Study on Trade Restrictiveness of Agricultural Policies in Iran

Transcription

1 Internatonal Journal of Agrultural Sene and Researh Volume, Number 1, Wnter 011(Seral #) Study on Trade Restrtveness of Agrultural Poles n Iran G. rouz 1 *; R. Moghaddas 1 ; S. Yazdan 1: Department of Agrultural Eonoms, Sene and Researh Branh, Islam Azad Unversty, Tehran, Iran : Department of Agrultural Eonoms, Collage of Agrultural and Natural Resoures, Unversty of Tehran, Iran ABSTRACT Reeved: May, 11, 011 Aepted: July, 4, 011 In ths paper the trade restrtveness of agrultural poles n Iran was measured by the TRI ndex for the nne major agrultural produts nludng wheat, re, barely, maze, soybean, meat (beef meat), soybean ol, sugar and banana over the perod 1981 to 007. The man results ndate that there s no trend n above ndex and hgh volatlty espeally n 006 and 007 years due to onsderable nreases n tarff of sugar, re, meat and ol. Based on alulated advalorem equvalents (AVE s ) of non-tarff barrers (NTB s ), these barrers have a sgnfant ontrbuton to the level of trade restrtveness measured by the TRI ndex. Thus, negletng the restrtveness of NTB s dentfy an be very msleadng. Moreover, two sets of proteted produts: Frst, Over-proteted produts (re, ol, sugar and banana) whh ther atual tarffs are hgher than ther optmum ones. Seond, Under-proteted produts (wheat, barely, maze, soybean and beef meat) whh ther atual tarffs are lower than ther optmum ones. Due to ths, we found that most produts are heavly proteted and trade poles n agrulture setor have the prohbtve role on welfare and mport. Ths an lead to the derease n domest welfare level as well as ountry s trade volume wth tradng partners. Thus, t s neessary to adjust Iran s agrultural trade tarff regme n order to brdge the gap between the urrent struture of proteton and optmum mport dutes. Arhve of SID Keywords: Tarff; n-tarff Barrers; Import Demand Funton; Ad-valorem of NTB; TRI ndex. *Correspondng Author Emal: ghnorouz@yahoo.om -

2 50 Internatonal Journal of Agrultural Sene and Researh Volume, Number 1, Wnter 011(Seral #) INTRODUCTION Fndng the best way to measure the average level of tarffs and non-tarff measures whh restrt trade n an eonomy s an mportant subjet beause trade restrtons have a substantal nfluene on the alloaton of resoures and on the level and dstrbuton of soal welfare n a tarff-mposng ountry. Trade restrtons of ourse also affet mports from the ountry s tradng partners and, therefore, the welfare n these ountres (Loyd & Malaren, 008). Indators of trade restrtveness are used to study the effets of trade poly on growth, poverty, soal welfare and trade dstortons. They are also an essental nput to trade negotators, and to any study attemptng to understand the nsttutonal and poltal determnants of trade proteton (Grossman & Helpman, 1994). Most studes use suh measures as smple or trade-weghted average tarffs, non-tarff barrer overage ratos, pre wedge method and measures of tarff dsperson that are not well grounded n trade theory, and often, they only vaguely defne the aspets of the restrtveness of the trade regme (Anderson & Neary, 1998; Kee et al., 008). The easest way to measure a ountry s formal trade barrers s the mport-weghted average tarff rate, whh an be readly alulated by dvdng the revenue from mport dutes by the value of total mports (Irwn, 009). Unfortunately, ths measure has four rtal shortomngs that make t a poor ndator of the tarff s heght and stat welfare ost. Frst, the average tarff s downward based: goods that are subjet to hgh tarffs reeve a low weght n the ndex, and goods that are subjet to prohbtve tarffs wll not be represented at all. Seond, the average tarff understates the welfare ost of a gven tarff struture beause t gnores the dsperson n mport dutes aross goods. Thrd, the average tarff laks any eonom nterpretaton: an average tarff of 50 perent may or may not restrt trade more (or generate deadweght losses larger) than an average tarff of 5 perent. Fourth, the average tarff wll not reflet the mpat of non-tarff barrers, suh as mport quotas, n restrtng trade (Irwn, 009). Gven these problems, Anderson and Neary (1994, 1998, 005) provded several ndes of trade barrers that have a well-defned theoretal bass n terms of eonom welfare and the volume of trade. They usng omputable general equlbrum (CGE) models ntrodued ndes of trade restrtveness ndex (TRI) and merantlst trade restrtveness ndex (MTRI). The trade restrtveness ndex (TRI) refers to the unform tarff whh, f appled to all goods, would yeld the same welfare level as the exstng tarff struture (Anderson & Neary, 1994). The merantlst trade restrtveness ndex (MTRI) refers to the unform tarff that would yeld the same volume of mports as the exstng set of tarffs (Anderson & Neary, 005). The TRI ndex measures the dstortons mposed by eah ountry's trade poles on ts own welfare. It answers the followng queston: What s the unform tarff hat f appled to mports nstead of the urrent struture of proteton would leave home welfare at ts urrent level? The MTRI (n ths paper labeled Overall Trade Restrtveness Index (OTRI)) measures the mpat of eah ountry's trade poles on ts aggregate mports. It answers the followng queston: What s the unform tarff that f mposed on home mports nstead of the exstng struture of proteton would leave aggregate mports at ther urrent level? Feenstra (1995) developed a smplfed partal-equlbrum verson of the TRI that an be alulated wthout resortng to omplex general equlbrum smulatons. Kee, Nta, and Olarreaga (005, 008, 009), loyd and Malaren(008), Irwn(009) and Yu et al., (009) have used ths approah to evaluate the trade restrtveness. They fnd that the TRI and OTRI are hghly orrelated but that the TRI s hgher than the OTRI beause of the varane n tarff rates. Followng Kee, Nta and Olarreaga's studes (008, 009), n ths paper we measure Trade Restrtveness of Agrultural Poles n Iran usng the TRI ndex over the perod 1981 to 007. Ths man objetve an be dvded nto the followng spef objetves: Estmatng ad-valorem equvalents (AVE s ) of non tarff barrers (NTB s ) n the agrultural Setor of Iran. Computng trade restrtveness ndex (TRI) of traded agrultural produts of Iran Arhve of SID

3 Study on Trade Restrtveness of Agrultural Poles n Iran 51 Measurng the nfluene trade restrtveness of Agrultural Poles on welfare level MATERIALS AND METHODS Ths seton desrbes the methodology used to estmate ad-valorem of NTB s and trade restrtveness ndex too. Wth regard to trade restrtveness ndex summarze mpat of all trade barrers (nomnal tarff lne and non- tarff barrers) n unform tarff-equvalent thus we requres frst to brng all types of trade poly nstruments nto a ommon metr. Seton 1 omputes AVE s of NTB s of seleted agrultural produts. Estmatng AVE s of NTB s To obtan the AVE s of NTB s, we frst estmate the quantty-mpat of NTB on mports usng log-lnear model of mport demand wth assumng onstant return tehnologes smlar to Kee et al.,(005, 008). Then we onvert the quantty mpat of NTB s on mports nto a tarff equvalent based on mport demand elasttes. Log-lnear model used and adopted n ths study s the followng: ln m 0 1 lngdp 1 D ln NPR ln(1 t) Pw (1) Where m s the mport volume of good, GDP s real value added of agrultural setor, D s a dummy varable ndatng the presene of a ore non-tarff suh as tehnal barrers, mport quota system and mport lense, NPR s agrultural domest support (nomnal proteton rate), t s the nomnal tarff mposed on good n ountry and p w s domest equvalent of world pre of good. In equaton (1), 1 and are the oeffents that aptures the quantty mpat of NTB on mports and s mport demand elastty. Ths model allows for both tarff and NTB to deter trade wth effets that vary by mportng ountry and good. In ths paper, nomnal proteton rate (NPR) whh s ndator of agrultural domest support obtans the followng: NPR Pd P. E 100 Where p d s domest pre of good, p s CIF pre of traded good n ountry and E s formal exhange rate n ountry. To make NTB s omparable wth ad-valorem tarffs, one needs to transform the quantty mpat nto pre-equvalents. Ths s referred to as an AVE of NTB, and s defned as follows: Arhve of SID d log( P d ) AVE (3) dntb Aordng to equatons (1) and (3) and wth respet to non-tarff parameters of D and NPR : () d d log( m ) d log( m ) d log( P ) D d. AVE (4) dd d log( P ) dd d d log( m ) d log( m ) d log( P ) NPR d. AVE (5) d log( NPR ) d log( P ) d log( NPR )

4 5 Internatonal Journal of Agrultural Sene and Researh Volume, Number 1, Wnter 011(Seral #) Where AVE D and AVE NPR are the ad-valorem equvalents of ore NTB and domest support mposed on good respetvely. Solvng (4) and (5) for two NTB, we obtan tarff equvalent of non-tarff barrers as follows: D 1 ln m 1 D 1 (6) AVE.. 1 D NPR 1 ln m 1 NPR 1 (7) AVE.. NPR Overall AVE for eah good s obtaned by smply addng the two AVE of NTB omponents, and s denoted AVE : AVE AVE AVE (8) D NPR Estmatng trade restrtveness ndex The overall level of proteton mposed on mports of good s gven by: T t AVE (9) Where T s the overall level of proteton that ountry mposes on mports of good, t and AVE are nomnal tarff and ad-valorem non tarff mposed on good respetvely. In ths paper, based on the partal equlbrum model and Followng Feenstra's method (1995) and Kee, Nta and Ollareaga (008, 009), we alulate the measure of trade restrtveness (TRI) n a smlar method. The TRI summarzes the restrtons mposed by eah ountry's trade poles on ts own welfare. It answers the followng queston: What s the unform tarff that f appled to mports nstead of the urrent struture of proteton would fx welfare at ts urrent level? The TRI s (mpltly) defned by: TRI 0 : W ( TRI ) W ( T ) W n n (10) Arhve of SID Where W n, s the welfare assoated wth mports of good n n ountry and W 0 s the urrent level of aggregate welfare n ountry gven ts proteton struture. It s well known that n a partal equlbrum setup a seond-order lnear approxmaton to the welfare ost s gven by: 1 W m.. T (11) Thus, the extent of welfare loss nreases wth the elastty of mport demand, mports, and the squared of the level of proteton (overall tarff). Totally dfferentatng (10), usng (11) and solvng for TRI yelds:

5 Study on Trade Restrtveness of Agrultural Poles n Iran 53 k 1 m n, n, Tn, TRI (1) m k n, n, In ths paper, a smplfed and transformed TRI for ountry ' s agrultural setor and an be expressed as: TRI m T m 1 =1,, n (13) Where m s mport volume of produt, s mport demand elastty of produt and T s overall tarff mposed on produt. Equaton (13) shows that the partal equlbrum TRI s the weghted sum of squared proteton levels, where weghts are gven by the elastty of mport demand and mports (Kee et al., 008). Kee et al., (008, 009) statement that the squared of the partal equlbrum TRI an be deomposed nto the squared of the mport weghted average tarff, the tarff varane, and the ovarane between tarff squared and mport demand elasttes. Thus a hgher tarff varane and hgher levels of proteton on goods wth large mport demand elasttes lead to a hgher TRI. In ths paper, we study Trade restrtveness of agrultural poles n Iran for the nne major agrultural produt nludng wheat, re, barely, maze, soybean, meat (beef meat), soybean ol, sugar and banana. These produts own more 70 perent of the value of Iranan agrultural mports among 00 to 007. Data of ths paper are annual data for durng perods whh there are obtaned from dfferent soures. Tarff data s olleted from tarff shedule were presented n the annually mport and export rules whh publshed by mnstry of ommere. Exhange rate, Pre ndex and real GDP data are avalable n entral bank database of Iran. Domest pre of agrultural produts are ganed from agrulture mnstry of Iran. World pre and trade data of agrultural produts are obtaned from FAO's statstal database. RESULTS AND DISCUSSION In ths seton, we frst dsuss the estmates of AVE of NTB for eah seleted produt and then present the estmates of trade restrtveness ndex (TRI). Arhve of SID ad-valorm equvalents of non- tarff barrers( AVE s of NTB s ) In order to measurng tarff equvalent of non- tarff barrers, we frst estmate mport demand funton for eah good based on equaton (1). Eah of these regressons provded us the oeffents that measure the mpat of the two dfferent types of NTB s (dummy varable for ore non-tarff barrers and agrultural domest support) on mports. These are the oeffents n front of the two NTB varables ( 1, ) nterated wth a onstant oeffents and mport demand elastty ( ) that allow us to measure the mpat of world pre hanges on mports. The results of the estmatons of mport demand funtons for 9 agrultural produts are shown n Table (1). As the table shows, most varables n the regressons have the expeted sgns and are statstally sgnfant. The F-statst ndates the overall sgnfane of the sample regresson and t suggestng that the overall ft of the mport equatons s very good.

6 54 Internatonal Journal of Agrultural Sene and Researh Volume, Number 1, Wnter 011(Seral #) Table 1: The results of the estmated mport demand funtons for 9 agrultural produts varable ln GDP D lnnpr LnP d R D.W F-statst wheat (.0001) * (.0001) (.3) (.05) re (.0001) (.91) (.33) (.00) barely (.087) (.98) (.1) (.07) maze (.0001) (.0) (.90) (.0001) soybean (.0001) (.05) (.70) (.034) meat (.06) (.005) (.003) (.0001) ol (.1) (.53) (.83) (-.006) sugar (.8) (.70) (.088) (.0) banana 1.09 (.0001) -.59 (.311) -.36 (.06) -.1 (.008) * Fgures n Parenthess are sgnfane level Gven the fndngs of table (1) and followng equaton (6) and (7), the AVE s of NTB s for seleted produts estmated and the results are presented n table (). The results shows the average AVE s of NTB s for 9 group produts are: -3.43% for wheat,./61% for re, 1.74% for barely, 1.1% for maze, 3.13% for soybean, 1.67% for meat,./47% for ol,.18%for sugar and 7.9% for banana. Based on the results reported n table () we an also onlude AVEs of all produts are postve exept wheat. Table : AVE of NTB s for the seleted produts and trade support produt Own Pre AVE of Trade support Average of NPRelastty of NTB s (%) NPR (%) 100(%) demand wheat no re Yes=protetve barely Yes=protetve maze Yes=protetve soybean Yes=protetve beef meat Yes=protetve ol Yes=protetve sugar Yes=protetve banana Yes=protetve Table soure: alulated author Domest pre support Yes Yes Yes Yes Arhve of SID Trade restrtveness ndes Consderng the results of table () and based on equatons (13) and (16) we ompute trade restrtveness ndes (TRI and OTRI) for agrulture setor of Iran. In order to evaluate the effet of non-tarff barrers, the TRI ndex s estmated n two ases: frst wth onsder to both tarff and NTB (AVE of non-tarff) and seond, usng nomnal tarff data only. Table (3) provdes our estmates of TRI and OTRI ndes ompared to weght average of tarff for agrulture setor of Iran durng

7 Study on Trade Restrtveness of Agrultural Poles n Iran 55 Table 3: trade restrtveness ndex and weghted average of tarff year TRI (%) (wth NTB s ) TRI (%) ( wthout NTB s ) Weghted average of tarff (%) Average Table soure: alulated author As already been stated, the TRI s the unform tarff that f appled to mports nstead of the urrent struture of proteton would fx welfare at ts urrent level. In the other words, t s sowelfare measure. Therefore, we an reognze the effet of urrent tradng poles mposed on agrultural produts on the welfare level usng TRI ndex. Table (4) ndates the prohbtve mpat of trade tarff regme on Welfare n agrulture setor of Iran. As ths tables show, optmal tarff to keep welfare n aordane wth the TRI ndex s 14.9%. Arhve of SID Table 4: The nfluene of agrultural trade poles on welfare n Iran Produt Average of TRI ndex (%) Condton analyze hange n nomnal tarff (%) welfare wheat./ TRI>nomnal tarff re TRI<nomnal tarff - barely TRI>nomnal tarff maze TRI>nomnal tarff soybean TRI>nomnal tarff meat TRI>nomnal tarff ol TRI>nomnal tarff sugar TRI<nomnal tarff - banana TRI<nomnal tarff - Table soure: alulated author CONCLUSION In ths paper we measure Trade Restrtveness of Agrultural Poles n Iran usng the TRI ndex over the perod 1981 to 007. For ths purpose, we frst to brng all types of trade poly nstruments nto a ommon metr. On the other hands, we quantfy non-tarff restrtons.

8 56 Internatonal Journal of Agrultural Sene and Researh Volume, Number 1, Wnter 011(Seral #) The man results show the Tarff equvalents of all produts are postve exept wheat. Ths means non-tarff restrtons mposed on these produts have been protetve. Based on the results reported n table () we an also onlude AVEs of most produts are orrelated wth NPR oeffents (exept barely, maze and soybean). Ths reflets that when domest support of agrultural produts s done and are sgnfant, the trade poles also have a protetve effet. Consderng the results of AVE of NTB s of seleted produts and based on equaton (13) we ompute trade restrtveness ndex (TRI) for agrulture setor of Iran n two ases: frst wth onsder to both tarff and NTB s and seond, usng nomnal tarff data only. Aordng our estmates of TRI ompared to weght average we an make several mportant observatons as follows: In over ths perod, trade restrtveness ndexes no onsstent trend and s volatlty espeally n 006 and 007 years due to onsderable nreases n tarff of sugar, re, meat and ol. Ths represents the fat that the TRI depends almost entrely on the mean and the varane of tarff rates (Irwn, 009). n-tarff barrers (NTB s ) have a sgnfant ontrbuton to the level of trade restrtveness measured by the TRI ndex. Aordng table (3), the estmated TRI value wth respet to non- tarff barrers s hgher than the TRI value usng nomnal tarff data only n the all years. Thus, negletng the restrtveness of NTB an be very msleadng. Table (3) shows the annual devaton of the TRI from the weghted average tarff measure. In the other words, the average tarff an understate or unpredted the TRI by a sgnfant margn beause the mport-weghted average tarff does not nlude the varane of the tarff rates aross goods. Over ths perod, TRI exeeds the average tarff by about 79.5 perent, on average (14.9 % aganst 8.3%). Of ourse, TRI and average tarff ndex tend to move losely together when the number of ommodtes s small and when the dsperson of tarffs s low. Other alulatons have found devatons of smlar magntudes: Anderson and Neary (005) alulate that the TRI s about 50 perent hgher than the average tarff for the Unted States n 1990, and Kee, Nta, Olarreaga (008) found that the TRI s about 80 perent hgher than the mport-weghted average tarff, on average, aross many ountres. Ths suggests that n trade negotatons and n order to determne optmal tarff lnes, most ountres would prefer to use the TRI to evaluate ther own trade poles. Arhve of SID Based on the results of TRI ndex n table (4), we dentfy two sets of proteted produts: Frst, Over-proteted produts (re, sugar and banana) whh ther atual tarffs are hgher than ther optmum tarffs. Seond, Under-proteted produts whh ther atual tarffs are lower than ther optmum tarffs nludng wheat, barely, maze, soybean, ol and beef meat. Due to ths, we dsover that some produts are heavly proteted and ther trade poles have deterrent effet. Ths an lead to the derease of domest welfare level as well as ountry s trade volume wth tradng partners. Thus, t s neessary adjust Iran s agrultural trade tarff regme n order to brdge the gap between the urrent struture of proteton and optmum mport dutes.

9 Study on Trade Restrtveness of Agrultural Poles n Iran 57 REFERENCES 1. Anderson, J., Neary, P. (005). The Merantlst ndex of trade poly. Internatonal Eonom Revew, 44(), pp Anderson, J., Neary, P. (1998). A new approah to evaluatng trade poly. Revew of Eonom Studes, 63(1), pp Anderson, J., Neary, p. (1994). Measurng the restrtveness of trade poly. World Bank Eonom Revew, v.8, pp Feenstra, R. (1995). Estmatng the effets of trade poly. Handbook of Internatonal Eonoms., v.3, Amsterdam. Marh Food and Agrultural Organzaton.(008). Trade and pre statsts. Retreved from Grossman, G. M., Helpman, E.(1994). Proteton for sale. Ameran Eonom Revew. 84:4, pp Irwn, D. (009). Trade restrtveness and deadweght losses from U.S. tarffs, Natonal Bureau of Eonom Searh., Cambrdge Unversty. Januray Kee, H.L., Nta, A., Olarreaga, M. (009). Estmatng trade restrtveness ndes. Eonom Journal, Royal Eonom Soety. 119(534), pp Kee, H. L., Nta, A., Olarreaga, M. (005). Import demand elastty and trade dstortons. Revew of Eonoms and Statsts. Pp , Kee, H. L., Nta, A., Olarreaga, M. (008). Estmatng trade restrtveness ndes. World Bank. September Loyd, P., Malaren, D. (008). A trade restrtveness ndex of the level of proteton n Australan manufaturng. Unversty of Melbourne, Department of Eonoms., Workng Papers, number 104, pp.1-3, Deember Yu, E., Wales, E., Chavez, E. (009). A quanttatve analyss of trade poly responses to hgh agrulture ommodty pres. Agrultural Eonoms Conferene. Bejng, Chna, August 16-, 009. Arhve of SID