Asymmetric Price Transmission between Imported Wheat and Domestic
Asymmetric Price Transmission between Imported Wheat and Domestic Flour Price based on the Threshold Estimation of Price Equation Jung. Hoon Han
Contents Ⅰ. Introduction Ⅱ. Background Ⅲ. Estimation method Ⅳ. Result Ⅴ. Conclusion
Ⅰ. Introduction Previous studies 1. Classical model Tweeten & Quance (1969) Wolffram (1971) Houck (1977) 2. Threshold approach Goodwin & Holt (1999) Goodwin & piggott (2001) Robert J. Myers. , and T. S. Jayne. (2011)
Ⅰ. Introduction Basic hypothesis ⅰ) Is there a significant threshold point of imported wheat price that has different impact on domestic flour price? If there is, what is the level of estimated threshold point? ⅱ) How does asymmetric price behavior appear based on these threshold points? Critical points ⅰ) We use a threshold variable of input price, imported wheat price. ⅱ) We identify both demand supply shifters based on theoretical and statistical evidence ⅲ) We control the endogeneity problem by using instrument variable, Soybean oil price.
Ⅱ. Background Why wheat and flour price? ⅰ) Consumption amount is large enough. Food production record Foods Output Amount (Ton) Amount of money (1, 000₩) 1 Flour 1, 595, 694 1, 025, 722, 758 2 White sugar 1, 291, 383 720, 197, 539 3 Carbonated drink 1, 269, 254 1, 126, 338, 861 4 Mixed drink 735, 318 705, 668, 560 5 Fruits and vegetable drink 484, 726 472, 037, 547 6 Starch syrup 469, 748 267, 698, 229 7 Soybean oil 410, 557 462, 750, 205 8 Fruit sugar 397, 119 189, 898, 380 9 Grain processed food 375, 561 470, 338, 548 10 Noodles 349, 230 1, 085, 505, 021
Ⅱ. Background Why wheat and flour price? ⅱ) The rate of dependence on imports is high. The trend of the self-sufficiency ratios of major food crops Rice Wheat Bean Potato 2003 97. 4 0. 3 7. 3 98. 1 2004 96. 5 0. 4 7. 1 97. 1 2005 102. 0 0. 2 9. 7 98. 6 2006 98. 5 0. 2 13. 6 98. 5 2007 95. 8 0. 2 11. 2 98. 4 2008 94. 3 0. 4 8. 6 98. 3 2009 101. 1 0. 5 9. 9 98. 7 2010 104. 6 0. 9 10. 1 98. 7 2011 83. 2 1. 0 7. 9 96. 9 2012 86. 1 0. 7 10. 3 96. 2
Ⅱ. Background Why wheat and flour price? ⅲ) Oligopolistic market condition -There are 11 milling factories in South Korea. - Furthermore, only 70% of them are running in 2013. - high fixed costs and level of accumulated technical know-how.
Ⅱ. Background The reason why this study is meaningful Government spends large amount of cost to protect domestic industry. If we can define thresholds which have different impacts of imported wheat price to domestic flour price, protecting policies can be changed. More efficient price policy can be specified if government set up the different policies in each regimes based on the thresholds.
Ⅲ. Estimation method Data Summary Variable Explanation of variable Flour Consumer Price Index of domestic flour market (2010=100) Wheat Imported wheat price per Kg (Won) Elec Producer Price Index of electricity (2010=100) Wage index in processing industry (2010=100) Inter Price Index of intermediate goods in industry (2010=100) Ramen Consumer Price Index of Ramen (2010=100) Bread Consumer Price Index of Bread (2010=100) Meat Consumer Price Index of Meat (2010=100) Rice Consumer Price Index of Rice (2010=100) Income The growth rate of nominal income (%)
Ⅲ. Estimation method Price Equation - Demand function QID = f. ID(Pr, Prm, Pb) QDD = fdd(Pf, Inc, Pm, Prc) where Pr : flour price, Prm: ramen price, Pb : bread price where Inc : income, Pm : meat price, Prc : rice price QD = fd(Pf, Prm, Pb, Inc, Pm, Prc) - Supply function Qs = fs(Pf, Pw, W, Pe, Pi) where Pf : flour price, Pw : wheat price, W : level of wage, Pe: price of electricity, Pi : price of intermediate material - Optimal condition QD = Q S - Price equation Pf = f(Pw, Prm, Pb, Inc, Pm, Prc, W, Pe, Pi)
Ⅲ. Estimation method Threshold estimation = α 1 + Φ 1 Ptinput + β 1 DSt + γ 1 SSt + εt Ptoutput = α 2 + Φ 2 Ptinput + β 2 DSt + γ 2 SSt + εt = α 3 + Φ 3 Ptinput + β 3 DSt + γ 1 SSt + εt if Ptinput < C 1 if C 1 ≤ Ptinput < C 2 if Ptinput > C 2 - Price transmission between input price(imported wheat price) and output price(domestic flour price) - Compare the coefficients of wheat price in each regimes
Ⅲ. Estimation Method Empirical procedures 1. Divide samples into two groups. - Sample 1 : January 1993 ~ January 2008 - Sample 2 : January 1993 ~ March 2014 2. Estimate Price Equation - Based on industrial and statistical background : Redundant variable test - Control both demand supply shifters. - Try to overcome endogeneity problem : Soybean oil price 3. Threshold Estimation - Two significant threshold in both two sample groups. - Split each sample into three regimes 4. Compare price equations in each regimes - Impact of input price in output price is more powerful in higher level
Ⅳ. Result Estimation result of price equation using sample 1(1993~2008) Variables Constant Wheat Wage Inter Elec Ramen Model 1 Model 2 Model 3 -40. 9502*** -46. 4145*** (9. 3057) 0. 1384*** (0. 0146) -0. 0516 (0. 1064) 0. 7143*** (0. 2262) -0. 0987 (0. 2092) 0. 4833** (0. 1862) (7. 6497) 0. 1393*** (0. 0149) -52. 8999*** (7. 3157) 0. 1278*** (0. 0152) 0. 6009*** (0. 2186) 0. 8981*** (0. 1465) 0. 4635*** (0. 1509) 0. 2986*** (0. 0967) Soybean oil -0. 0537 (0. 1799) 0. 3096 (0. 2675) -0. 0908 (0. 2092) 0. 1301 (0. 1218) 3. 2431* (1. 9317) 0. 7480*** (0. 0675) 0. 2622 (0. 2653) 0. 3833 (0. 2655) 3. 2508** (1. 8652) 0. 8139*** (0. 0571) 0. 5817 (1. 7477) 0. 8326*** (0. 0546) R-squared 0. 982809 0. 982321 0. 982364 Adjusted R-squared 0. 981684 0. 981708 0. 981752 Akaike Info criterion 4. 529317 4. 501769 4. 499342 Bread Income Rice Meat Dummy 1 AR(1)
Ⅳ. Result Estimation result of price equation using sample 2 (1993~2014) Variables Constant Wheat Wage Inter Elec Ramen Model 1 -42. 8752*** (10. 2790) 0. 0301*** (0. 0008) -0. 1161 (0. 0819 0. 8170*** (0. 1620) 0. 1697 (0. 1085) 0. 1851 (0. 1862) Model 2 -43. 7698*** (8. 1177) 0. 0297*** (0. 0087) Model 3 -15. 1568 (17. 9340 0. 0267*** (0. 0085) 1. 0007*** (0. 1530) 0. 8174*** (0. 1751) 0. 3339*** (0. 1175) 0. 2340*** (0. 0659) Soybean oil 0. 1734 (0. 1530) 0. 1878 (0. 2307) R-squared 0. 0290 (0. 1590) 0. 1593 (0. 2367) -0. 0248 (0. 1192) 0. 0416 (0. 0889) 7. 2285*** (1. 8805) 19. 5963*** (1. 7639) 0. 8893*** (0. 0343) 0. 994308 6. 1939*** (1. 8645) 18. 7704*** (1. 7424) 0. 9148*** (0. 0307) 0. 994099 6. 1252*** (1. 8447) 17. 9899*** (1. 6880) 0. 9761*** (0. 0190) 0. 994375 Adjusted R-squared 0. 994025 0. 993931 0. 994214 Akaike Info criterion 4. 615523 4. 612224 4. 564407 Bread Income Rice Meat Dummy 1 Dummy 2 AR(1)
Ⅳ. Result Threshold Estimation Result of Sample 1 (1993~2008) Total Sample Regime A Regime B Null Hypothesis No threshold in sample 1 No threshold in regime A under the threshold estimate No threshold in regime B upper the threshold estimate Number of Bootstrap Replication 1000 Trimming Percentage 0. 01 Threshold Estimate 208. 8435 194. 6068 232. 9253 F-test for no threshold -181 -115 -69 Bootstrap P-value 0. 023 0. 119 0. 050
Ⅳ. Result Threshold Estimation Result of Sample 2 (1993~2014) Total sample Regime A Null Hypothesis No threshold in sample 2 No threshold in regime A under the threshold estimate Number of Bootstrap Replication 1000 Trimming Percentage 0. 01 Threshold Estimate 464. 0359 344. 2669 F-test for no threshold -255 -243 Bootstrap P-value 0. 001 0. 077
Ⅳ. Result Price equation of each regime using sample 1 (1993~2008) Regime 1 Regime 2 Regime 3 (Pw < 208. 84) (208. 84 < Pw < 232. 92) (Pw > 232. 92) Constant 31. 4372* (15. 9581) 65. 0170 (68. 9396) -10. 4986 (34. 2843) Wheat -0. 0008 (0. 0113) 0. 1351* (0. 0718) 0. 2078*** (0. 0175) Income 0. 0811 (0. 0990) -0. 3922 (0. 3717) -1. 6849 (1. 7928) Inter 0. 1396 (0. 1660) -0. 2227 (0. 4969) -0. 7355 (0. 7769) Soybean oil 0. 1156*** (0. 0545) 0. 0322 (0. 2439) 1. 2890*** (0. 3994) AR(1) 0. 9926*** (0. 0056) 0. 9805*** (0. 0397) -0. 0345 (0. 2644) R-squared 0. 997772 0. 973882 0. 955831 Adjusted R-squared 0. 997660 0. 966201 0. 942840 Akaike Info criterion 2. 116706 3. 688603 5. 449814 Variables
Ⅳ. Result Price equation of each regime using sample 2 (1993~2014) Regime 1 Regime 2 Regime 3 (Pw < 344. 26) (344. 26 < Pw < 464. 03) (Pw > 464. 03) Constant -45. 0704*** (4. 8917) 204. 6912*** (21. 7893) 61. 1925 (50. 9954) Wheat 0. 0616*** (0. 0115) 0. 0850*** (0. 0242) 0. 1852*** (0. 0527) Income 0. 1295 (0. 1789) -0. 5601 (0. 3389) 6. 1926 (1. 9709) Inter 1. 0668 (0. 1009) -1. 4695*** (0. 1843) 0. 7435 (0. 5031) Soybean oil 0. 1859*** (0. 0692) 0. 2303 (0. 1512) -1. 2489*** (0. 4051) AR(1) 0. 8479*** (0. 0406) 0. 4420*** (0. 0963) -0. 3613 (0. 4374) R-squared 0. 992777 0. 838917 0. 779654 Adjusted R-squared 0. 992578 0. 817722 0. 596032 Akaike Info criterion 3. 765121 5. 253446 7. 304188 Variables
Ⅳ. Result Poutput regime 1 regime 2 208. 84 regime 3 232. 92 sample 1(1993~2008) Pinput regime 1 regime 2 344. 26 regime 3 464. 03 sample 2(1993~2014) Pinput
Ⅴ. Conclusion 1. There are two significant thresholds in both sample groups. So, we can divide total sample into three regimes based on the thresholds 2. Price transmission effect between imported wheat and domestic flour is strong when the level of imported wheat price is getting higher. 3. There can be a possibility of asymmetric welfare distribution, market failure, existence of monopolistic intermediate merchant if the price transmission is asymmetric. 4. Price policies have to be different in each regime because the impacts of input price on output price are quite different based on the threshold points.
Ⅴ. Conclusion Limitation 1. Difficulty of interpretation. 2. Skewness of sample splitting 3. Reasonability of result Further study - Change threshold variable as the exchange rate
References Abdulai A. (2000). Spatial price transmission and asymmetry in the Ghanaian maize market. Journal of Development Economics, 63(2), pp. 327 -349. Barry K. Goodwin. , and Matthew T. Holt. (1999). Price Transmission and Asymmetric Adjustment In the U. S. Beef Sector. American Journal of Agricultural Economics, 81(3), pp. 630 -637. Bruce E. Hansen. , and Byeongseon Seo. (2002). Testing for two-regime threshold cointegration in vector errorcorrection models. Journal of Econometrics, 110(2), pp. 293 -318 Engle R. F. , and Granger C. W. J. (1987). Cointegration and error correction : representation, estimation and testing. Econometrica, 55, pp. 251 -276. ERNST R. Berndt. (1991). The practice of econometrics : Classic and contemporary. ADDISON-WESLEY. Giliola Frey. , and Matteo Manera. (2005). Econometric Models of Asymmetric Price Trasnmssion. Working paper. G. R. Griffith. , and N. E. Piggott. (1994). Asymmetry in beef, lamb and pork farm-retail price transmission in Australia. Agricultural Economics, 10(3), pp. 307 -316. Herry W. Kinnucna. , and Olan D. Forker. (1987). Asymmetry in Farm-Retail Price Transmission for Major Dairy Products. American Journal of Agricultural Economics, 69(2), pp. 285 -292. HAL R. Varian. (1992). Microeconomic Analysis. Third Edition. NORTON. James P. Houck. (1977). An approach to specifying and estimating nonreversible functions. American Journal of Agricultural Economics, 59(3), pp. 570 -572.
References Korean Agricultural Trade Information (KATI) Korean Statistical Information Service (KOSIS) Robert J. Myers. , and T. S. Jayne. (2011). Multiple-Regime Spatial Price Transmission With an Application To Maize Markets in Southern Africa. American Journal of Agricultural Economics, 94(1), pp. 174 -188. Ruey S. Tsay. (2005). Analysis of Financial Time Series. Second Edition. WILEY. Sam Petzman. (2000). Price rise faster than they fall. Journal of political economy, 108(3), pp. 466 -502. Seung-Churl Choi. , and Kyeong-Soo Jeong. (1999). A test of asymmetry in meat pricing. Journal of Korea Association Of Livestock Management, 15(1), pp. 65 -75. Sunoong Hwang. , and Moonsoo Park. (2013). Analysis of Asymmetric price transmission along the chicken and egg marketing channel. Agricultural Economic Research, 54(3), pp. 45 -70. Teahoon Kang. (2009). Empirical Analysis of Asymmetric Price Transmission of Agricultural Products. Agricultural Economics. 32(5), pp. 63 -81. Teahoon Kim. , and Baesung Kim. (2009). Impulse lag and asymmetric price transmission of cereal price. Agricultural Economics, 32(1), pp. 21 -40. Tweeten L. G. , and Quance L. (1969). Positive measures of aggregate elasticities : some new approaches. American Economic Review, 59, pp. 175 -183.
References Von Cramon-Taubadel S. , and Meyer J. (2004). Asymmetric price transmission : a survey. Journal of Agricultural Economics, 50, pp. 581 -611. Wolffram. R. (1971). Positive measures of aggregate elasticities : some new approaches-some ciritical notes. American Journal of Agricultural Economics, 53, pp. 356 -359. Yair Mundlak. , and Donald F. Larson. (1992). On the Transmission of World Agricultural World Bank Economic Review, 6(3), pp. 399 -422. Prices. Yong Duck Kwon. (2008). The Impact of Import Liberalization Policy on Transmission of Beef Price in Korean Journal of Agricultural Management and Policy. 35(1), pp. 72 -90. Yong-Kwang Shin. , Tae-Hun Kim. , Byeong-Il Ahn. , and Young-Gu Park. Test for asymmetric price transmission between imported and domestic products : application to red pepper, garlic and onion. Agricultural Economics, 51(2), pp. 1 -15.
Appendix Redundant variable test using sample 1 Redundant variables : Bread, Wage, Meat, Rice, Elec Value Df Probability F-statistic 0. 954350 (5, 168) 0. 4475 Likelihood ratio 5. 041330 5 0. 4109 Redundant variable test using sample 2 Redundant variables : Bread, Wage, Meat, Rice, Elec Value Df Probability F-statistic 1. 770350 (5, 241) 0. 1196 Likelihood ratio 9. 161987 5 0. 1028
- Slides: 25