet and Falck ( 2006) . This model could, of course, be applied to

211 examine the effects of emigrant’ s remittances, since they constitute a major component as an origin of “capital inflows”. In Figure 1, the horizontal axis exhibits non- tradable while the vertical one shows tradable. The curve P- P represents the initial transformation curve between tradable and non-tradable. Point A is an initial equilibrium, where the transformation curve is tangential to the social indifference curve ( not drawn) and the slope of the curves, i.e., the relative price of non-tradable to tradable, is fixed at that point. The transformation curve shifts upwards to P-PF with the introduction of the capital inflows ( emigrant’ s remittances in this study) shown at point F, since the supply of non-tradable is constant and the availability of tradable expands with higher disposal income. There would be excess demand for non- tradable with unchanged relatives price of non- tradable to tradable shown at point A’ , if we assume positive income elasticity of non- tradable. The price of non- tradable, therefore, has to go up to clear the market, and the relative price of non- tradable to tradable also rises, since the price of tradable is determined in the world market. This effect is referred to as an appreciation of real exchange rate ( spending effect) . The rise of relative price, then, encourages the movement of production factors from the tradable sector to the non- tradable sector, and leads to an expansion in the output of non- tradable and a decline in that of tradable from point A’ to point B (resource movement effect). Bourdet and Falck ( 2006) added the following story from the longer- term perspective. They considered the role of capital accumulation, and argued that the transformation curve could shift further towards P’ - P’ when an economy utilized capital inflows for domestic capital accumulation. As a consequence, the relative price of non-tradable might be expected to fall from point B to point C, thereby facilitating the recovery of tradable sector. Thus, the “ capital accumulation effect” might offset or mitigate the economic damages caused by original Dutch Disease effect. To sum up, the basic theory tells us that capital inflows ( emigrant’ s remittances in this study) reduce the production of tradable through real currency appreciation by the sectoral resource movement. In the longer-term, however, capital inflows would lead to the increase in the outputs of both tradable and non- tradable due to capital accumulation, i. e. , the intertemporal resource movement. In short, capital inflows are not compatible with economic growth under the Dutch Disease, but could be friendly with growth under the capital accumulation in the longer-term.

212 5.
Empirics This section represents empirics for examining sectoral and intertemporal impacts of emigrant’s remittances with a focus on the CLMV economies under the aforementioned theoretical framework, containing the description of the data for key variables, methodologies for a VAR model estimation, and the estimation outcomes with its interpretation. 5.1
Data for Key Variables At the beginning, we identify the following economic variables for a VAR model estimation to examine sectoral and intertemporal impacts of emigrant’s remittances based on the theoretical framework in Section 4. The variable of emigrant’s remittances (roy) is expressed as a percentage of Gross Domestic Product (GDP). The data are retrieved from the item “Personal remittances, received (% of GDP)” of World Development Indicators (WDI) published by the World Bank.80 For examining sectoral impacts of the remittances, namely, the Dutch Disease effect, consumer prices (cpi) and manufacturing-service ratio (mos) are introduced to investigate “spending effect” and “resource movement effect”, respectively. The consumer prices are assumed to be a proxy of real exchange rate. The reason for not using exchange rate is that the CLMV have adopted peg system rather than floating one as their currency regimes for most of the sample periods, according to Ilzetzki et al. (2011). Frankel (2010) argued in the context of Dutch Disease that the real currency appreciation takes the form of money inflows and inflation if the country has a fixed exchange rate, whereas taking the form of nominal currency appreciation if the country has a floating exchange rate. Following this argument, consumer prices are adopted instead of exchange rate. The data is retrieved from the item “Consumer price index (2010 = 100)” of WDI. The manufacturing- services ratio, for assessing “resource movement effect”, is assumed to be a proxy of tradable-nontradable production ratio following the empirical method of Lartey, et al. (2012). The ratio is derived by dividing “manufacturing in value-added term” by “services in value-added one” in the database of UNCTAD Stat.81 The spending effect is identified if consume prices are positively affected by the remittances, while the resource movement effect is confirmed if the manufacturing-services ratio is negatively influenced by the remittances. 80 See the website: http://data.worldbank.org/. 81 See the website: http://unctadstat.unctad.org/EN/.

213 To assess intertemporal impacts of the remittances, namely, capital accumulation effect presented by Bourdet and Falck (2006), the investment- consumption ratio (ioc) is introduced in the estimation model. The ratio is produced by dividing “gross fixed capital formation” by “final consumption expenditure” in the database of UNCTAD Stat. The capital accumulation effect is verified if the investment-consumption ratio is positively affected by the remittances. Another category of variables are those to control their sectoral and intertemporal impacts and to extract pure effects of the remittances. Those are net inflows of official development assistance (oda), foreign direct investment (fdi) and natural resource rents (nrr), all of which are expressed at a percentage of GDP. These variables should be contained as control variables since they also have the Dutch Disease effect and capital accumulation effect as the components of capital inflows as described in Section 4, and also since these components have some weights comparable to received remittances in the CLMV economies as shown in Table 3. The data are retrieved from the items of “Net ODA received (% of GDP)”82, “Foreign direct investment, net inflows (% of GDP)” and “Total natural resources rents (% of GDP)” of WDI, respectively. Figure 2 displays the overviews of three key variables: remittance-GDP ratio (roy), manufacturing-services ratio (mos) and investment-consumption ratio (ioc) in the CLMV. It appears by rough observation that there are no clear correlation between remittance-GDP ratio, and manufacturing-services ratio and investment-consumption ratio. It might be because those endogenous variables of manufacturing-services and investment-consumption ratios are also affected by the other components of capital inflows. Thus, the variable interactions should be statistically tested by a more sophisticated manner, i.e., a VAR model estimation, by containing necessary control variables in the following Section. For the VAR estimation, we construct a panel data with the four CLMV economies for the period from 1984 to 2015.83 5.2
Methodologies for a VAR Model Estimation We herein conduct a VAR model estimation. The reason why we adopt a VAR model for our remittance assessment is that the VAR model allows for potential endogeneity between the variables of concerns, and also for tracing out the dynamic responses of variables to exogenous shocks overtime. 82 The WDI have only “Net ODA received (% of GNI)”, and so the variable is recalculated by replacing GNI with GDP. 83 The data availability of the remittances in WDI differs in the CLMV: Cambodia for 1994-2015, Lao PDR for 1984-2015, and Myanmar and Vietnam for 2000-2015.

214 We now specify a VAR model with panel data for estimation in the following way.  
 
   
 (1) where  is a column vector of the endogenous variables with country i and year t, i. e. ,  
 

 ′ for examining the Dutch Disease effect, and  
 
 ′ for examining the capital accumulation effect,  is a vector of the control variable, i. e. ,  
 

′,  is a constant vector, each of  and  is a coefficient matrix,  is a vector of the lagged endogenous variables, and  is a vector of the random error terms in the system. The lag length ( - 1) is selected by the Schwarz information criterion with maximum lag equal to (-2) under the limited number of observations. Based on the VAR model estimation specified in ( 1) , we examine the Granger causalities and impulse responses from remittance- GDP ratio ( roy) to consumer prices (cpi) and manufacturing-services ratio (mos) for the analysis of the Dutch Disease effect, and those from remittance- GDP ratio ( roy) to investment- consumption ratio (ioc) for the analysis of the capital accumulation effect. When the causalities and impulses response from roy to cpi and mos are identified ( positively to cpi and negatively to mos), we could argue that the CLMV economies have suffered the Dutch Disease effect from their received remittances. If the causality and impulse response from roy to ioc are confirmed in positive ways, we could argue that the CLMV have enjoyed the capital accumulation effect from their remittances. 5.3
Estimation Outcomes and Its Interpretation Table 4, Table 5 and Figure 3 respectively report estimation outcomes of the VAR model, the bilateral Granger causalities and the impulse responses.84 Regarding the Granger causalities shown in Table 5, as far as the analysis of Dutch Disease effect is concerned, it was only the causality from remittance- GDP ratio ( roy) to manufacturing- services ratio ( mos) that was identified at the conventional ( 95 percent) level of significance. Considering the estimated VAR model in Table 4, this causality was supposed to be a “ negative” one. As for the analysis of capital accumulation effect, the causality from remittance-GDP ratio (roy) and investment- consumption ratio ( ioc) was verified, but at the weak ( 90 percent) 84 In the VAR model estimation in Table 4, the constant term was excluded since its parameter was not significant and its inclusion made the model performance get worse.

215 level of significance. The causality was also supposed to be a “negative” one judging from the estimated model in Table 4. The impulse response analysis was focused on the two cases where the Granger causalities were identified above: the one from remittance- GDP ratio ( roy) to manufacturing-services ratio (mos) and the other from remittance-GDP ratio (roy) to investment- consumption ratio ( ioc) . According to Figure 3, the manufacturing- services ratio (mos) responded negatively to the shock of remittance-GDP ratio (roy) within a 95 percent error band after two- year lag, and the investment- consumption ratio (ioc) also responded negatively to that shock but within a 90 percent error band. The implications of the estimation outcomes above are summarized as follows. First, the CLMV might suffer the Dutch Disease effect from their received remittances, judging from the causality and impulse response from remittance- GDP ratio to manufacturing-services ratio. It should be, however, noted that an increase in remittances did not accompany real exchange rate appreciation in terms of the hike of consumer prices, judging from no causality from the remittances- GDP ratio to consumer prices. Thus, the usual Dutch Disease mechanism of “spending effect” and “ resource movement effect” suggested by Corden and Neary ( 1982) might not be applicable to the CLMV case. Instead, the Dutch Disease effect in this case could be interpreted such that received remittances shrank tradable sector directly through a decline in labor supply as Acosta, et al. (2009) suggested as an additional transmission mechanism ( See Section 3) . Second, for the intertemporal impact of received remittances, capital accumulation effect might be deteriorated by the remittances in the CLMV, although the negative causality and impulse response from remittance- GDP ratio to investment-consumption ratio was not so robust. 6.
Recommendation: To Mobilize Remittances for Productive Use The precious section implied that the received remittances might cause the Dutch Disease effect and even deteriorate capital accumulation effect in the CLMV. This section proposes some policy recommendation on how to reform the remittance flows in such the received remittances could be utilized for capital accumulation in manufacturing sector in the CLMV. Some countries with high dependence on migrant remittances have ever employed a large variety of policy measures to enhance the development impact of remittances, and international organization such as the World bank have also proposed different policy options and recommendations on mobilizing remittance- earnings for productive use. Carling (2004) categorized these policy measures into the direct and indirect frameworks. In the direct framework, the home government for emigrants

216 straightly intervenes the remittance flows for their public use. The typical example is that the government imposes taxes or levies on remittance transfers and to use their revenues for public infrastructure. The second scheme is an indirect support by the government for remittance flows to be used for private investment. The style of the government support is, for instance, to create such frameworks as the sale of remittance bonds, opening of foreign currency and premium interest rate accounts for remittances, promoting remittance transfers through microfinance institutions, etc. When we consider the cases of the CLMV, we focus on the latter indirect scheme, since the direct intervention might not be suitable for the CLMV due to their weak government governance. According to the Worldwide Governance Indicators by the World Bank in 201585, the CLMV shows the rankings far behind the 215 countries in the world in terms of each item: control of corruption, government effectiveness, political stability and absence of violence/terrorism, regulatory quality, rule of law and voice and accountability (See Table 6). Regarding the indirect scheme to mobilize the remittances, Pant ( 2008) introduced the progressive cases of South Asian countries, i. e. , Bangladesh, India, Pakistan and Sri Lanka: their governments have devised a framework in which emigrants are granted such facilities as a) maintenance of bank accounts in both foreign and local currencies without tax burden; b) investments in securities and shares, and deposits with local firms and companies; and c) investments in immovable properties in their home countries. To be specific, India has shifted remittance flows from informal channels to formal ones by eliminating the black- market premium. Bangladesh has issued “ Wage Earners’ Development Bond” to attract domestic investment from remittance revenues. 86 Pant ( 2008) also proposed “ remittance- linked microfinancing services” under the economies in which banking system is still premature at its development stage. His idea could be illustrated in Figure 4: a) emigrant workers save their money in the branch banks of their home country; b) the banks of home country invest their saved money for microfinance institutions; and c) the microfinance institutions finance the small and family business for returned workers and/or family members of emigrant workers. The framework of South Asian countries and the remittance- linked microfinancing services proposed by Pant ( 2008) could be applicable and advantageous to the CLMV for mobilizing their received remittances for more 85 See the website: http://data.worldbank.org/data-catalog/worldwide-governance-indicators. 86 Apart from the South Asian cases, World Bank (2016) introduced the case of Vietnam in which Vietnamese overseas have been permitted to buy houses in Vietnam by the change in law in 2015.

217 productive use. First, the framework would contribute to capital accumulation in the sense that the savings from remittance revenues could be utilized for private investment. Second, the framework would also contribute to an effective usage of human capitals by creating job opportunities in small and local businesses for returned emigrant workers. The knowledge and skills gained by emigrant workers in foreign countries could be effectively utilized for local businesses in their home countries. Third, the remittance-linked microfinancing services are useful for the CLMV. Most of rural part of the CLMV is still out of access from formal banking networks, and easy access to microfinancing services would remove financial bottlenecks for local entrepreneurs. 7.
Concluding Remarks This paper examined the sectoral and intertemporal impacts of international emigrant remittances by using a vector auto-regression (VAR) estimation focusing on the CLMV economies. The reason for targeting the CLMV countries is that they have still depended largely on remittance- earnings from their emigrant workers in their economies, and that the macroeconomic impacts of received remittances would be critical for their sustainable growth. The empirical study identified the decline in manufacturing- service ratio ( the Dutch Disease effect) as a sectoral effect of remittances, and also the decline in investment- consumption ratio ( the deteriorated capital accumulation effect) as their intertemporal effect, judging from the causalities and dynamic responses from remittances to both ratio in the VAR estimation outcomes. The strategic implication is that the CLMV countries should establish a framework to mobilize their remittance- earnings for more productive use. The bridging scheme between savings from remittance revenues and private investment such as remittance- linked microfinancing services should be highly recommended in the CLMV economies.

218 References (1)
Acosta, P.A., Calderon, C., Fajnzylber, P. and Lopez, H. (2008). What is the Impact of International Migrant Remittances on Poverty and Inequality in Latin America? World Development, 36, 89-114. (2)
Acosta, P.A., Lartey, E.K.K. and Mandelman, F.S. (2009). Remittances and the Dutch Disease. Journal of International Economics, 79, 102-116. (3)
Aggarwal, R. , Demirguc- Kunt, A. and Peria, M. ( 2006) . Do Workers’ Remittances Promote Financial Development? World Bank Policy Research Working Paper No. 3957, World Bank. (4)
Adams, R. and Cuecuecha, A. (2010). The economic impact of international remittances on poverty and household consumption and investment in Indonesia. World Bank Policy Research Working Paper 5433. Washington, DC: World Bank. (5)
Adams, R. and Page, J. (2005). Do International Migration and Remittances Reduce Poverty in Developing Countries? World Development, 33, 1645-69. (6)
Barham, B. and Boucher, S. ( 1998) . Migration, remittances and inequality: estimating the net effects of migration on income distribution. Journal of Development Economics, 55(3), 307–331. (7)
Bourdet, Y. and Falck, H. (2006). Emigrants’ Remittances and Dutch Disease in Cape Verde. International Economic Journal, 20(3), 267-284. (8)
Carling, J. (2004). Policy Options for Increasing the Benefits of Remittances. Paper presented at the Ninth International Metropolis Conference, Geneva, September 27-October 1. (9)
Chowdhury, M. (2011). Remittances Flow and Financial Development in Bangladesh. Economic Modelling, 28 (6), 2600-8. (10)
Corden, W.M. and Neary, J.P. (1982). Booming sector and de- industrialization in a small open economy. Economic Journal, 92, 825–848. (11)
Cox-Edwards, A. and Ureta, M. (2003). International migration, remittances and schooling: evidence from El Salvador. Journal of Development Economics, 72(2), 429–461. (12)
Fayad, G. (2011). Remittances: Dutch disease or export-led growth? OxCarre (Oxford Centre for the Analysis of Resource Rich Economies) Research Paper 57.

219 (13)
Frankel, J. (2010). The Natural Resource Curse: A Survey. NBER Working Paper Series No.15836. (14)
Gill, I. and Kharas, H. (2007). An East Asian Renaissance. IBRD and World Bank. (15)
Giuliano, P. and Ruiz- Arranz, M. ( 2009) . Remittances, financial development and growth. Journal of Development Economics, 90( 1) , 144– 152. (16)
Gorlich, D., Omar, M. and Christoph, T. (2007). ‘Explaining Labour Market Inactivity in Migrant- Sending Families: Housework, Hammock, or Higher Education? Kiel Working Paper 1391, Kiel Institute for the World Economy, Kiel, Germany. (17)
Ilzetzki, E., Reinhart, C.M. and Rogoff, K.S. (2011). The Country Chronologies and Background Material to Exchange Rate Arrangements into the 21st Century: Will the Anchor Currency Hold. In http://personal.lse.ac.uk/ilzetzki/data/ERA-Country_Chronologies_2011.pdf. (18)
Lartey, E. K. K. , Mandelman, F. S. and Acosta, P. A. ( 2012) . Remittances, Exchange Rate Regimes and the Ditch Disease: A Panel Data Analysis. Review of International Economics, 20(2), 377-395. (19)
Lokshin, M. , Bontch- Osmolovski, M. and Glinskaya, E. ( 2010) . Work- related migration and poverty reduction in Nepal. Review of Development Economics, 14(2), 323–332. (20)
Pant B. ( 2008) . Mobilizing Remittances for Productive Use: A Policy- oriented Approach. NRB Working Paper No. 4, Nepal Rastra Bank. (21)
Taylor, J. E. , Mora, J. , Adams, R. Jr. and Lopez- Feldman, A. ( 2005) . Remittances, inequality and poverty: evidence from rural Mexico. Working Paper 05–003. Davis, CA: Department of Agricultural Economics, University of California. (22)
Woodruff, C. and Zenteno, R. (2001). Remittances and Microenterprises in Mexico. Graduate School of International Relations and Pacific Studies Working Paper No. 43, University of California, San Diego. (23)
World Bank (2016). Migration and Remittances: Recent Developments and Outlook.

220 (24)
Yang, D. (2005). International Migration, Human Capital, and Entrepreneurship: Evidence from Philippine Migrants’ Exchange Rate Shocks. World Bank Policy Research Working Paper No. 3578, World Bank.

221 Table 1 International Migrant Stock by Origin and Destination in CLMV (2015) Population Destination of Emigrants (thousand) (thousand) Origin of World 1st Coutry 2nd County 3rd Country Emigrants 15,578 (% of Pop.) (% of World) Cambodia 6,802 1,187 (7.6) (% of World) (% of World) 53,897 France Lao PDR 93,448 1,345 (19.8) Thailand U.S. 63 (5.3) 169,725 805 (67.8) 167 (14.1) France Myanmar 2,882 (5.3) Thailand 43 (3.2) 969 (72.0) U.S. Saudi Arabia Vietnam 2,559 (2.7) 200 (14.9) 203 (7.0) Canada CLMV Thailand Malaysia 183 (7.2) 1,978 (68.6) 252 (8.9) Australia U.S. 227 (8.9) 1,303 (50.9) 7,973 (4.7) - World 7,349,472 243,700 (3.3) - Source: Based on International migrant stock 2015 by United Nations: http://www.un.org/en/development/desa/population/migration/data/estimates2/estimates15.sht ml

222 Table 2 International Remittances Received by CLMV Cambodia remitttances, mil. USD 2000 2015 2015/2000 Lao PDR GDP, mil. USD 103 397 3.9 Myanmar 3,654 18,050 4.9 Vietnam remitttances / GDP, % 2.8 2.2 - CLMV remitttances, mil. USD 1 93 1,731 12,369 140.9 World GDP, mil. USD 0.0 0.8 7.1 remitttances / GDP, % 102 387 - remitttances, mil. USD 8,905 62,601 3.8 1.1 0.6 7.0 GDP, mil. USD 1,340 13,200 - remitttances / GDP, % 33,640 193,599 9.9 remitttances, mil. USD 4.0 6.8 5.8 1,545 14,077 - GDP, mil. USD 47,930 286,619 9.1 remitttances / GDP, % 3.2 4.9 6.0 remitttances, mil. USD 121,312 552,998 - 33,551,372 74,292,304 4.6 GDP, mil. USD 0.4 0.7 2.2 remitttances / GDP, % - remitttances, mil. USD GDP, mil. USD remitttances / GDP, % Sources: Based on World Development Indicators by World Bank Data: http://data.worldbank.org/

223 Table 3 Remittances, ODA, FDI and Natural Resource Rents (NRR) Received by CLMV (Percentage of GDP, 2015) Remittances ODA FDI NRR Cambodia 2.2 3.8 9.4 2.0 Lao PDR 0.8 3.8 8.7 11.6 Myanmar 0.6 1.9 6.5 4.8 Vietnam 6.8 1.6 6.1 2.6 Sources: Based on World Development Indicators by World Bank Data: http://data.worldbank.org Figure 1 Theoretical Framework of the Dutch Disease Tradable Resource Movement Effect P’ C Long-term Effect PF A’ B by capital Accumulation P A Spending Effects F Income Inflows Non-tradable P P’ Sources: This diagram is based on Corden and Neary (1982) and Bourdet and Falck (2006).

224 Figure 2 Overviews on Key Variables in CLMV Cambodia roy mos ioc 4.00 60.00 3.00 40.00 2.00 1.00 20.00 0.00 0.00 Lao PDR roy mos ioc 1.50 60.00 1.00 40.00 0.50 20.00 0.00 0.00 2000 2000 2001 2001 2002 2002 2003 2003 2004 2004 2005 2005 2006 2006 2007 2007 2008 2008 2009 2009 2010 2010 2011 2011 2012 2012 2013 2013 2014 2014 2015 2015 Myanmar ioc roy mos 3.00 2000 2000Vietnamioc 60.00 2.00 2001 2001 40.00 1.00 2002 2002roy mos 20.00 0.00 2003 2003 0.00 2004 2004 10.00 2005 2005 60.00 8.00 2006 2006 40.00 6.00 2007 2007 20.00 4.00 2008 2008 0.00 2.00 2009 2009 0.00 2010 2010 2011 2011 2012 2012 2013 2013 2014 2014 2015 2015 Source: World Development Indicators (World Bank) and UNCTAD Stat.

225 Table 4 Estimated VAR Model Dutch Disease Effect roy cpi mos roy -1 0.854*** 0.540 -0.004 ** cpi -1 [13.026] [0.855] [-2.334] mos -1 -0.003 1.026*** oda [-1.353] [40.385] -0.000 fdi 4.067 [-1.428] nrr 1.891* [0.378] 1.052*** [1.695] -13.781 [33.816] adj. R^2 -0.006 [-0.894] 0.022 [-0.702] 0.118 [0.511] 0.006 [0.434] -0.000 [0.231] 0.537** [-0.018] 0,019 [2.482] 0.000 [0.875] 0.982 [1.404] 0.811 0.968 Capital Accumulation Effect roy ioc roy -1 0.889*** -0.004* ioc -1 [14.777] [-1.704] oda 1.058*** fdi 0.893 [22.916] nrr [0.917] 0.066 -0.421 [0.935] adj. R^2 [-0.281] 0.000 0.013 [0.639] [0.534] -0.000 0.002 [-0.510] [0.084] 0.886 0.821 Note: * * * , * * , * denote rejection of null hypothesis at the 99% , 95% and 90% level of significance, respectively. Table 5 Granger Causality Tests mos Lags Null Hypothesis Chi-sq roy & mos 1 roy does not Granger Cause mos 5.448** 1 mos does not Granger Cause roy 2.873* roy & cpi 1 roy does not Granger Cause cpi 0.731 1 cpi does not Granger Cause roy 1.833 cpi & mos 1 cpi does not Granger Cause mos 2.040 ioc 1 mos does not Granger Cause cpi 0.142 Chi-sq roy & ioc Lags Null Hypothesis 2.906* 1 roy does not Granger Cause ioc 0.840 1 ioc does not Granger Cause roy Note: * * * , * * , * denote rejection of null hypothesis at the 99% , 95% and 90% level of significance, respectively.

226 0.010 Figure 3 Impulse Responses to Remittance Shock 8 0.000 -0.010 Impulse Response on Manufacturing-Service Ratio (mos) -0.020 -0.030 1234567 -0.040 -0.050 0.020 Impulse Response on Investment-Consumption Ratio (ioc) 8 0.010 1234567 0.000 -0.010 -0.020 -0.030 -0.040 -0.050 -0.060 -0.070 -0.080 -0.090 Note: The coarse and fine dotted lines denote a 90 and 95 percent error band, respectively, over 8-year horizons.

227 Table 6 World Governance Indicators in CLMV in 2015 Ranking in 2015 Cambodia Lao PDR Myanmar Vietnam among 215 countries 183 168 174 127 94 Control of Corruption 109 139 Government Effectiveness 156 133 188 113 182 Political Stability and Absence of 119 84 189 127 Violence/Terrorism Regulatory Quality 136 160 194 Rule of Law 173 156 193 Voice and Accountability 166 195 177 Average 156 149 186 Sources: Worldwide Governance Indicators (World Bank) Figure 4 Idea on Remittance-Linked Microfinancing Services Foreign Countries Home Country Banks of Home Country Microfinance Institution Savings Finance Starting Small Business Emigrant Workers Returning Workers (Source) produced by author

228 PART 3 Conclusion and Policy Recommendations Piti Srisangnam87, Ph.D. This research has been conducted with cooperation between academics from countries in mainland ASEAN countries, comprising Cambodia, the Lao People's Democratic Republic, Myanmar, Thailand, and Vietnam, as well as from Japan. It is divided into two sections: Section 1 sketches an overall picture of human resources in the ASEAN member countries of sub- regional mainland Southeast Asia and the agreement on the movement of human resources within the ASEAN community; Section 2 provides an analysis of data and policies using statistical data and an economic model, giving insight into macroeconomic situations which will likely occur when there is a transfer of human resources and changes in human capital formation in mainland ASEAN countries. The overall picture of the human resource landscape in mainland ASEAN is summarized as follows: 1. In respect to population structure, mainland ASEAN countries, except for Thailand, present a high potential in human resource growth. The population growth rate of Cambodia, Laos, Myanmar, and Vietnam means that they are not expected to experience the problems of an aging society until the year 2050. The labor force of these four ASEAN members has been continually expanding due to continuous economic expansion, particularly since the 1990s, when they started instituting serious economic reforms, resulting in their achieving lower middle income status according to the definition given by the World Bank. This, in turn, has allowed for the continual development of social security systems, especially health care systems with improved medical treatment. These countries can be expected to enjoy a ‘demographic dividend’ between 2030 and 2050. Another phenomenon that these four mainland ASEAN countries have in common is a decrease in the Labor Force Participation Rate, since a certain percentage of the growing labor force has come into the labor market more slowly. This is because 87 Faculty of Economics, Chulalongkorn University. ASEAN Studies Center, Chulalongkorn University.

229 they now receive more education, especially obligatory education that extends to secondary education. 2. In the case of Thailand, the situation of population growth is reversed. When we consider the population pyramid, it becomes evident that Thailand is an aging society, with a total dependency ratio at 40 percent in 2015 (Youth Dependency Ratio = 25.2% and Elderly Dependency Ratio = 14.8%). Thailand’s children aged between 0-14 years are at 6,503,133 persons, or 16.93 percent of the total population, whereas senior citizens of over 65 years old are at 7,238,248 persons, which is 10.58 percent of the entire population in 2017 (CIA Factbook, 2018). It should be noted that while the total population of Thailand has been growing at a sharply reduced rate, reaching as low as 0.4 percent per year (2017 data), the proportion of senior citizens has been rising at an alarmingly high rate as people aged 6 0 and over are increasing at approximately 5 percent each year, with the population aged 80 and over increasing by over 6 percent per year. It is estimated that the year 20 1 9 will be the first year in which Thailand will have more seniors (aged 60 and over) than children (aged below 15 years) (Institute for Population and Social Research, Mahidol University, 2017). In the near future, the number of senior citizens will continue to surge and Thailand will become a full-fledged aged society, perhaps as soon as 2021, when the percentage of the population aged 6 0 and over exceeds 2 0 percent of the total population. This is because the population cohort born during 1 9 6 3 - 1 9 8 3 , the generation which experienced high birth rates of over a million persons per year, will soon become senior citizens (right now, this portion of the population is aged between

230 3 5 - 5 5 years) . If no significant changes occur or measures are taken to tackle this problem, Thailand will most likely enter the super-aged society stage in 2031, when the population over 6 0 years old reaches 2 8 percent of the total Thai population ( Institute for Population and Social Research, Mahidol University, 20 1 7 ) . Towards the end of the 2030s, this major population cohort will be aged 55-75 years, and about a third of the population will be older than 60 years. As a result, Thailand will face a very important challenge in regards to human capital: a shortage of labor resulting from this demographic structure. The World Bank has forecasted that a decrease of the Thai labor force in this manner will surely deteriorate its economic performance and growth (World Bank, 2016). 3. Every mainland ASEAN country has established public policies at the national level in the form of national demographic plans and at the level of ministries and government agencies with a focus on the development of human resources through education management systems which concentrate on the improvement of the population in each country in each stage of life, ranging from pre- schooling, schooling, and on the improvement of labor quality when people go into working age. Also, all mainland ASEAN countries have attempted to establish a system of social welfare including health care. In the case of Thailand, in addition to such measures, it has already started making preparations for becoming an aging society. Nevertheless, the researchers of this study found that the implementation of such policies in those countries have not fully achieved their objectives due to a lack of coordination between state agencies regarding planning and implementation. Additionally, it has been found that analysis regarding improvements to the implementation of other policy phases has not been comprehensively conducted and some projects or policies are overlapping or sometimes contradictory. More importantly, coordination and collaboration has not been concrete enough, and what has been worked on is all under the framework of the ASEAN Socio-Cultural Community (ASCC), which stresses the enhancement of human resources through collaboration for the improvement of social welfare and social justice within the region. 4. Although every mainland ASEAN country has instituted policies on human resources towards achieving the Millennium Development Goals (MDGs) set by the United Nations at a satisfactory level (albeit some goals have yet to be attained by some countries) , advancement in social aspects, enhancement of human capital, and increase in the Human Development Index (HDI) have occurred in every member country. Furthermore, every member country has formulated policies establishing

231 goals in accordance with the United Nations’ Sustainable Development Goals (SDGs). Yet, the researchers of this study have discovered that progress in poverty reduction and human development has been uneven across the region and in particular among ethnic groups, especially between the diverse ethnic groups of the seven states in Myanmar where the majority is not Burmese and the seven regions where it is. In the case of the Lao People's Democratic Republic, progress has been clustered around Vientiane, the capital city, whereas ethnic groups living in other areas, excluding the Lao- Tai group, have not been developed sufficiently. This underdevelopment also applies to the central area and high-mountain areas in Vietnam. 5. As regards the development of human capital, every mainland ASEAN country has established policies for the development of the education system, especially education in the form of Technical and Vocational Education and Training ( TVET) . However, this research found that the current formal and vocational education being provided lacked the necessary personnel and resources to boost investor and employer confidence, and it is felt that it will take some time before the education systems in mainland ASEAN countries can produce graduates that can fulfill the demand for human resources. As a result, new policies for the development of human capital which include effective and efficient investments in education, health, labor and migration, as well as social development mechanisms that increase the levels and equitable distribution of human and social capital are strongly recommended. The concept of ‘ lifelong learning’ can be seen as a key factor in addressing the education needs of the region. Human capital development should start before the schooling stage to ensure that the later stages of education produce a capable and competent workforce that can be productive in the further stages of life. 6. Although the development of human resources is brought up regularly and set as an urgent need for developing countries, budget limitations are often a serious constraint on improvement of the education system, the development of human resources and learning material, and an amelioration of social welfare. Normally, the budget used to develop human resources which are regarded as important public products and services comes from the country budget where the government of each country has allocated. However, being a part of the world's economic system places other multi- faceted financial burdens on governments of developing countries. Those burdens include investment in improving basic infrastructure and addressing serious

232 environmental problems which occur as a result of the poor development planning, causing government expenses to increase. Meanwhile being a part of the global economic system involves being a committed member of the World Trade Organization, ASEAN, and/ or other trade agreements, which lowers state revenues from tariffs, often a major source of government income for many developing countries. Likewise, policies decreasing income tax rates, both for juristic persons and individuals, have been required in order to bring them in line income tax rates in other ASEAN member countries. These financial factors result in governments having more fiscal liabilities and more constraints on budget allocation for the social welfare system and the improvement of human resources. 7. Since virtually all mainland ASEAN member countries are considered developing countries which have only recently held the status of least developed country, the role of international assistance, known as “ Official Development Assistance” (ODA), especially from developed countries such as Japan, is important for improving the economy. In order to study relationships between assistance and vital macroeconomic variables such as economic growth rates, international trade variables, and variables associated with the development of human capital, the second section of this research employed the Computable General Equilibrium Model (CGE). Statistical data and the model showed that in the short term, international aid had little significant direct impact on expansion of international trade and the expansion rate of the economy in Cambodia, the Lao People's Democratic Republic, and Myanmar. Meanwhile, for Vietnam, assistance was found to have a positive impact ( though not in a high proportion) on the economic growth rate, but had significant negative effects on expansion of international trade. This is because the assistance has boosted domestic production, which reduces reliance on imports. Meanwhile, its trading partner countries have tended to be less diverse. Nonetheless, in the long term assistance can be expected to affect the capital formation of countries who receive it in regards to both physical capital and human capital. 8. The CLMV countries ( Cambodia, Laos, Myanmar and Vietnam) tend to have limited industrial production without a particularly large service sector. The yields of their agricultural sector are also relatively low, resulting in a large percentage of the workforce from these countries going to work overseas. Whether working abroad legally or not, these workers often send considerable income back to their homeland in the form of remittances, increasing the quality of life of their families.

233 However, the high transfer value from abroad puts mainland ASEAN countries at risk of the ‘Dutch disease’. The foreign currency that flows in from emigrant workers may be in considerable enough to make the currency of the destination country so strong that it affects export capability and economic expansion is affected as the main currency is overvalued. Based on the statistical data and economic- dimensional model in the second part of this research, the CLMV might suffer the Dutch disease effect from received remittances, judging from the causality and impulse response from Remittance- GDP ratio to Manufacturing- Service ratio. Nevertheless, the risk of this occurring is still not high, as destination currency values have not become far stronger and simultaneously have not raised the cost of living of the population in the country so significantly the consumers have become heavily burdened ( as can be seen from no causality from the Remittance- GDP ratio to Consumer Prices) . However, the Dutch disease effect that could occur in the CLMV counties would happen in the form of a contraction in manufacturing for international trade as a result of a decreased demand for labor. Another possible effect is that capital accumulation might deteriorate due to remittances, although the negative causality and impulse response from the Remittance-GDP ratio to Investment-Consumption ratio is not so robust. 9. As mentioned above, it can be concluded that a defining feature of the human resource landscape in mainland ASEAN countries is what researchers would call ‘triple disparities,’ which include three levels of disparities: i.
Disparity between demand and supply for human resources in each country ii.
Disparity between problems of skilled labor and unskilled labor iii.Disparity between populations in different geographical areas It was found that in terms of the skilled labor market, the CLMV countries, four of which have lower middle income or developing economies, have a common demand problem, i. e. a problem attracting and retaining entrepreneurs. In other words, businesspeople and investors have not been able to acquire adequate skilled labor with knowledge and expertise in these countries or hired workers often leave their jobs as they are able to earn higher payment when changing jobs. Therefore, finding people with knowledge and capabilities corresponding to needs and working environment, as well as maintaining their loyalty, is difficult for businesses. The problem is one of excess demand. In contrast to skilled labor, unskilled labor in the CLMV countries is in excess supply, or a situation where the number of job applicants is larger than the number of job openings. Currently, the problem with excess supply is a key factor driving workers in these countries to emigrate to work abroad, despite the fact that their

234 domestic income rate is no longer different from that in other neighboring countries. The same income rate can be found in the capital cities and major cities of Cambodia, the Lao People's Democratic Republic, and Myanmar in particular. However, one problem that has arisen is that there are insufficient job positions for job applicants, notably the unskilled labor group and semi-skilled labor group, which includes driver and foreman positions. Nonetheless, the situation in Thailand is somewhat the opposite because its economic development is on the upper middle income/ developing economy level, which enables Thailand to provide social welfare for education for its population across the country. Yet, a problem with educating its workforce is that they are not being provided with the specific skills needed by employers, which has triggered excess supply for skilled labor in the labor market. This prevents a certain percentage of students who have graduated from secondary education from finding jobs. Even if they get hired, they still tend to be underemployed, as the labor market requires more workers with a vocational background. Thailand's unskilled labor market has been facing a shortage of young workers since the country's population structure began transitioning into an aging society and many of its people of working age began going on to higher education and were disinterested in working as unskilled labor. This led to a disparity between population and geography since the countries that share land borders- - Cambodia, the Lao People's Democratic Republic, Thailand, and Vietnam- - experience differing problems regarding human resources. Moreover, within each country there are differences in human resources resulting from ethnic and minority groups not receiving equal opportunities to improve themselves, differences in improvement of human resources in different geographical regions, and differences in developing human resources for different types and sizes of businesses including properly registered businesses and those in the informal sector. Also, mainland ASEAN countries struggle with issues of human rights, human trafficking, child and spousal abuse, and neglect of the disabled and underprivileged. 1 0 . The triple disparities can be largely be addressed through ASEAN agreements such as the ASEAN Political- Security Community ( APSC) , which focuses on non- traditional security or ‘ human security’ , the ASEAN Socio- Cultural Community (ASCC), which was formed with the aim of developing human resources within ASEAN, and the ASEAN Economic Community ( AEC) , by promoting labor mobility in a more efficient manner. Marcus Tullius Cicero, noted scholar and statesman of the Roman Empire, held that immigration makes a country stronger. Labor mobility is one factor which

235 mitigates the disparity among the member countries of ASEAN. In Thailand, the problem of a lack of unskilled labor can be addressed by recruiting workers from neighboring countries which have excess labor, whereas CLMV countries that have a problem recruiting and retaining skilled labor can import and hire skilled workers who are unemployed in Thailand. Currently, AEC has agreements on labor mobility in two forms. The first form is the Agreement on Movement of Natural Persons ( MNP) , which was ratified in 2012, but has not been enforced in every member country. Skilled labor has been permitted to work in other ASEAN member countries without any work visa or work permit required. Another form is ASEAN Mutual Recognition Arrangements (MRA), which have established professional standards for eight professions, namely the MRA on Engineering Services ( 2005) , MRA on Nursing Services ( 2006) , MRA on Architectural Services ( 2007) , Framework Arrangement for the Mutual Recognition of Surveying Qualifications (2007), MRA on Medical Practitioners (2009), MRA on Dental Practitioners ( 2009) , MRA on Accountancy Services ( 2014) , and MRA on Tourism Professionals ( 2012) . Under these MRAs, workers in eight professions can request a professional license in various ASEAN member countries if they are fully qualified under the MRA. Other professions are being studied and negotiated as there is a pressing need for implementation of additional MRAs that will cover a variety of professions. While ASEAN has not yet agreed or negotiated on free trade of unskilled and semi- skilled labor mobility, Thailand, as a country that relies on labor most from its neighboring countries, has signed Memorandums of Understanding ( MOU) on Cooperation in the Employment of Workers bilaterally with Myanmar, Cambodia, and the Lao People's Democratic Republic. However, a problem arising at the present time is the one of fully enacting and implementing the variety of agreements, namely MNP, MRA, and MOU. Furthermore, many businesses in mainland ASEAN countries operate in the informal economy, causing a problem with data collection and storage. In addition, a lack of understanding of workers’ rights and strong sense of social justice, including among those charged with enforcing the laws, have resulted in ineffective labor mobility between mainland ASEAN countries, not allowing these countries to reach their full potential for labor mobility that each of the countries may have.