EOOK OF ABS1RACTS
IITERNATI .i
Co-Orgonizer : Co-Hosl: Supporled By: ffi&,,,.] #O# ffi Sip'4, # ffi.,.,.it".u, 6lttF"iirt' tr*#lmandrn #,[f'n (D rnern@rv WETCOMING REMARKS ICAIOS (Internatipnal centre forAceh and Indian ocean studies) It is a distinct glqat pleasure and honour for everyone us at ICAIos to welcome all and to the 5tn L'te""ational conferenc" on 4""r, and India., o""u, studies, as rcAros, here at the rslamic ;:HT::viated state university (urN) Ar_Raniry,
The International centre for Aceh ancl Indian ocean studies (lcAlos) is joinr between three state universities a effort in Aceh (syiah Kuala university, uIN University of Malikussaleh), Ar-Raniry and the Government of Aceh, Indonesia and Ministry of Research Technology' and a number of international academic institutions and scholars. establishment was facilitated by It BRR Aceh-Nias in zooT,and since operation from zoog has started its Darussalam campus, here, in BandaAceh. since then, ICAIos has come to be known as inter-national and inter-university research training and research focusing on social and humanities centre, as well as inter-disciplinary studies. With all of your supports, we have now witnessed the young srh tCAtOS in the centre,s very sth y*"" anniversary since itu irr".,g.r.*r"* A- first was hosted by Nias and NUS singapore BRR Aceh- in Banda Aceh in zoo7, the ,".orrJ*";;;;d;;_Raniry NUS in 2oog, the third by syiah and l(uala university (uNSylAH) and at Frankfurt in zort, Goethe University the fourth by Malikussaleh university (UNIMAL) University in zor3. and Harvard
The fifth conference is schedulecl to be held in zeir5. However, roth since December zor4 is the anniversary of the zoo4 Indlan ocean Trrrr#i, it is consiclered that it is timely hold the confe.ence in zor4' This to 1'ear contbrence is co-organized by Raniry, and the centre, ulN fu.- The Earth Observatory of Singapore (EOS).
This year conference has more significance as H/e commemorate the ,oth y"", of,the eoo4 sumatra-Indian ocean-Boxing oay rslna-i"";;; n?f,"""' of Helsinki peace Mou. decade iong reconstruction in post-conflict A and post-disaster can now in term of be better exarnined changes it has made to our society in Aceh and beyond as well as assessing sustainability' Hence, the conference its theme is "confliet, Disaster, and Beyond; sustainability' and change, Interconnectedness in the Indian ocean the Regions,,. The focus of conference will be on how sciences, both natural and social sciences, contribute (or efforrs in seeking a berter and :;H::::Lx}]ri" safer life in posr-disaster and posr- we thank ulN Ar-Raniry, syiah I(uala universiry Mahkussaleh universiry, Government of and the Aceh for their continuing supports to the centre. our gratitude arso goes to Eos through Prof' I(erry sieh and Prof, Isaac I(erlow for their great contribution make this conference possible. to
ICAIOS lll we thank all keynote and invited speakers, special and contributed panel speakers and all other speakers foToTt"q to share your research findings, insights, and knowledge. We hope you have a froductive and insightful discussion iuring the conference ,,Charming and that you can make time to enjoy our BandaAceh,,.
Sincerelyyours, Saiful Mahdi, Ph.D. Executive Director/ICAIOS V Convener
IV TCAIOS A*ah *niversiti*s tr+w*n F*rnhina Internatlenal Board *s$.$$rti$*i {tsoarC oi Trustees} Prot$yarnsuJ Riu l. Headeti try Prof. R€dor, Syiatr Kssls Univwsity. Uniuersily +f *yiali K*atx Or. Flusaini lbrahim, $yiah Kusta Univesky ljniversity t{ Malikt}ss.{ieh F.,nlho*y ft*id Frof FBrid Wajdi lbrahln, Rector, tAtN AFRaniry ProfYusny tilamiq State ln$ilute Saby. tAtN Ar-Ranirv Regter al S)"'iah Kuals ll. AFr'dar. Rector, Uniwrsity of M;likus$ateh of Ar-Raniry Go$snor 8€cior si tAif,l Ar-Eani[d ZainiAbdutiah, A*h Govsnmenl Teuks Kama$ilaafl sn,Aceh Civil Socie|/ R6ct*r 0l l4alikusraiet iJ tndoflesiafl Miniqier of Res€€mh and Tffiimotogy Frof . h'lichilsi Feeqet Ptof lffian Abdullah, UniBsitt ot Gadiah M6da Rc€er Toi Bf. Pro{ (Emtritus}Anthony fl eid.'Austraiian Narional UnNersity Pfof. Michael F€ner. Aria Rssrch tfrstitute, NUS, Sinsapsre PrBf Henk Schult$Nordholi. KITLV, t-eiden Dr RogBr Tot, KtTlV, Jakarla 0r F€nci$cus Verell*n, EF€e, pari$ 0r Oaniet Penet, EFSO, Jakirta Prof Michael Leigh, Melboume UniveEity, AustrHlia lfs:. l{oji Yiyaakir Tokyo Unive.sity af Foreisft $rudieb, JaFan Prof. lsmail Ha&ki Goksoy, Suleyman Oemlret Unlvffiitv. T;rkev Pr6f VietorT. KrBg, Unrwrsity ot Le€ds, Grsrt Sdilain Managing Eoard Br Annabet Gallop, Briiish UbadrJ Frof BVrs GoOd. t'ts{vard Untvqsiiy, USA Eka *r. $rimriyani PrBf.lfury-Jo G€od. Haruard Univehity, US,S 1js6r$i$,lctul Aeiz Prof Su$anne Sshroebr, Godhe Univ&siu in FrenKturt, 6ffinsny 8r. Edward.&sFinatl A fias; $sl./a$ 0ah'earviyeh BBba, Univsrsiv Ksbangsaan Metay$iq Mdsysi8 $*!f ;rdd{* Farta*Y*r'"t ls&6ndarfuihqr'lain **rd&rh{r $upervis*ry Eaard Prsf. 0r. Ru*idiAli i.trhsftmEd - TABLE OF CONTENTS
Welcoming Remarks Fakhriati M. Thahir Icaios (lnternational Centre For Aceh Conflict Refl ection Between Ulama And And Indian Ocean Studies) ------l iii Umara In The rgth Century: A StudY On Sirajuddin Manuscript ------l 7 Sambutan Universitas Islam Negeri (UIN) HninAye Ko Ar-Raniry ------l v inter Religious Conflicts And The Im- poftance Of Reform Process For Sus- About ICAIOS ------l tainable Peace In Mvanmar Transition "ii Period ------l 8 Icaios Governing Body ------l ix Inggrid, Siana Halim, And Indriati Table Of Contents ------l x Nioto Bisono Assessing The Impact Of The Indian funirul Hadi Ocean Tsunami On The Economy: Evi- History, Memory And ldentity: The Of- dence From Indonesia And fice Of Wali Nanggroe In Aceh ------l r Thailand ------l q
Anthony Reid Juanda Djamal Europeans In Pre-r873 Aceh: Help Or Post-War Peacebuilding Through Im- Hindrance To Independence? ------l z plementing The Frameu'ork On New Aceh Scenario (NAS) ------l ro Asae SayakaAnd Salina l(halid Mangrove Rehabilitation Through Com- Ken Miichi munity Involvement In Post-Tsunami Plai{ul Relief:The Role Of Folk Per- Affected Area In Phang-Nga Province, forming Arts In japan After The zou Thailand ------l I Tsunami ------i rt
Bryan Rochelle Laina Hilma Sari, Patrick Daly, Capacity-Building And Resident Capa- Saiful Mahdi, Desi Safriana bilities In Post-Post-Tsunami Acehnese User Modifications And Aiterations Of Villages: Preliminary Findings Of The Post-Disaster Housing In Banda Aceh, Acarp Ii Village Survey ------l + Indonesia ------l u
Craig Thorburn Lely Safrina, Risana Rachmatan, Normalisasi: A Degrading Enabling Envi- Rizanna Rosemary, Maya l(hairani ronment For Community Empowerment Health Promotion To Stregthen The And Engagement In Development Plan- Mental Health Service ln Aceh (An Ac- ning And Implementation In Post-Post- tion Research) ------l r3 TsunamiAceh ------l 5 M. HasbiAmiruddin Eka Srimulyani Jihad Untuk Membangun "Trvo Steps Forward One Step Back"?: Peradaban ------l t+ The Dlnamics Of Rural Women's Agency In Post-Tsunami Aceh ------l 6 \ ICATOS Masaya Iga Rokhmat Hidayat, Teuku Cut Mahmud Evaluating The Econorrr,ic Recovery Of Aziz, ZaraYunizar Post-Tsunami Aceh province: Spatlal Re_ The Use Of Alternative EnergyAnd The structuring Of ShrimpAnd Fish Supply Improvement Of Economic Value Of Chains ------l rf Beef Cattle In Tingkeum Manyang Village, I(uta Blang Sub-District, Bireuenbistrict _ Misri A. Muchsin Aceh ------l z5 IAIN-UIN Ar-Raniry Dalam perspektif Historis ------l .6 Suhartono Intervention Analysis For Evaluating The Mokbul Morshed Ahmad Impact Of Disaster ------l z6 The Struggle Against Salinity Intrusion: Exploring The Alternative Coping Mech_ Syaifullah Muhammad, Saiful anisms ForThe Mahdi, CoastalcommrrrriW J., Patrick Daly And Cut Salfiana Bangladesh -----:l rZ The Effectiveness And Sustainabilitv Of Livelihood Asset Replacement post Mota, In Tsu_ L. And Sugianto S. nami BandaAceh ------l zt Tourism, TrainingAnd Tsunami ______l rg Abdul Manan MuharnedYunus Fenomena Duduk Ngangkang Naik Motor Contested Humanitarian Space And Bagi Wanita (Respon Masyarakat Tentang Inter-Communal Violence In Rakhine Efektivitas Implementasi Seruan Walikota State ------l r9 Lhokseumawe) ------| zg
Muhibbutthabry Afrizal Tioetra Reposisi Dan peran Revitalisasi Uin Peran Strategis OMS Dan pembangunan Ar-Raniry Dalam Implementasi Syari at Perdamaian Di Aceh ------l Islam Di Aceh ------l ,o "g Afrizal Tjoetra, Nurlian, Riki yulianda Nur Anisah And Rahmat Saleh Advokasi I(ebijakan Tentang penanggulan- The Intercultural Communication Be_ gan Kemiskinan Di Aceh (Studi Kasus tween The Expatriates And Acehnese Di I(ecamatan Samatiga Post-Tsunami In BandaAceh --_____l zr Aceh Barat) ------l t" Rahmat Saleh And Nur Anisah Al Chaidar Javanese Migrant Workers In Banda Aceh Terrorism In Aceh: A Social-Cultural Anal_ Post Tsunami (A Cross Culture Commu_ ysis OnAcehnese People On Refusing The nication Perspective) ------l zz Radical Movement Of Tandhim eaidltul Jihad ------l It Rizanna Rosemary, Febri Nurrahmi, Muallifah Anton Widyanto Media Coverage On Disasters-Related Conflicts And Negotiation In Educational Issues In Aceh's Local Newspaper (Case Policymaking: Examining process In Inter_ Study Of The Aceh's Earthquak e 2orz Institutions Policymaking To Implement Coverage In Serambi Indonesia) -___l "l Civic Education InAceh ----=--l i, Rizki AffiatAnd Maida Irawani Female Education Leaders For peace In .\ceh ------j z4 ICAIOS XI ffi Asrul Sidiq Ibnu Mundzir Rural Development Policy Impacts And Perceptions On Unintended Impacts Disparity In Rural Areas Of Aceh Province, Of Humanitarian Aid And Post Trau- Indonesia ------l ZZ matic Stress Disorders Among Con- flict And Tsunami-Affected Adults In Danil Alcbar Taqwadir?, Sulaiman, Aceh Besar ------l +r Taqwaddin Husin Bantuan Hukum Struktural Terhadap Kor- Inayatullah AbdoerrahMan ban Pelanggaran HakAsasi Manusia Oleh Anak Dan Bencana: Menguji Penera- Lbh Banda Aceh: tggs-zar1 -----l Z+ pan Pengurangan Resiko Bencana Di Sekolah Dasar Dan Madrasah lbtidai- Diah Pawestri MaharaniAn{t A. Nanda yah (Studi Analisis Di Kabupaten Aceh Saraswati Besar Dan Kota Banda Aceh) ----l +n Perlindungan Hak Ulayat Dan Hak-Hak I(eperdataan Masyarakat Adat Aceh Atas Intan Dewi Kumala, Kartika Sari, Tanoh Gampong/Tanoh Mukim Pasca Ben- Ibnu Mundir", And Haiyun Nisa cana Tsunami ------I lf Efektivitas Dan l(eberlanjutan Pro- gram Desa Siaga Sehat jiwa (DSSJ) - Ella Suzanna, Iskandar Zulkarnaen, Studi Kasus Pada Program DSSJ And Mardiana Mohamad Kabupaten Bireun ------l +l Understanding Acehnese Adolescents ro Years Post Disasters: Current Issues, Chal- Jamaluddin Abdullah Yasin, Mu- lenges And Coping ------l 16 hammad Nazaruddin, Nanda Arna- lia Chairuddin, And Laila M. Rasyid Eriko Kameyama "Cerai Liar" Dalam Masyarakat Aceh: ReflectingAid With LOnger Term Perspec- Studi Sosio-Legal Tentang Hu- tive: Case Study Of A Disaster Manage- kum PErceraian Dalam Perspektif ment Project In Post-Tsunami Aceh, Masyarakat, Ulama Dan Hakim Mah- Indonesia ------l lZ kamah Syar'iyah ------l ++
Euis Silvia Sundarti and Teuku Cut Kamaruddin And Renaldi Mahmud Aziz Safriansyah Disaster Mitigation Of Landslide In Suka- Analisis Pemberdal'aan Ekonomi damai Village Timang Gajah Sub-District Masyarakat Berbasis Masjid: Studi I(a- Bener Meriah District ------l :8 sus l(ota BandaAceh ------l +6
Fauzan Kusmawati Hatta Public Service Innovation Of 'Paterf At Media Dan Ilmu Komunikasi Untuk Sub District Levels In The District Of Aceh Pengurangan Resiko Bencana Yang Utara ------l lg Efektif Dan Berkelaniutan ------l +l
Ftraiyun Nisa, Zarina Akbar, Dian Yeron- Malahayati Rahman, Amrizal ika Sakti Kaloeti, And Evelin Witruk J. Prang ro Years Of Tsunami Disaster: Long-Term The Concept Of Legal Protection Of Psychological Consequences For Disaster Aceh Government Concerning Women Survivors InAceh, Indonesia ------l +o Workers In Mala,vsia In Post-Disaster And Post-Conflict ------l a8
xlt ICATOS MarzukiAbubakar Sabirin Dayah Post Tsunami:pari 'Modernis TradiSionalis Ke Pemb ------l erdayaan Masyarakat pascakonfl ik +q Berbasis Meunasah __-___l S+ Muhammad Ansor, Muhammad Bin Sehat Abu Bakar, yaserAmri Ihsan Shadiqin Orang Remaking Jawa Di Aceh: Memori Kedatangan, Aceh's Identity: Bendera Bintang Perkembangan Desa, Dan Sejarah ti" ti o.' Konflik ------l i""i"":l-*l,f$ *:';" ::illi 1^"'" Sg Yenny Rahmayati Mukhlisuddin llvas Aceh's Resurrection: post_Tsunami Potret Bekas Desa-produksi Ganja New DiAceh Housing And Its Cultural ------l S, Implications -----__l s6 Onanong Thippimol Yoshifumi Azuma The Political Role Of Acehnese UlamaAnd Forgotten Debates Datagzoo6_7 Iom Data On Over Islamic Law tn Aceh, ,9a5_ Ache t96z ------l Gam Combatants - Lessons From 5z Empirical Data ----___l 57 Rizki Amalia Af,fiat Short Biography The Trauma Is Ours: Of Reshaping Collective The MemoryThrough Inv-iteD Speakers ______l The Narratl"ls Of Wom_ Sg en Survivors In Aceh ______I S: Short Biography Of Icaios Executing Board ______l 6+
Programs, Date, And Venue ______l 63
ICAIOS xlll
Assessing the Impact of the Indian Ocean Tsunami on the Economy: Evidence from Indonesia and Thailand
Inggrid,†& Siana Halim,‡ and Indriati Njoto Bisono‡
Abstract— Recent research in developed countries shows posits that natural disasters lead to adverse and permanent an adverse effect of natural disasters on the economy. This effects on growth trajectories [4]-[5]. paper aims at examining whether this is also relevant for However, conducting cross-country studies to evaluate the developing countries. Applying a counterfactual approach to actual impact of natural hazards gives rise to two main provincial data for Indonesia and Thailand, we find that the problems. First, from growth theory, this means that they Indian Ocean tsunami of 2004 negatively affects per capita impose the strong assumption of parameter homogeneity [6]. gross domestic product (GDP) of the exposed provinces. It is Therefore, the effects of population growth, physical and also shown that the effect is heterogeneous within the country. human capital, as well as the initial level of income on income These results seem straightforward to reconcile with previous growth are the same for all countries in the analysis. In fact, evidence using developed countries data. this assumption is very strong and unrealistic. For instance, it
is very unlikely that different types of natural disasters Keywords—natural disaster, economic impact, developing country. produce similar effect on the economy. Second, country-level studies unable to capture the spatial distributional effect of the I. INTRODUCTION disaster. This paper seeks to fill the gap by investigating the causal SMALL but growing literature has been devoted to study effect of the tsunami catastrophic disaster in 2004 on the the economic consequences of disasters with the A regional economy of Indonesia and Thailand, the two most evolution of gross domestic product (GDP) as the central affected countries. It was 26 December 2004 at 00.59 GMT topic. The other common characteristic is the level of analysis (just before 08.00 a.m. Jakarta time), when a powerful focusing on cross-country studies. Intriguingly, existing earthquake with magnitude of 9.0 on the Richter scale hit empirical studies produce mixed-results. Following Sumatra Island of western Indonesia. The earthquake neoclassical growth frameworks, natural disasters are subsequently generated devastating tsunami waves, yielding predicted to have a positive effect on the GDP trajectory. In the tallest wave as high as 24.4 meters. The tsunami totally contrast, endogenous growth models provide less clear-cut slammed Aceh Province of Sumatra, the closest area to the explanation of disaster effects. A class of endogenous growth epicenter of the earthquake, whereas Nias Island of North models à la the Schumpeterian creative destruction process Sumatra Province was less affected. The successive tsunami reaches an agreement with the neoclassical theory. Several moved to the west to hit coastal areas of the other Asia earlier works seem to support favorable effects of natural countries (India, Malaysia, Maldives, Myanmar Srilanka, and disasters [1]-[3]. Yet, the AK-type endogenous growth models Thailand) and several African countries (Kenya, Somalia, and predict trivial impacts of disasters on the growth rate even Tanzania). In Thailand, the impacts of the tsunami were more though the economy that experiences a destruction of the pronounced in the southern part, especially Phuket, Krabi, capital stock will never go back to its pre-disaster growth path. Phang Nga, Trang, Ranong, and Satun [7]. Another variant of the endogenous growth theory with a Looking at the data, it was reported that Indonesia production function that exhibits increasing returns to scale experienced by far the highest number of fatalities than
Thailand (over 165,000 versus 8,300) representing about 70% * This research was made possible by funding from the Indonesian Directorate General of Higher Education (DIKTI) under SP-DIPA- of all deaths. Although these countries suffered from the 023.04.2.415015/2014. misery, the macroeconomic impact on Indonesia and Thailand † Department of Business Management, Faculty of Economics, Petra in 2005 was predicted to be small because Aceh’s GDP was Christian University, Surabaya, Indonesia. ‡ Department of Industrial Engineering, Faculty of Industrial Technology, approximately 4% of Indonesian GDP whereas the combined Petra Christian University, Surabaya, Indonesia. six provinces of Thailand accounted for only 2.7% of the & To whom correspondence should be addressed: Department of Business national GDP (The Economist Intelligence Unit, 2005). Yet, Management, Faculty of Economics, Petra Christian University, Surabaya, Indonesia (e-mail: [email protected]). preliminary findings reported that the tsunami had a sizeable impact on the regional economy of Aceh in Indonesia and
Phuket and Krabi in Thailand [8]-[9]. control group or the donor pool for the affected province. We use the synthetic control method (SCM) to estimate our Here, we let T0 =2004 be the year when the tsunami struck causal of interest [10]-[11]. SCM is an extension of the Indonesia and Thailand. We denote Y I as the regional GDP original difference-in-differences (DiD) but it is less stringent it N with respect to the identical trend assumption and it allows for per capita in the presence of the tsunami, while Yit is the the presence of unobservable time-variant provinces regional GDP per capita if the tsunami had not occurred. It is characteristics. The method is suitable in our case since the generally acceptable to assume that the disaster does not have tsunami is considered as a large shock influencing a single any effects on the outcome prior to its occurrence at time T0 . province. Hence, = for t∈[0,…, -1]. The economic effect of the This current work enriches fairly limited study available on T0 the economics of natural disasters in developing countries. tsunami for province i at time t is written as: I N The findings of our work also complement a recent study it Yit Yit (1) based on developed country data [12] and corroborate disaster We also have D , the binary variable that takes a value of theories about a non-linear relationship between a country’s it income per capita disaster shocks. one if province i is exposed to the tsunami at time t and zero This paper proceeds as follows. In Section 2, we give an otherwise. overview of estimating the distributional effect of the tsunami We can observe the post-tsunami outcome for province i at by utilizing SCM. Section 3 presents the main findings of the time t as: N paper. The last section concludes. Yit Yit it Dit (2) For each model, we assume that the only first province in II. SYNTHETIC CONTROL METHODS Indonesia and Thailand hit by the tsunami after . Therefore, We are interested in examining whether the Asian tsunami 1 𝑖푓 𝑖 = 1 푎푛푑 푡 > 푇 퐷 = { 0 has a substantial influence on the provincial GDP per capita of 푖푡 0 표푡ℎ푒푟푤𝑖푠푒
Indonesia (i.e. Aceh and North Sumatra) and Thailand (i.e. Our goal is to estimate it for the eight affected provinces (i Phuket, Krabi, Phang Nga, Trang, Ranong, and Satun). The =1) and for all t > , or: fundamental problem we have is to find an unexposed I N N (3) province that best reproduces the characteristics of those 1t Y1t Y1t Y1t Y1t I exposed provinces. Given that none of the other comparison The above equation implies that Y1t is observed in the provinces follow the identical time trends as the provinces of period 2005-2012, whereas Y N is unobserved. We need to interest; our strategy is to take a weighted average of all 1t N potential comparison provinces as a control group of the estimate Y1t which is the counterfactual of the exposed affected provinces. Therefore, the economic effect of the provinces or the synthetic control units. disaster is estimated through the difference in the regional It is shown in [12] that: GDP per capita between the two groups after the tsunami. This Y N Z (4) method is well-known as the synthetic control method (SCM). It t t i t i it where is an unobserved common time-dependent factor, We formalize the concept of the synthetic control method as t t follows. is a vector of unobserved parameters, Zi is a vector of Suppose that we observe n provinces (n =24 provinces for observed covariates for important ingredients for a growing 1 2 Indonesia and n =35 provinces for Thailand ) for the period GDP that is not affected by the tsunami, is unknown t =1995,…,2004,…,2012. Let i =1 be the exposed province, t common factors, is a province-specific unobservable, and and i =2,…, n be the other provinces that serve as the potential i are the error terms which represent unobserved transitory it 1 Since the introduction of the Regional Autonomy Law in 1999, the number shocks at the level of province ( E(it ) 0 for all i and t ). of provinces has been proliferating in Indonesia. Maluku and Papua have split For constructing the synthetic control unit, we define a (rx1) into two provinces since 1999. The new provinces are North Maluku and ' West Papua. A year later, the other three provinces were established, i.e. vector of weights W (w2 ,....wn ) such that wi 0 for i Bangka Belitung of South Sumatra, Banten of West Java, and Gorontalo of n North Sulawesi. Riau and South Sulawesi were separated to Kepulauan Riau =2,…, n and w 1. Each value of W indicates a i2 i in 2002 and West Sulawesi in 2004 respectively. The latest was North Kalimantan which was previously the part of East Kalimantan before 2012. potential synthetic control unit for each exposed provinces. Overall, there were 34 provinces in 2012. To maintain consistency, we We thus state the outcome for each synthetic control as: amalgamate these proliferated provinces with their original provinces and n n n n leave us with 26 provinces. However, we exclude DKI Jakarta and East (5) wiYit t t wi Zi t wi i wiit Kalimantan from the donor pool since these two provinces have extremely i2 i2 i2 i2 high per capita GDP among the other provinces. * * ' 2 Thailand has 76 provinces and is geographically divided to seven regions, We need to choose a set of weights (w2 ,...., wn ) that best i.e. Bangkok and Vicinities (6 provinces), Northern (17 provinces), North reproduces pre-tsunami characteristics of the exposed eastern (19 provinces), Southern (14 provinces), Eastern (8 provinces), Western (6 provinces), and Central (6 provinces). We only use the four last provinces such that: regions in the analysis due to their similar socioeconomic characteristics.
n n n synthetic counterfactual in 2005 and -27.02% on average w*Y Y ,..., w*Y Y and w*Z Z (6) i i1 11 i iT0 1T0 i i 1 during the period from the occurrence of the tsunami. Looking i2 i2 i2 It is proved that, as long as the condition in (6) holds and the at Phuket, per capita GDP is 21.95% lower in 2005 and 3.08% number of pre-tsunami observations is large as compared with lower on average. In general, the table also suggests that the the level of the transitory shocks [11], then economic effect of the tsunami is larger in Indonesia than n Thailand (reducing per capita GDP by 7.31% and 4.98% in Y N w*Y (7) 2005 respectively). 1t i it i2 To test the validity of our results, we perform a four Ultimately, the estimator for for t∈[ +1,…, ] is 1t T0 T different type of placebo exercises (i.e. placebo tests among given by untreated unit, placebo tests in time, treatment extremity test, n ˆ Y w*Y (8) and leave-one-out tests) to falsify several underlying 1t 1t 12 i it assumptions. These placebos should not respond uniformly to It should be noted that equation (2) can hold precisely under false interventions as the real treated unit does to the true the condition intervention if the causal effect is unquestionable. These (Y ,...,Y Z ' )Iˆ{(Y ,...,Y ,Z ' ),.....,(Y ,.....,Y ,Z ' )} 5 11 1T0 , 1 21 2T0 2 i11 i1T0 i1 falsification tests further strengthen our findings. However, in some cases, it is often possible to select the TABLE I SUMMARY OF THE TSUNAMI IMPACT IN INDONESIA AND THAILAND synthetic control W * to approximately satisfy condition (6). To assess the validity of our causal results, we conduct a 2005 Average series of placebo tests aimed at testing the underlying Gap % Gap % identification assumptions of our models. However, our Indonesia -816.21 -7.31 -1,245.45 -10.36 falsification tests must depend on permutation inference since Aceh -1,744.82 -16.24 -3,014.25 -27.02 the small samples used in SCM. North Sumatra 112.40 1.62 523.36 6.30 Thailand -9,285.87 -4.98 -1,534.04 -1.17 III. RESULTS AND DISCUSSION Phuket -49,445.71 -21.95 -6,757.41 -3.08 The essence of SCM is to construct a counterfactual unit or Krabi -8,863.03 -11.31 -446.27 -0.63 a synthetic control unit that closely replicates the pre-tsunami Phang Nga -732.29 -0.91 4,797.00 5.30 characteristics of the affected provinces. This is defined as a Trang 988.23 1.52 -4,773.15 -6.48 weighted average of unexposed provinces whose per capita Ranong 2,392.44 2.86 -921.82 -0.80 GDP is akin to the affected provinces if it had not been hit by Satun -54.84 -0.08 -1,102.58 -1.34 the tsunami. Figure 1 shows that the levels and trends of per Notes: Gap is the difference in per capita GRDP between the exposed capita GDP between the exposed province and the synthetic province and the synthetic control unit (in 1,000 Rupiah for Indonesia and in 3 Baht for Thailand). % is the ratio of Gap to per capita GRDP of the synthetic control unit in all eight cases are very similar. The values of control. Average is averaged over the post-tsunami period. per GDP ingredients of the exposed provinces before the onset of the tsunami do not diverge significantly to those of the 4 synthetic units. These findings suggest that the current IV. CONCLUSION exercises satisfy the identifying assumptions of SCM. The exposed and synthetic provinces are fairly comparable after We investigate the effects of the regional economic exposure to a catastrophic disaster in Indonesia and Thailand the tsunami period. What about the economic impacts of the tsunami? Figure 1 in the aftermath of the 2004 Indian Ocean tsunami. We find that Aceh, Phuket, Krabi, and Phang Nga experience a clearly shows that the tsunami has a negative effect on per nontrivial decline in their per capita GDP, whereas the capita GDP in Aceh, Phuket, Krabi, Phang Nga, and Satun, whereas it turns to be small and positive in North Sumatra, economy of North Sumatra, Trang, Ranong and Satun are less affected. Trang, and Ranong. However, it is should be noted that To the best of our knowledge, this is the first study applying between Aceh and Phuket, the two most affected provinces, a-quasi-experimental strategy and focusing exclusively on the evolution of per capita GDP is remarkably different. Aceh macroeconomic data from developing countries to identify the appears to experience a persistent decline in its GDP per capita while Phuket is able to recover from the catastrophic causal effects of a large natural disaster on the short- and the medium-term of income per capita. However, a major disaster and moves toward an upward trend. limitation of the macroeconomic framework as our current Table 1 presents summary statistics of the per capita GDP gaps between the affected provinces and the synthetic units. work is that it does not give a detailed explanation of the total welfare loss from the disaster. The study of microeconomic Given the level of Aceh’s actual GDP per capita, per capita GDP in this province seems to be 16.24% lower than in the data apparently helps to identify utility losses together with many other multifaceted dimensions (such as education, health, and poverty). This analysis is especially suitable for 3 We use a different length of the pre-tsunami period to minimize the root developing countries, like Indonesia and Thailand because the mean squared prediction error (RMSPE) for each case because per capita GDP of some province fluctuated in the late 1990s. 4 The predictor balance tests available upon request. 5 Results available upon request.
Fig. 1 Per capita regional GDP (in log): affected provinces and synthetic control units consequences of large disasters are more serious, but there is no adequate insurance coverage to protect households from [6] S. N. Durlauf, A. Kourtellos, and A. Minkin, “The local Solow growth model,” Europ. Econ. Rev., vol. 45, no. 4–6, pp. 928–940, May 2001. such extreme events. For this reason, an investigation of the [7] P.C. Athukorala, and B.P. Resosudarmo, “The Indian Ocean Tsunami: distributional impacts as well as insurance mechanisms against Economic impact, disaster management, and lessons,” Asian Econ. the economic costs of natural disasters deserves further Papers, vol. 4, no. 1, pp. 1-39, Winter 2005. [8] World Bank and GFDRR, Indonesia: preliminary damage and loss attention in the future research. assessment - The December 26, 2004 natural disaster. Washington, D.C.: The World Bank, 2005. REFERENCES [9] B. Nidhiprabha, “Adjustment and recovery in Thailand: Two years after the Tsunami,” unpublished. [1] C. Benson and E. Clay, “Understanding the economic and financial [10] A. Abadie, and J. Gardeazabal, “The economic costs of conflict: A case impact of natural disasters”. Washington D.C.: The World Bank, 2004. study of the Basque country,” Amer. Econ. Rev., vol. 93, no. 1, pp. [2] J.C. Cuaresma, J. Hlouskova, and M. Obersteiner, “Natural disasters as 113–132, Mar. 2003. creative destruction? Evidence from developing countries,” Econ. [11] A. Abadie, A. Diamond, and J. Hainmueller, “Synthetic control methods Inquiry, vol. 46, pp. 214-226, Apr. 2008. for comparative case studies: estimating the effect of California’s [3] S. Hallegatte, and P. Dumas, “Can natural disasters have positive tobacco control program”, J. Am. Stat. Assoc., vol. 105, no. 490, pp. consequences? Investigating the role of embodied technical change,” 493–505, June 2010. Ecological Econ., vol. 68, pp. 777-786, Jan. 2009. [12] W. duPont, and I. Noy, “What happened to Kobe? A reassessment of the [4] I. Noy, “The macroeconomic consequences of disasters,” J. Dev. Econ., impact of the 1995 earthquake in Japan,” to be published. vol. 88, pp. 221-231, Mar. 2009. [5] E. Strobl, “The economic growth impact of natural disasters in developing countries: Evidence from hurricane strikes in the Central American and Caribbean regions,” J. Devel. Econ., vol. 97, pp. 131-140, Jan, 2012.
APPENDIX A: DATA DESCRIPTION Inggrid is a lecturer and researcher in the Department of Business We describe the data used in the analysis and provide Management, Faculty of Economics, Petra Christian University, Indonesia. sources. The data are at the provincial level for the period She did her postgraduate degree in economics at Uppsala University, Sweden. 1995-2012. She is interested in program evaluation (with main focus on impact evaluation), poverty and inequality, human capital formation, and economic Indonesia: modeling. Per capita regional GDP (millions of Rupiah). Source: Central Bureau of Statistics (BPS). The data are obtained by dividing Siana Halim is a lecturer and researcher in the Department of Industrial the value of GDP in a particular province by its total Engineering, Faculty of Industrial Technology, Petra Christian University, Indonesia. She received her Ph.D. in statistics from Technische Universität population. Kaiserslautern, Germany, in 2005. Her research interests are statistical Sectoral shares (%). Source: Central Bureau of Statistics modeling, data analysis, and non-parametric statistics. (BPS). It consists of the value added of 9 economic sector, that is, agriculture, hunting, forestry, and fishing, mining and Indriati Njoto Bisono is a lecturer and researcher in the Department of Industrial Engineering, Faculty of Industrial Technology, Petra Christian quarrying, manufacturing, electricity, gas, and water, University, Indonesia. She has been doing her Ph.D. in applied statistics at the construction, trade, hotel, and restaurant, transportation and University of Melbourne, Australia. Her teaching and research interest include telecommunication, finance, real estate, and services. The Statistical Modeling and Design of Experiment. share of each sector is obtained by dividing the value added of each sector by the total provincial GDP. Population density (persons per square kilometer). Source: Central Bureau of Statistics (BPS). It is calculated as total population divided by land area in kilometre square. Human capital (%).Source: Central Bureau of Statistics (BPS). It includes educational attainment of the population (i.e. adult literacy rates, primary school, junior high school, senior high school, and university). Physical capital (%).Source: Central Bureau of Statistics (BPS). It is the share of fixed capital formation in the provincial GDP. Thailand: Per capita regional GDP (millions of Bath). Source: National Statistical Office of Thailand (NSO). The data are obtained by dividing the value of GDP in a particular province by its total population. Sectoral shares (%).Source: National Statistical Office of Thailand (NSO). It consists of the value added of 16 economic sector, that is, agriculture, mining and quarrying, manufacturing, electricity, gas and water supply, construction, wholesale and retail trade, hotels and restaurants, transport, storage and communications, financial intermediation, real estate, renting and business activities, public administration and defence; compulsory social security, education, health and social work, other community, social and personal service activities, and private households with employed persons Population density (persons per square kilometer). Source: Ministry of Interior. It is calculated as total population divided by land area in kilometre square. Human capital (%).Source: National Statistical Office of Thailand (NSO). It includes educational attainment of the population (i.e. preschool, primary school, junior high school, senior high school, and university). Credit to GDP ratio (%).Source: Bank of Thailand (BoT). It is the ratio of domestic credit provided by financial sector to provincial GDP.