External Liberalization, Poverty, and Inequality in Vietnam: The - - PowerPoint PPT Presentation

SLIDE 1

December, 2004

External Liberalization, Poverty, and Inequality in Vietnam: The Role of Price Dispersion

David Roland-Holst

UC Berkeley and Mills College

World Bank Conference on Putting Development Back into the Doha Agenda The Hague 2 December 2004

SLIDE 2

December, 2004 Roland-Holst

2

Contents

• 1. Introduction
• 2. Overview of Data and Methodology
• 3. Price Dispersion and Development
• 4. Preliminary Results
• 5. Conclusions and Extensions

SLIDE 3

December, 2004 Roland-Holst

3

Per Capita Income in East Asia

(quartiles)

SLIDE 4

December, 2004 Roland-Holst

4

Poverty Incidence by Province

(percent of population, quartiles)

National average 37%, NU average 59%.

SLIDE 5

December, 2004 Roland-Holst

5

Human Resources, Trade, and Income

Vietnam’s economic potential is far from being realized.

GDP

Exports

Population

Indonesia Malaysia Philippines Singapore Thailand Vietnam

SLIDE 6

December, 2004 Roland-Holst

6

Data Overview

Data are developed at three levels and integrated in the model to aggregate consistently.

National – a new SAM for Vietnam Regional – a regional SAM for the Northern

Uplands

Local - LSMS data

SLIDE 7

December, 2004 Roland-Holst

7

National Data: A New 2000 SAM for Vietnam

Incorporates the new 2000 GSO Input-Output Table 112 domestic production activities 114 commodities (includes trade and transport margins) 14 factors of production

12 labor categories Capital Land

16 household types, sampled from the VLSS

(Farmer, SelfEmp, Worker, UnEmp)x(Rural, Urban)xGender

3 enterprises (Private, Public, and Foreign) State (detailed fiscal instruments) Consolidated capital account 194 international trading partners

SLIDE 8

December, 2004 Roland-Holst

8

Primary Vietnam Household Surveys

Name Period of data collection Sample size Lowest level

• f

representation Types of income data collected 1993 Vietnam Living Standards Survey 1992-1993 4800 Seven regions Crop area, production, and sales; crop by-products; livestock income, fisheries income; livestock and fisheries expenses; forestry income; non-farm enterprises revenue and costs; wage income most important current wage jobs and most important wage jobs over past 12 months; remittances; land rent; social security; other income 1998 Vietnam Living Standards Survey 1997-1998 6000 Ten strata (7 regions and 3 types of urban areas) Almost identical content and structure as the 1993 VLSS. 2001-02 Vietnam Household Living Standards Survey 2001-2002 75,000 61 provinces Similar to VLSS, but less detailed, particularly the sections on wage income and enterprise income.

SLIDE 9

December, 2004 Roland-Holst

9

Notes on Methodology

To better understand these linkages, we apply economic analysis at three levels

1.

National – An economywide forecasting model to elucidate structural and price adjustments – this model generates national price, income, and sectoral adjustments

2.

Regional (Northern Mountain Region) – analogous to the national level, but with more localized incidence – this model maps national results to provinces

3.

Local – econometric estimation and micro- simulation or imputation of household responses/impacts

SLIDE 10

December, 2004 Roland-Holst

10

Overview of Methodology

National CGE Model Econometric Analysis Poverty Analysis Micro Simulation External liberalization will help the poor if its growth returns are transmitted to asset classes associated with them, mainly unskilled labor and land. To capture linkages across the economy and from the top down, a four-fold analytical framework is used.

SLIDE 11

December, 2004 Roland-Holst

11

Detailed Methodology

WTO Accession Doha, FTAs, External Shocks Regional and Local Market Data

Household Surveys

Prices Sectoral Adjustments Aggregate Incomes Occupational choice Production technology Consumer behavior Household Incomes, Expenditure, Output Factor use Poverty Indices Mapping Taxes/subsidies Investment

• Ag. Services, Credit,

Producer Support, Labor/land regulation

National CGE Model

Econometric Analysis Poverty Analysis

Micro Simulation

• Data
• Results
• Policy Intervention

SLIDE 12

December, 2004 Roland-Holst

12

Price Dispersion: Motivation

Geographic variation in prices is substantial in developing countries, and can play an essential role in inequality and poverty. A great deal of the development literature focuses on market failure. This is not synonymous with price failure. Prices embody information about economic structure and market functionality. The challenge is to extract this information and make use of it.

SLIDE 13

December, 2004 Roland-Holst

13

Price Dispersion: Causes

The underlying causes of price dispersion are many, but generically they can be divided into three categories:

• Geography – spatial and physical characteristics

that affect distribution margins

• Information – spatial asymmetries in market

relevant information induce heterogeneity in production and consumption decisions

• Institutions – price differences for both producers

and consumers are likely to be affected by their abilities to participate in formal and informal, public and private institutions that represent economic interests

SLIDE 14

December, 2004 Roland-Holst

14

Price Dispersion: Four Perspectives

• 1. Absolute Prices – resource costs,

comparative advantage, PPP

• 2. Relative Prices – production/consumption

patterns, incentives (subsistence)

• 3. Price Transmission – market segmentation

and margins, information content of prices

• 4. Price Volatility – risk and insurance

(subsistence)

SLIDE 15

December, 2004 Roland-Holst

15

Absolute Prices: Rural Price Indices by Province

Rice Other Crops Pigs Cattle Poultry Rice Coffee/ Tea Sugar Cane Other Crops Pigs Cattle Poultry

1 HaGiang 1.09 1.24 .52 .75 .40 1.07 .53 1.77 .73 1.00 .81 1.02 2 CaoBang 1.02 1.12 .71 .35 .54 1.15 .51 1.46 .91 .94 .70 .97 3 LaoCai 1.02 .90 .75 .74 .61 1.04 .55 .95 .63 .96 .52 1.03 4 BacCan 1.05 1.13 .71 .73 .63 1.04 .51 1.88 .77 .90 .56 .91 5 LangSon 1.03 1.08 .74 .41 .71 .98 NA 1.21 .97 .96 .52 1.09 6 TuyenQuang 1.04 1.01 .77 .60 .65 1.03 .51 .81 .49 .91 .69 .95 7 YenBai 1.04 1.01 .76 .78 .78 1.02 .55 .56 .64 .93 .45 1.03 8 ThaiNguyen 1.03 1.08 .81 .73 .63 1.04 .51 .72 .76 .94 .80 1.00 9 PhuTho 1.08 .97 .80 .84 .77 1.01 .51 1.02 .69 .92 .65 .99 10 BacGiang 1.09 .80 .78 .96 .88 1.11 .51 1.99 .90 .95 .70 .98 11 QuangNinh 1.03 .99 .90 .94 1.09 1.06 NA 1.66 .75 .99 .54 1.25 12 LaiChau 1.08 .99 .72 .90 .36 1.04 1.23 1.32 .57 1.02 .43 1.07 13 SonLa 1.00 .81 .86 1.23 .52 1.21 .96 1.26 .41 1.00 .45 1.03 14 HoaBinh 1.00 1.11 .79 .63 .60 1.04 .51 1.11 .33 .95 .82 1.02 Producer Prices by Commodity Consumer Prices by Commodity

Producer prices below national average: 75% of cases. Consumer prices below national average: 82% of cases.

SLIDE 16

December, 2004 Roland-Holst

16

Absolute Prices: Spatial Price Correlation (1998)

Source: Ligon and Roland-Holst

Price levels do not appear to be correlated nationally, i.e. local prices exhibit persistent and significant differences.

SLIDE 17

December, 2004 Roland-Holst

17

Relative Prices: Agricultural Terms of Trade

Margins affect relative prices for farmers in two ways:

• 1. Producer prices (numerator) decline with

market distance

• 2. Prices of agricultural inputs (denominator)

rise with distance In other words, rural balance sheets are leveraged from both directions.

SLIDE 18

December, 2004 Roland-Holst

18

Dac Lac Gia Lai Son La Lai Chau Nghe An Lao Cai Kon Tum Song Be Thanh Hoa Lam Dong Cao Bang Lang Son Ha Giang Yen Bai Minh Hai Ha Tinh Bac Thai Binh Thuan Ha Bac Quang Binh Dong Nai Vinh Phu Binh Dinh Quang Nam-Da Nang Long An Hoa Binh Kien Giang Phu Yen Quang Tri Tuyen Quang Quang Ninh Quang Ngai Can Tho Khanh Hoa An Giang Ninh Thuan Soc Trang Nam Ha Dong Thap Ha Tay Thua Thien-Hue Hai Hung Ben Tre Tra Vinh Ninh Thuan Tien Giang Thai Binh Ninh Binh Ba Ria - Vung Tau Kien Giang Dac Lac Gia Lai Son La Lai Chau Nghe An Lao Cai Kon Tum Song Be Thanh Hoa Lam Dong Cao Bang Lang Son Ha Giang Yen Bai Minh Hai Ha Tinh Bac Thai Binh Thuan Ha Bac Quang Binh Dong Nai Vinh Phu Binh Dinh Quang Nam-Da Nang Long An Hoa Binh Kien Giang Phu Yen Quang Tri Tuyen Quang Quang Ninh Quang Ngai Can Tho Khanh Hoa An Giang Ninh Thuan Soc Trang Nam Ha Dong Thap Ha Tay Thua Thien-Hue Hai Hung Ben Tre Tra Vinh Ninh Thuan Tien Giang Thai Binh Ninh Binh Ba Ria - Vung Tau Kien Giang Dac Lac Gia Lai Son La Lai Chau Nghe An Lao Cai Kon Tum Song Be Thanh Hoa Lam Dong Cao Bang Lang Son Ha Giang Yen Bai Minh Hai Ha Tinh Bac Thai Binh Thuan Ha Bac Quang Binh Dong Nai Vinh Phu Binh Dinh Quang Nam-Da Nang Long An Hoa Binh Kien Giang Phu Yen Quang Tri Tuyen Quang Quang Ninh Quang Ngai Can Tho Khanh Hoa An Giang Ninh Thuan Soc Trang Nam Ha Dong Thap Ha Tay Thua Thien-Hue Hai Hung Ben Tre Tra Vinh Ninh Thuan Tien Giang Thai Binh Ninh Binh Ba Ria - Vung Tau Kien Giang Dac Lac Gia Lai Son La Lai Chau Nghe An Lao Cai Kon Tum Song Be Thanh Hoa Lam Dong Cao Bang Lang Son Ha Giang Yen Bai Minh Hai Ha Tinh Bac Thai Binh Thuan Ha Bac Quang Binh Dong Nai Vinh Phu Binh Dinh Quang Nam-Da Nang Long An Hoa Binh Kien Giang Phu Yen Quang Tri Tuyen Quang Quang Ninh Quang Ngai Can Tho Khanh Hoa An Giang Ninh Thuan Soc Trang Nam Ha Dong Thap Ha Tay Thua Thien-Hue Hai Hung Ben Tre Tra Vinh Ninh Thuan Tien Giang Thai Binh Ninh Binh Ba Ria - Vung Tau Kien Giang

Poverty Specialization

• Mkt. Distance

Ag TOT

Poverty and Its Correlates

Unfavorable Favorable

Source: Author’s estimates

SLIDE 19

December, 2004 Roland-Holst

19

Mapping: Poverty and Market Access

Source: Epprecht:2004

SLIDE 20

December, 2004 Roland-Holst

20

Price Transmission: Spatial Correlation of Price Changes: 1993-98

Source: Ligon and Roland-Holst

Price adjustment exhibits some statistically significant correlation, but at very low levels.

SLIDE 21

December, 2004 Roland-Holst

21

National Model Simulations

Scenarios:

• 3. Trade Liberalization (UTL) from 2005,

Assuming China Implements its WTO

Commitments

• 7. UTL with Respect to Doha 1
• 9. UTL with Respect to Doha 2
• 11. UTL with respect to GTAP

SLIDE 22

December, 2004 Roland-Holst

22

Export and Import Composition by Partner

(2000 SAM Data)

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Vietnam Exports ASEAN Exports US Imports from ASEAN EU Imports from ASEAN Japanese Imports from ASEAN Agriculture Energy ProcFood TextApprl Manufact Services

SLIDE 23

December, 2004 Roland-Holst

23

Results: Unilateral Liberalization is Beneficial, Global Effects Seem Negligible

Scenario 2005 2010 2015 2020 1 UTL 3.01% 8.85% 19.20% 36.88% 2 UTL+Doha 1 3.01% 8.82% 19.16% 36.81% 3 UTL+Doha 2 3.01% 8.82% 19.17% 36.82% 4 UTL+GTL 3.01% 8.84% 19.22% 36.91%

SLIDE 24

December, 2004 Roland-Holst

24

Micro-simulation

Two primary channels for rural participation in urban growth:

• 1. Labor/Migration
• 2. Diversification to higher income elasticity

agricultural products The micro-simulation model extends current approaches by including multi-input, multi-

• utput CES/CET household production

SLIDE 25

December, 2004 Roland-Holst

25

The Role of Labor Income

The second primary source of rural HH income, labor earnings, has important direct and indirect linkages to the household enterprise:

• 1. Cash constraint – migrants are the most

important source of this

• 2. Migrant remittances are strongly correlated

with emergence from subsistence

SLIDE 26

December, 2004 Roland-Holst

26

Micro-simulation Results

All Households Urban Rural Before After % Change Before After % Change Before After % Change GE0

52.0 53.4

1.4

69.7

76.0 6.3

40.2

38.4

• 1.8

GE1

65.3 65.1

• 0.3

85.8

89.8 4.0

56.1

53.7

• 2.4

Gini

52.5 53.4

0.9

61.4

69.0 7.6

48.8

46.4

• 2.4

p0

62.8 53.4

• 9.4

53.5

40.0

• 13.5

68.4

60.9

• 7.5

p1

27.6 21.2

• 6.3

22.7

16.5

• 6.2

30.4

23.9

• 6.5

p2

15.3 11.7

• 3.6

13.4

9.4

• 4.0

16.5

13.0

• 3.5

Crossing the Poverty Line Freq. Percent Cumulative np/np

4,727,075 31.18 31.18

p/ p

8,840,273 58.31 89.49

np/p

154,928 1.02 90.51

p/ np

1,438,684 9.49 100.00 15,160,960 100.00

Poverty is reduced in both Urban and Rural areas. Inequality increases in Urban areas, but falls in Rural areas.

SLIDE 27

December, 2004 Roland-Holst

27

Calibrated Distribution Margins

Endogenous specification of margins is still in

• progress. For illustration, we calibrated a

simple margin w.r.t. market distance (d)

) (

) (

d

e d d m d m

β α + −

+ =

SLIDE 28

December, 2004 Roland-Holst

28

R a d i u s A l l H o u se h o l d s U r b a n R u r a l

1 0 0 0 k m

B e f o r e A f t e r % C h a n g e B e f o r e A f t e r % C h a n g e B e f o r e A f t e r % C h a n g e G E 0 5 2 . 0 0 5 3 . 4 3 1 . 4 3 6 9 . 7 0 7 6 . 0 2 6 . 3 1 4 0 . 2 0 3 8 . 3 8

• 1 . 8 2

G E 1 6 5 . 3 4 6 5 . 0 6

• 0 . 2 8

8 5 . 8 0 8 9 . 8 2 4 . 0 1 5 6 . 1 0 5 3 . 6 8

• 2 . 4 3

G i n i 5 2 . 4 7 5 3 . 3 7 0 . 9 0 6 1 . 4 0 6 9 . 0 0 7 . 6 0 4 8 . 8 0 4 6 . 3 6

• 2 . 4 4

p 0 6 2 . 8 0 5 3 . 4 1

• 9 . 3 9

5 3 . 5 0 4 0 . 0 0

• 1 3 . 5 1

6 8 . 4 0 6 0 . 9 3

• 7 . 4 7

p 1 2 7 . 5 6 2 1 . 2 4

• 6 . 3 1

2 2 . 7 0 1 6 . 5 3

• 6 . 1 7

3 0 . 4 0 2 3 . 9 2

• 6 . 4 8

p 2 1 5 . 3 1 1 1 . 7 1

• 3 . 6 0

1 3 . 4 0 9 . 4 1

• 4 . 0 0

1 6 . 5 0 1 3 . 0 4

• 3 . 4 6

3 0 0 k m

G E 0 4 7 . 1 0 4 8 . 3 1 1 . 2 1 6 9 . 7 0 7 5 . 8 8 6 . 1 8 4 0 . 2 0 3 8 . 8 5

• 1 . 3 5

G E 1 6 5 . 9 0 6 5 . 6 5

• 0 . 2 5

8 5 . 8 0 8 9 . 7 9 3 . 9 9 5 6 . 1 0 5 4 . 4 3

• 1 . 6 7

G i n i 5 2 . 4 0 5 3 . 1 4 0 . 7 4 6 1 . 4 0 6 8 . 6 5 7 . 2 5 4 8 . 8 0 4 7 . 1 3

• 1 . 6 7

p 0 6 6 . 1 0 5 8 . 4 7

• 7 . 6 3

5 3 . 5 0 4 0 . 0 7

• 1 3 . 4 3

6 8 . 4 0 6 3 . 3 6

• 5 . 0 4

p 1 2 9 . 3 0 2 3 . 7 2

• 5 . 5 8

2 2 . 7 0 1 6 . 6 5

• 6 . 0 5

3 0 . 4 0 2 6 . 1 7

• 4 . 2 3

p 2 1 6 . 0 0 1 2 . 8 0

• 3 . 2 0

1 3 . 4 0 9 . 6 2

• 3 . 7 8

1 6 . 5 0 1 3 . 9 1

• 2 . 5 9

2 0 0 k m

G E 0 4 7 . 1 0 4 7 . 9 8 0 . 8 8 6 9 . 7 0 7 4 . 7 0 5 . 0 0 4 0 . 2 0 3 9 . 2 9

• 0 . 9 1

G E 1 6 5 . 9 0 6 5 . 7 0

• 0 . 2 0

8 5 . 8 0 8 9 . 2 9 3 . 4 9 5 6 . 1 0 5 4 . 9 9

• 1 . 1 1

G i n i 5 2 . 4 0 5 2 . 9 7 0 . 5 7 6 1 . 4 0 6 7 . 8 9 6 . 4 9 4 8 . 8 0 4 7 . 6 4

• 1 . 1 6

p 0 6 6 . 1 0 6 0 . 5 5

• 5 . 5 5

5 3 . 5 0 4 2 . 2 1

• 1 1 . 2 9

6 8 . 4 0 6 4 . 6 4

• 3 . 7 6

p 1 2 9 . 3 0 2 4 . 7 9

• 4 . 5 1

2 2 . 7 0 1 7 . 3 4

• 5 . 3 6

3 0 . 4 0 2 7 . 5 5

• 2 . 8 5

p 2 1 6 . 0 0 1 3 . 4 1

• 2 . 5 9

1 3 . 4 0 1 0 . 2 0

• 3 . 2 0

1 6 . 5 0 1 4 . 6 9

• 1 . 8 1

1 0 0 k m

G E 0 4 7 . 1 0 4 7 . 6 1 0 . 5 1 6 9 . 7 0 7 3 . 6 4 3 . 9 4 4 0 . 2 0 3 9 . 8 8

• 0 . 3 2

G E 1 6 5 . 9 0 6 5 . 7 9

• 0 . 1 1

8 5 . 8 0 8 8 . 5 7 2 . 7 7 5 6 . 1 0 5 5 . 6 6

• 0 . 4 4

G i n i 5 2 . 4 0 5 2 . 6 9 0 . 2 9 6 1 . 4 0 6 6 . 5 1 5 . 1 1 4 8 . 8 0 4 8 . 3 7

• 0 . 4 3

p 0 6 6 . 1 0 6 3 . 2 9

• 2 . 8 1

5 3 . 5 0 4 5 . 1 5

• 8 . 3 5

6 8 . 4 0 6 6 . 9 0

• 1 . 5 0

p 1 2 9 . 3 0 2 6 . 8 9

• 2 . 4 1

2 2 . 7 0 1 8 . 9 5

• 3 . 7 5

3 0 . 4 0 2 9 . 3 4

• 1 . 0 6

p 2 1 6 . 0 0 1 4 . 5 0

• 1 . 5 0

1 3 . 4 0 1 1 . 0 5

• 2 . 3 5

1 6 . 5 0 1 5 . 7 4

• 0 . 7 6

Poverty Indices with Margins

SLIDE 29

December, 2004 Roland-Holst

29

Whose poverty was alleviated?

SLIDE 30

December, 2004 Roland-Holst

30

Extensions

• 1. National Level:

1.

Endogenous growth “X-factors”

2.

Financial flows

3.

Calibration and “envelope” problems

• 2. Micro-level:

1.

Better strategies for HH sampling

2.

Econometrics of HH production/consumption

3.

Endogenous price transmission

4.

Migration/remittances

SLIDE 31

December, 2004 Roland-Holst

31

Discussion

SLIDE 32

December, 2004 Roland-Holst

32

Preliminary Econometric Results

Because of its rich data resources, we want to use Vietnamese data to elucidate price dispersion as well as household consumption and production. Estimates thus far support two conclusions:

• Price levels do not appear to be correlated

nationally, but price shocks exhibit correlation with

• distance. Thus national market forces may be at

work, but local prices exhibit persistent and significant differences.

• Consumption growth has very low correlation

across the country. In other words, there is no nationally representative household, i.e. local characteristics matter very much.

SLIDE 33

December, 2004 Roland-Holst

33

Where Cit = consumption expenditure of household i in period t CIit = crop income index Hit = household size αj = dummy for region j

Consumption Regression

SLIDE 34

December, 2004 Roland-Holst

34

Consumption Regression

Source: Ligon and Roland-Holst