Скачать презентацию Estimation and Decomposition of Agricultural Productivity Growth in Скачать презентацию Estimation and Decomposition of Agricultural Productivity Growth in

104d56ade90b037b937d2e1f2fd28b3d.ppt

  • Количество слайдов: 31

Estimation and Decomposition of Agricultural Productivity Growth in Asia Supawat Rungsuriyawiboon Faculty of Economics Estimation and Decomposition of Agricultural Productivity Growth in Asia Supawat Rungsuriyawiboon Faculty of Economics Thammasat University

Introduction “When all people at all times have both physical and economic access to Introduction “When all people at all times have both physical and economic access to sufficient food to meet their dietary needs for a productive and healthy life ” (USAID) l Food crisis and food security are back on policy agendas Food Price

Introduction l Some food price examples from the FAO Type 2003 2007 2008 $/ton Introduction l Some food price examples from the FAO Type 2003 2007 2008 $/ton White Thailand rice (second grade) 198 323 854 Yellow corn 105 160 250 (+58%), (+36%) Wheat 144 207 401 (+64%), (+48%) 1, 835 3, 288 4, 750 521 714 Powdered milk Soy oil (+77%), (+62%) (+61%), (+30%) 1, 400 (+63%), (+49%)

Introduction l Food commodity price indices have increased across the board 52% Oil&Fat 48% Introduction l Food commodity price indices have increased across the board 52% Oil&Fat 48% Cereals 32% Dairy

Introduction l Numerous factors are influencing this price rise l Supply side: difficult seasonal Introduction l Numerous factors are influencing this price rise l Supply side: difficult seasonal conditions in the major production regions and increased input costs. l l Demand side: increasing food demand, rising demand for grain for biofuels Given the current world food situation, it is clear from the global perspectives that each world region must have a sufficient supply in agricultural products to meet the growing food demand

Asia has the potential to supply a substantial share of the expected growth in Asia has the potential to supply a substantial share of the expected growth in food demand Many countries undergone from CPE to a free market economy

Introduction Production of Wheat, Corn and Rice l Asia has experienced impressive growth in Introduction Production of Wheat, Corn and Rice l Asia has experienced impressive growth in rice and wheat production l The Green Revolution was achieved through the application of the high-yielding varieties of major cereals and irrigation system l Increased input use cannot guarantee a long-run sustainable growth rate of yields and output l Given the potential sources of factor inputs are being exhausted, future growth in agriculture will not only rely on mobilizing inputs but will also require rising productivity l Understanding the state of productivity improvements in Asia is important

Literature Review l A number of studies examine intercountry differences in productivity growth: - Literature Review l A number of studies examine intercountry differences in productivity growth: - The availability of new panel data sets - The development of frontier analysis l Two types of frontier analysis: - Stochastic Frontier Analysis (SFA): A parametric approach - Data Envelopment Analysis (DEA): A nonparametric approach l This frontier analysis allows to not only calculate productivity, but also decompose productivity growth l Both SFA and DEA models conducted in many studies to investigate intercountry differences in agricultural productivity growth in Asia using the panel data from the FAO

Literature Review l A nonparametric DEA model: - l A parametric SFA model: - Literature Review l A nonparametric DEA model: - l A parametric SFA model: - l Bureau, Färe, and Grosskopf (1995) Fulginiti and Perrin (1997) Arnade (1998) Suhariyanto and Thirtle (2001) Trueblood and Coggins (2003) Coelli and Rao (2005) Fulginiti and Perrin (1993) Craig, Pardey and Roseboom (1997) Wiebe et al (2000) Liu and Wang (2005) Because of data problems of transition countries in Central Asia, previous studies just ignored these countries

Objectives l First, this study formulates a general model using a parametric technique to Objectives l First, this study formulates a general model using a parametric technique to measure productivity growth This approach allows to uncover what sources attributing to productivity growth. l Second, this study measures productivity growth in Asian countries This study includes 27 countries for 25 years. The size of this sample allows us to examine productivity for almost all major nations in Asia over time.

Theoretical Framework l Performance of a firm A study about an ability of a Theoretical Framework l Performance of a firm A study about an ability of a firm to convert inputs into outputs given a technology in the production process l Performance measurement is a relative concept l A simple measure of performance is a productivity ratio l Productivity is defined as the ratio of outputs to inputs Productivity l = outputs inputs The greater value implies the better performance

Productivity Measurement l If a production technology consists of multiple inputs and outputs, a Productivity Measurement l If a production technology consists of multiple inputs and outputs, a measure of productivity becomes more complex l Productivity measured from the multi-input and multi-output production technology is called total factor productivity (TFP) l TFP can be measured using a concept of index number TFP index =output index input index

Other Method to Measure Firm’s Performance l Another method to measure the performance of Other Method to Measure Firm’s Performance l Another method to measure the performance of a firm is to use a concept of firm’s efficiency l In practice, the terms, productivity and efficiency have been used interchangeably. l However, they are not precisely the same things. l Efficiency of a firm is measured using a production frontier.

A Measure of Technical Efficiency l l l Consider a simple production process in A Measure of Technical Efficiency l l l Consider a simple production process in which a single input (x) is used to produce a single output (y) Line OF’ represents the maximum output attainable from each input level. The line OF’ is called a production frontier Consider three firms, that is A, B and C, are operating as follows • Firm A is operating beneath the frontier OF’ whereas firm B and C are operating on the frontier OF’ • Firm B and C are technically efficient • Firm A is technically inefficient • Technical efficiency (TE) can be measured by the distance. TE is equal to 0 A/0 B or 0 C/0 A

Distinction between Technical Efficiency and Productivity l From the figure, firm A is technically Distinction between Technical Efficiency and Productivity l From the figure, firm A is technically inefficient whereas firm B and C are technically efficient l Productivity of these firms are measured by the slope of the rays from origin l Firm C has higher productivity than firm A and B. Firm C has the highest productivity l Point C is the point of technically optimal scale. Operation at any other point on the production frontier results in lower productivity. Point C indicates an operation at scale economies

Distance function l Consider a production technology when multiple inputs are used to produce Distance function l Consider a production technology when multiple inputs are used to produce multiple outputs l Production frontier can not use to describe this production technology l Shephard (1953, 1970) proposes a distance function to describe the structure of production technology with multiple inputs and outputs l Two types of distance function. 1 Input distance function, DI. 2 Output distance function, Do

Output Distance Function (Do) l The minimum amount by which an output vector can Output Distance Function (Do) l The minimum amount by which an output vector can be deflated and still remain producible with a given input vector. l Output distance function Do(x, y) is defined as l where P(x) = {y: (y, x) Є T} Consider M = 2 This figure shows that the output vector y is producible with input x, but so is the radially expanded output vector (y/μ*) B A So, D 0(x, y) = μ * = OA/OB ≤ 1 D 0(x, y) = TE 0

Properties of Output Distance Function l )i) Do(x, 0) = 0 and Do (0, Properties of Output Distance Function l )i) Do(x, 0) = 0 and Do (0, y) = ∞ l )ii) Do (x, λy) = λDo (x, y) for λ > 0 (HOD+1 in y) l (iii) Do (λx, y) ≤ Do (x, y) for λ ≥ 1 (non-increasing in x) l (iv) Do (x, λy) ≤ Do (x, y) for 0 ≤ λ ≤ 1 (non-decreasing in y) l (v) Do (x, y) is convex function in y

Methodology l Total Factor Productivity (TFP) growth: Residual growth in outputs not explained by Methodology l Total Factor Productivity (TFP) growth: Residual growth in outputs not explained by growth in input uses l Färe et al. (1989) proposed a Malmquist TFP index to measure productivity growth using the output distance function l The output distance function at period t represents the minimum amount by which yt can be deflated and still remain producible with xt

Methodology l The Malmquist TFP index in period t l The Malmquist TFP growth Methodology l The Malmquist TFP index in period t l The Malmquist TFP growth index between t and t + 1 Period t Malmquist TFP growth (MTC) Period t+1 Technical Efficiency Change (TEC) Technical Change (TC) Scale Efficiency Change (SEC)

TFP growth decomposition TFP growth decomposition

Methodology l Orea (2002) employs a parametric technique to derive a generalized MPC decomposition. Methodology l Orea (2002) employs a parametric technique to derive a generalized MPC decomposition. l The output distance function taking the Translog functional form l Young’s theorem requires linear homogeneity in outputs

Methodology l The decomposition of MTC can be calculated as Technical Efficiency Change (TEC) Methodology l The decomposition of MTC can be calculated as Technical Efficiency Change (TEC) Technical Change (TC) Scale Efficiency Change (SEC)

Data l The empirical analysis in this study focuses on agricultural production of 27 Data l The empirical analysis in this study focuses on agricultural production of 27 Asian countries over the period from 1980 -2004 l The primary source of data is obtained from the website of the Food and Agricultural Organization (FAO) acquired from the AGROSTAT system l Production technology consists of two output variables and five input variables

Data Output Variables: l The output series are derived by aggregating detailed output quantity Data Output Variables: l The output series are derived by aggregating detailed output quantity data on 115 cropping commodities and 12 livestock commodities expressed in terms of the international average prices (in US dollars) Input Variables: l Land: Arable land in hectare includes both land under permanent crops as well as the area under permanent pasture l Tractor: the total number of wheeled- and crawler tractors used in agriculture l Labor: the number of economically active people in agriculture l Fertilizer: the commercial use of nitrogen, potassium and phosphate fertilizers in nutrient-equivalent terms expressed in thousands of metric tons l Livestock: the sheep-equivalent of the six categories of animals (buffaloes, cattle, pigs, sheep, goats and poultry)

Country Profile Region Central Asia (CA( East Asia (EA( West Asia (WA( Southeast Asia Country Profile Region Central Asia (CA( East Asia (EA( West Asia (WA( Southeast Asia (SEA( South Asia (SA( Country Kazakhstan (KAZ( Kyrgyzstan (KGZ( Tajikistan (TKM( Turkmenistan (TJK( Uzbekistan (UZB( China (CHN( Japan (JPN( Republic of Korea (PRK( Mongolia (MNG( Iraq (IRQ( Israel (ISR) Saudi Arabia (SAU) Syrian Arab Republi (SYR( Cambodia (KHM( Indonesia (IDN( Lao PDR (LAO( Malaysia (MYS) Myanmar (MMR) Philippines (PHL) Thailand (THA( Vietnam (VNM( Bangladesh (BGD( India (IND( Islamic Rep of Iran (IRN( Nepal (NPL( Pakistan (PAK( Sri Lanka (LKA(

Estimated Parameters of the Output Distance Model Parametera Estimates t-Statistic β 0 0. 277 Estimated Parameters of the Output Distance Model Parametera Estimates t-Statistic β 0 0. 277 8. 781** 0. 490 -0. 099 -0. 184 -0. 192 -0. 224 -0. 334 0. 331 20. 114** -7. 126** -15. 228** -8. 222** -16. 310** -11. 067** 5. 253** -0. 101 0. 033 0. 151 -0. 022 -0. 228 0. 043 -0. 103 0. 048 0. 035 0. 195 -0. 060 -0. 128 -0. 214 -0. 008 0. 296 -0. 051 0. 189 0. 114 -0. 008 -0. 001 -0. 008 -0. 006 -0. 001 -7. 517** 3. 321* 2. 455* -3. 161** -2. 034 5. 147** -4. 426 5. 470** 1. 179 8. 454** -7. 818** -4. 866** -10. 331** -0. 103 12. 564** -2. 115* 10. 061** 2. 067* -6. 887** -2. 590* -6. 996** -2. 564* -0. 410 βy 1 (crop) βx 1 (land) βx 2 (tractor) βx 3 (labor) βx 4 (fertilizer) βx 5 (livestock) βy 1 y 1 βx 1 x 1 βx 2 x 2 βx 3 x 3 βx 4 x 4 βx 5 x 5 βx 1 x 2 βx 1 x 3 βx 1 x 4 βx 1 x 5 βx 2 x 3 βx 2 x 4 βx 2 x 5 βx 3 x 4 βx 3 x 5 βx 4 x 5 βx 1 y 1 βx 4 y 1 βx 5 y 1 βt βtt βx 1 t βx 5 t βy 1 t

MTC and Decomposition for All Asian Countries Region TEC TC SEC MTC 1980 -1985 MTC and Decomposition for All Asian Countries Region TEC TC SEC MTC 1980 -1985 -0. 598 1. 422 -0. 481 0. 343 1985 -1990 0. 371 1. 897 -0. 494 1. 775 1990 -1995 -0. 218 2. 376 -0. 300 1. 857 1995 -2000 -0. 885 2. 847 0. 061 2. 023 2000 -2004 0. 835 3. 245 -0. 165 3. 916 1980 -2004 Asia Period -0. 138 2. 321 -0. 280 1. 902 l TFP growth across all of Asia was positive and nearly 2% l The high TFP growth has relied on TC. l The high TFP growth for Asia is largely driven by rises in TFP during the past 5 years. l TFP growth has been pulled down due to declining TEC and SEC. l This decline may be due to the continued rise in off-farm employment. l Asian TFP growth was relatively robust and rising. This is good news for those concerned about keeping balance in Asia and world food markets.

MTC and Decomposition for Each Region (in %) Region Period TEC TC SEC MTC MTC and Decomposition for Each Region (in %) Region Period TEC TC SEC MTC A) SA 1980 -2004 -0. 176 2. 456 -0. 064 2. 216 B) SEA 1980 -2004 0. 292 0. 825 -0. 050 1. 066 C) WA 1980 -2004 -0. 402 0. 081 -0. 056 -0. 376 D) EA 1980 -2004 -0. 218 2. 739 -0. 495 2. 026 E) CA 1992 -2004 -0. 087 1. 940 -0. 509 1. 344 l SA and EA exhibited high TFP growth. TC was a major factor driving TFP growth would have been higher had efficiency levels not fallen l TFP growth rate in SEA was only 1. 1%. Both TEC and TC contributed to TFP growth in SEA. l WA was the only region exhibiting TFP regress. However, average TFP growth is small. Both TEC and SEC dragged down TFP growth. l Without including transition countries in CA, Asian TFP growth would have been lower. TFP growth rate in CA reached 1. 4%.

MTC and Decomposition by Transition Countries (in %) Transition Country Periods TEC TC SEC MTC and Decomposition by Transition Countries (in %) Transition Country Periods TEC TC SEC MTC A) China 1980 -2004 -0. 250 3. 209 -0. 358 2. 600 B) Mongolia 1991 -2004 0. 078 3. 983 -0. 347 3. 714 C) Vietnam 1986 -2004 -0. 062 0. 052 -0. 734 -0. 744 D) Laos 1986 -2004 -1. 320 0. 542 0. 544 -0. 234 E) Myanmar 1989 -2004 0. 008 1. 704 0. 545 2. 256 F) Kazakhstan 1992 -2004 0. 225 3. 412 -1. 689 1. 948 G) Kyrgyzstan 1992 -2004 -0. 219 0. 587 -1. 020 -0. 653 H) Tajikistan 1992 -2004 0. 517 0. 232 0. 268 1. 018 I) Turkmenistan 1992 -2004 0. 069 1. 529 0. 687 2. 285 J) Uzbekistan 1992 -2004 -0. 950 1. 215 0. 122 0. 387

Conclusion l With nearly half of the potential agricultural resources, Asia has the potential Conclusion l With nearly half of the potential agricultural resources, Asia has the potential to supply an increase in world food demand l By including more member countries into the analysis especially the transition economies, Asian countries exhibited a healthy TFP growth with a growth rate of 1. 9 per annum. l Investments in R&D was a major contribution to TFP growth in Asian agriculture l The healthy TFP growth in Asian agriculture is greatly enhanced by countries in EA and SA. l Focusing on transition countries , large differences exist in terms of the magnitude and direction of agricultural TFP growth during the past two decades. l Some transition countries such as China, Mongolia and Turkmenistan exhibited above average growth. Others, such as, Kyrgyzstan, Uzbekistan, Laos, and Vietnam did not do so well