USDA Agri India

8/4/2019 USDA Agri India

1/43

United StatesDepartment oAgriculture

Economic

ResearchService

EconomicResearchReportNumber 89

December 2009

Growth and Equity Eects

o Agricultural Marketing

Efciency Gains in IndiaMaurice R. Landes

Mary E. Burfsher


2/43

www

.ers

.us

da.govVisit Our Website To Learn More!

National Agricultural Library

Cataloging Record:

The U.S. Department of Agriculture (USDA) prohibits discrimination in all its programs

and activities on the basis of race, color, national origin, age, disability, and, whereapplicable, sex, marital status, familial status, parental status, religion, sexualorientation, genetic information, political beliefs, reprisal, or because all or a part of an

individual's income is derived from any public assistance program. (Not all prohibited

bases apply to all programs.) Persons with disabilities who require alternative meansfor communication of program information (Braille, large print, audiotape, etc.) should

contact USDA's TARGET Center at (202) 720-2600 (voice and TDD).

To file a complaint of discrimination write to USDA, Director, Office of Civil Rights, 1400

Independence Avenue, S.W., Washington, D.C. 20250-9410 or call (800) 795-3272

(voice) or (202) 720-6382 (TDD). USDA is an equal opportunity provider and employer.

Landes, MauriceGrowth and equity effects of agricultural marketing efficiency gains

in India.(Economic research report (United States. Dept. of Agriculture.Economic Research Service) ; no. 89)

1. AgricultureIndiaEconometric models. 2. Agriculture-Economic aspectsIndia.3. Agriculture and stateIndia. 4. Farm produce-Marketing-Government policyIndia.

I. Burfisher, Mary E. II. United States. Dept. of Agriculture. Eco-

nomic Research Service. III. Title.HD9016.I42

Indiawww.ers.usda.gov/Briefing/India/

Photos: Maurice R. Landes


3/43

United States

Department

of Agriculture

www.ers.usda.gov

A Report from the Economic Research Service

Abstract

Agriculture is the largest source o employment in India, and ood accounts or abouthal o consumer expenditures. Moving agricultural products rom the arm to consumersmore eciently could result in large gains to producers, consumers, and Indias overalleconomy. This analysis uses a computable general equilibrium model with agricul-tural commodity detail and households disaggregated by rural, urban, and income class

to study the potential impacts o reorms that achieve eciency gains in agriculturalmarketing and reduce agricultural input subsidies and import taris. More ecient agri-cultural marketing generates economywide gains in output and wages, raises agricul-tural producer prices, reduces consumer ood prices, and increases private consumption,particularly by low-income households. These gains could help to oset some o themedium-term adjustment costs or some commodity markets and households associatedwith reducing agricultural subsidies and taris.

Keywords: India, agriculture, policy reorm, marketing eciency, taris, subsidies,households, computable general equilibrium model.

AcknowledgmentsRoman Keeney (Purdue University), Ashok Gulati (International Food Policy ResearchInstitute), Holly Higgins (USDA, Foreign Agricultural Service), and two anonymousreviewers provided valuable comments on earlier versions o this report. The authorsthank Anita Regmi (USDA, Economic Research Service (ERS)) or coordinating the peerreview o the report, Linda Hatcher (USDA, ERS) or editing the report, and Cynthia A.Ray (USDA, ERS) or graphics design and layout.

Maurice R. Landes

Mary E. Burfsher

Growth and Equity Eects

o Agricultural Marketing

Efciency Gains in India

Economic

Research

Report

Number 89

December 2009


4/43

iiGrowth and Equity Eects o Agricultural Marketing Efciency Gains in India / ERR-89

Economic Research Service/USDA

Contents

Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Agricultural Policy, Investment, and Marketing Efciency in India . . 3

Potential Implications o Marketing Efciency Gains . . . . . . . . . . . . . . 6

Estimated Costs o Marketing Agricultural and Food Products. . . . . . 8

Household Income and Expenditure Patterns . . . . . . . . . . . . . . . . . . . 11

Impacts o Potential Marketing Efciency Gains . . . . . . . . . . . . . . . . . 15

Economywide Impacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

Agricultural Sector Impacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

Household Impacts: Income and Rural/Urban Distribution . . . . . . . . . 18

Comparing the Impacts o Increased Marketing Efciency

With Agricultural Input Subsidy and Tari Reorm. . . . . . . . . . . . . . 19

Indias Agricultural Subsidies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

Indias Agricultural Taris . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

Scenario Comparisons. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

Reerences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

Appendix 1: Model and Database . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

Appendix 2: Sector Aggregation From GTAP Database;

Mapping GTAP Data Into the India SAM . . . . . . . . . . . . . . . . . . . . . . 33

Appendix 3: Scenario Results or Impacts o Agricultural Marketing

Efciency, Input Subsidy, and Tari Reorm Scenarios in India . . . . . 34

Recommended citation ormat or this publication:

Landes, Maurice R., and Mary E. Burfsher. Growth and Equity Eects o

Agricultural Marketing Efciency Gains in India . ERR-89. U.S. Dept. o

Agriculture, Econ. Res. Serv. November 2009.


5/43

iii

Growth and Equity Eects o Agricultural Marketing Efciency Gains in India / ERR-89Economic Research Service/USDA

Summary

Growth in Indias economy and consumer buying power has acceleratedsharply since the early 1990s, when a balance-o-payments crisis instigatedmajor liberalizing reorms to exchange rate, trade, and domestic regulatorypolicies. As a result, Indias ood demand is expanding and diversiying,but Indias arm sector has not shared in the benets o policy reorm, andgrowth in the sector has remained sluggish. Because the arm sector accountsor a large share o total output and employment in the Indian economy, thispoor perormance raises concerns about Indias ability to sustain acceleratedincome growth, reduce pervasive rural poverty, and maintain ood security.

What Is the Issue?

Despite the recent robust expansion o Indias economy, lagging investmentand growth in its agricultural sector are raising concerns about the need or asecond round o reorms to stimulate the arm economy. Reorms o agricul-tural trade protection and producer subsidies are requent topics o discussionand analysis, but strong evidence supports the idea that ragmented and ine-

cient domestic agricultural marketing chains seriously hinder agriculturalcompetitiveness and growth. Measures to boost marketing eciency byreducing regulatory barriers that have impeded investment in agriculturalwholesale and retail trade services may also improve conditions or low-income producers and consumers, a priority or Indias policymakers. Thisstudy examines the perormance o Indias agricultural marketing system andanalyzes the economywide implications o improved marketing eciencythat might stem rom uture reorms to domestic market regulations andincreased investment in agricultural markets.

What Are the Major Findings?

Measures to improve agricultural marketing eciency in India can substan-tially and broadly benet Indias economy. Improved marketing eciencyhas the potential to generate economywide gains in output and wages, raiseagricultural producer prices, reduce consumer ood prices, and increaseprivate consumption, particularly by rural and low-income households.

The broad gains rom improving agricultural marketing eciency contrastsignicantly with the impacts o reducing agricultural subsidies and taris.Indian policymakers ace domestic and international pressures to reduceinput subsidies and taris in the arm sector. But reducing subsidies andtaris, while conerring economywide benets, may also, at least in themedium term, create adjustment costs or labor, land, and capital markets,

some commodity sectors, and households. In contrast, improved agricul-tural marketing eciency can benet the overall economy (see chart), aswell as low-income households. The results o this study suggest that policymeasures to improve the eciency o domestic agricultural markets may be avaluable complement to subsidy and tari reorms by helping to mitigate themedium-term losses that may stem rom such reorms.


6/43

ivGrowth and Equity Eects o Agricultural Marketing Efciency Gains in India / ERR-89


Greater investment and eciency in Indias agricultural supply chains also

have the potential to enhance agricultural growth over the longer term.Whether new policies lead to rapid investment by modern retailers andothers in transorming Indias markets or the process occurs gradually, moreecient agricultural marketing is likely to strengthen consumer demand orood and other goods. More ecient domestic marketing may also boost netagricultural exports, although this result does not account or the changes indemand likely to occur as higher income growth is sustained over the longerterm or or constraints on crop and livestock production that may emerge.

This analysis also does not ully assess the adjustment costs, including poten-tial employment and income losses in some areas o the traditional marketingsystem, that might result rom transorming Indias traditional wholesale and

retail markets into a more ecient sector. This transormation could lead toewer, but larger, vertically integrated, and more technologically advancedprocessing and marketing enterprises. Concerns with managing these adjust-ment costs are central to the current debate over regulating direct oreigninvestment in ood retailing in India. This analysis suggests, however, thatalthough some participants in the traditional marketing system would undoubt-edly ace adjustment costs, the impacts on economywide wages and welare, aswell as the welare impacts on low-income households, are positive.

For the United States, these results suggest that increasing investment andeciency in Indias agricultural markets is an important potential drivero broad-based income and demand growth in India, likely bringing long-

term benets to its trade partners. The analysis o the potential impacts oreorms to agricultural input subsidies and taris suggests why these issuesare sensitive or India in bilateral and multilateral negotiations. At least inthe medium term, reducing input subsidies lowers output o ood staples,leading to reduced welare or low-income households, while reducingtaris imposes adjustment costs on protected commodity sectors. But, theresults also suggest that measures to improve agricultural marketing e-ciency, perhaps including cooperation to strengthen Indias market institu-tions and investment climate, can help mitigate the adjustment costs aced

Economywide effects of alternative agricultural and food marketing

policy scenarios in India

-6.0 -4.0 -2.0 0.0 2.0 4.0 6.0

Percent change from base

50% increase in marketingefficiency

Eliminate input subsidies

Eliminate agricultural tariffs

Capital rents

Wages

Land rents

Producer price index

Real investment demand

Real household consumption

Real gross domestic product


7/43

v


by some households and commodity sectors rom such reorms. Further, thesubstantial increases in agricultural output and ood consumption that ariserom improved marketing eciency in this analysis suggest the potential orpositive returns to private investmentincluding U.S. private oreign invest-mentin Indian agribusiness.

How Was the Study Conducted?

The authors constructed a computable general equilibrium (CGE) modelo the Indian economy. The basic model structure was developed by theInternational Food Policy Research Institute (IFPRI), with expandedcommodity coverage, disaggregation o marketing and trade costs, disaggre-gation o rural and urban households by expenditure class, and other exten-sions added by the authors or this study. Model data were supplied by theGlobal Trade Analysis Project (GTAP) and ocial Indian sources. Modelstructure, extensions, and data sources are described in detail in an appendix.The study also draws on data and results rom studies o Indias agriculturalmarkets conducted recently by ERS, the World Bank, and other institutions.


8/43


9/43

1


Introduction

Growth in Indias real income has accelerated sharply since the early 1990s,when a balance o payments crisis instigated major liberalizing reormsto exchange rate, trade, and domestic regulatory policies. But, despite theresulting expansion and diversication o ood demand, growth in Indiasarm sector has been slowing. Because the arm sector accounts or largeshares o total output and employment in the Indian economy, this poorperormance raises concerns about progress in reducing pervasive ruralpoverty, the sustainability o Indias higher growth path, and the maintenanceo ood security. The successul reorms o the early 1990s enhanced thecompetitiveness and growth o Indias manuacturing and service sectors,but political consensus on a second round o reorms to stimulate the ruralsector and more directly address the needs o Indias rural and urban poorhas remained elusive. A major package o credit concessions and outlays oragriculture announced in the 2008/09 government budgetin advance onational elections in early 2009was evidence o continued concern witharm sector perormance.

The government has been and continues to be heavily involved in Indiasagricultural sector through policy interventions in the production, marketing,and trade o most major arm commodities. Most obviousand moststudiedhave been Indias border measures, including relatively high boundand applied taris and export controls or arm products, and its burgeoningsubsidies on arm inputs and on producer and consumer prices o wheatand rice.1 Less obviousand less studiedare Indias extensive array ocentral and state regulations that aect the movement, storage, processing,and marketing o arm commodities and that have served as disincentives orprivate investment in agricultural wholesale and retail marketing. As a result,Indias agricultural marketing system consists primarily o small-scale,nonintegrated, and inecient marketing enterprises that tax both producers

and consumers o arm products (Landes, 2008; Landes and Gulati, 2004).

A small but growing body o research on Indias agricultural wholesale andretail markets has identied a range o developments that are increasing orcould increase eciency in the supply chains that move agricultural prod-ucts rom producers to consumers. Vertical integration,2 led by the privatesector, has signicantly reduced marketing costs, reduced consumer prices,increased grower returns, and stimulated growth in Indias broiler industry(Landes et al., 2004). Reducing the governments role in marketing wheatby ully decentralizing and privatizing procurement or government distri-bution programs could reduce producer wholesale markups or wheat,as well as sharply lower budgetary costs (Jha et al., 2007). Deregulating

oilseed marketing and processing, which would allow the import ooilseeds, could improve eciency in Indias oilseed-processing industryand reduce consumer prices o edible oil while maintaining incentives oroilseed producers (Persaud and Landes, 2006; World Bank, 1997). Josephet al. (2008) have recommended a package o reorms needed to improvethe competitiveness and eciency o Indias agricultural supply chains,including the promotion o retailer cooperatives and associations to procuredirectly rom armers, developing proessional standards and productbranding, improving credit and banking services, and simpliying regulatoryand licensing requirements in the marketing chain.

1A bound tari is the maximum tari

rate allowed by the World Trade Orga-

nization. An applied tari is the actual

tari rate in eect at a countrys border.

2One business controls or owns di-

erent aspects o buying, processing,

selling, and delivering products and

services.


10/43

2Growth and Equity Eects o Agricultural Marketing Efciency Gains in India / ERR-89


The widespread underinvestment and ineciency in Indias agriculturalmarketing system, coupled with the high priority given to boosting agri-cultural growth, makes it important to explore the potential impacts oimprovements in marketing eciency that may occur as a result o reducedregulation and increased investment in domestic markets. This studyreviews and builds on industry studies o agricultural marketing eciencyto provide an economywide assessment o the impacts o potential increasesin marketing eciency. The analysis uses a computable general equilibrium(CGE) model to estimate the eects o increased agricultural marketingeciency on economywide production, consumption, prices, and welare.Because the equity implications o policy reorm are a key considerationor Indian policymakers, the ramework also accounts or the distribution oimpacts on 10 representative household types classied by rural and urbanlocation and by income class. The impacts o increased marketing eciencyare then contrasted with those or two controversial and more commonlyaddressed areas o agricultural reorm: liberalization o agricultural trade andremoval o agricultural input subsidies.

The results indicate that, when marketing costs are reduced, impacts

include higher agricultural producer prices, lower consumer ood prices,and increased real consumption or all householdshigh and low income,rural and urban. Increasing the eciency o agricultural marketing tends tohave greater positive impacts on consumers and producersparticularly onlower income householdsthan do other more traditional reorms, suchas reducing agricultural input subsidies or removing barriers to agriculturaltrade. A key implication is that measures to improve marketing eciencyoer a potentially valuable complement to agricultural subsidy and tradereorms.


11/43

3


Agricultural Policy, Investment, and

Marketing Efciency in India

Marketing services are the wholesale and retail trade services used to bringdomestically produced goods rom the producer to the point o sale orintermediate and nal demand. In Indias agricultural and ood markets,the marketing chain typically consists o a primary market where armers

sell their product, wholesale markets where wholesalers who purchase romprimary markets sell to processors and retailers, and retail markets whereretailers sell to consumers. In many cases, additional intermediaries maybe present in supply chains, including village collectors who purchase romarmers and sell in primary markets or subwholesalers who operate betweenwholesalers and retailers. Many, i not most, o the intermediaries in Indiasagricultural supply chains are small-scale, nonintegrated marketing enter-prises operating in the unorganized sector, or which little ocial data areavailable.

Inecient marketing systems tax producers by placing downward pres-sure on arm gate prices and tax consumers by raising retail prices, thus

representing a potentially important target or reorm. In India, agricultureaccounts or about 18 percent o gross domestic product (GDP) and 58percent o employment, and ood accounts or about 47 percent o consumerexpenditure (Reserve Bank o India, 2007; Government o India, Ministry oStatistics and Programme Implementation, 2005, 2008). Thus, in India, likeother developing countries, the economywide impacts o improved marketingeciency on ood demand, output, and trade are potentially larger than inmore developed countries where agriculture and ood account or smallershares o the economy. Improved perormance o agricultural marketing mayalso have desirable equity outcomes by reducing ood costs or low-incomerural and urban consumers, who tend to allocate the largest budget shares orood.

The signicance o high marketing costs as a tax on arm output is evidentin studies that have quantied government intervention in Indian agricul-ture, in which the combination o trade restrictions and high marketing costsoten result in producer prices below import parity prices or many o Indiasmajor arm commodities (U.S. Department o Agriculture, 1994; Mullenet al., 2005). There is also strong evidence that an array o central and stategovernment policies have tended to undermine incentives or private invest-ment by armers and agribusinesses. These policies may have contributed toan inecient marketing system characterized by small-scale, nonintegratedmarketing service enterprises (Landes, 2008; World Bank, 1999, 1997).These policy measures include the ollowing:

Movement, storage, and pricing restrictions on ood commodities . TheEssential Commodities Act allows central and state governments toadministratively impose movement, storage, pricing, and quality restric-tions on most ood commodities. These restrictions are a major source opolicy risk or agribusiness investment, particularly in larger scale rmsthat hold relatively large stocks or operate across state borders.


12/43



Agricultural marketing restrictions . State agricultural marketing lawshave historically required all arm produce to be sold in government-regulated markets and prohibited private investment in markets andvertical coordination or integration between agribusinesses and armers.These rules are now in the process o reorm but with ull implementa-tion in just a ew states. Government-regulated markets continue todominate agricultural marketing in most o the country.

Land tenure policies . In addition to ceilings on land ownership, manystates either prohibit arm land rental or provide insucient legal protec-tions to either party to a rental agreement to support an active rentalmarket. These policies impede ormation o larger operational holdingsand armer access to investment credit and complicate vertical coordina-tion between growers, processors, and traders.

Scale limitations on agricultural processing frms. Until the late 1990s,most agricultural processors were required to be small-scale industries,restricted to xed capital assets o no more than 10 million rupees (Rs)(about $247,000). Although most scale restrictions have been removed,agricultural marketing and processing remain dominated by small-scalerms.

Credit policies . The high cost and limited availability o credit likelyrestricted new investment in agriculture and agribusiness during the1990s. Terms and availability or institutional credit have improvedsignicantly since 2000, but a large share o armers and agribusinessesstill depend on high-cost credit rom traditional moneylenders.

Tax policies. Until 2000, high excise tax rates o 8-16 percent on mostprocessed agricultural products raised costs and deterred investment inood processing. These tax rates have been reduced sharply since 2001.

This policy environment, although changing since the late 1990s, hascontinued to be unriendly to private investment in agricultural marketingand processing rms, particularly larger, integrated agribusinesses, and hascorresponded with sluggish investment in Indian agriculture and agribusi-ness. During 2003-05, the share o investment in Indias GDP was 27 percentor the economy as a whole. The share o agricultural investment in agricul-tural GDP was just 6 percent or investment in production agriculture and 15percent or investment in production agriculture and agricultural marketinginrastructure (g. 1).3 Further, while total investment in the Indian economyhas been accelerating and grew about 6.5 percent annually in real termsduring 1997-2004, investment in agriculture has been slowing and grew just

2.2 percent annually during the same period. Additional evidence o the lowlevels o investment in Indian agribusiness includes the ollowing:

Although more than 90 percent o Indias ood grains receive primaryprocessing, mostly in medium- or small-scale mills, relatively smallshares o other ood products are processed. About 2.2 percent o ruitand vegetable production receives any processing, about 35 percento milk is processed (mostly as packaged fuid milk), and between 6and 21 percent o poultry and meat is processed (U.S. Department oAgriculture, 2008).

3Intercountry comparisons are

complicated by data limitations anddierences in denitions. I one uses

comparable broad denitions o invest-

ment in agriculture, Indias agricultural

investment-to-GDP ratio o 15 percent

or 2003-05 alls well below Brazils 48

percent. I one uses comparable narrow

denitions, Indias investment-to-GDP

ratio o 7 percent also alls below

Chinas 10 percent (Landes, 2008).


13/43

5


The scale o most o Indias soybean-processing plants is in the range o170 tons/day, ar below the average capacity o U.S. plants (2,700 tons/day) and the even larger average capacity o new plants in Brazil andArgentina (Reca, 2003).

Only 1-2 percent o ood products are sold through modern chain retailestablishments (U.S. Department o Agriculture, 2006).

The strong expansion o consumer ood demand, along with improvednational tax and credit conditions and implementation o agricultural

marketing reorm in some states, has improved the climate or investment inIndian agriculture since 2001. There is evidence o increased investment inmodern ood retailing by domestic and multinational rms (Landes, 2008;Reardon and Gulati, 2008). Findings rom research on Indian agriculturalmarkets indicate that these new private-sector initiatives could signicantlyimprove marketing eciency:

Vertical integration in Indias broiler industry is improving eciencyin production and, particularly, marketing. Variable production costsaveraged 10 percent lower and producer-to-retail marketing margins 60percent lower in regions where vertical integration accounted or thelargest shares o production (Landes et al., 2004).4

Domestic markets or wheat and rice are inecient because o largenumbers o intermediaries, high physical losses, and disincentivesor private investment (World Bank, 1999). The average producer-to-retail markup or unprocessed, ungraded, bulk wheat is estimated at 52percent o the producer price in the private sector, with the producer-to-wholesale markup or the public-sector Food Corporation o India (FCI)estimated at 51 percent o the producer price (Gandhi and Koshy, 2007).Reducing the role o the FCI and increased investment by the private

4The marketing margin is the dier-

ence between the retail and arm price.

Figure 1

Gross fixed capital formation (GFCF) in India

as share of gross domestic product (GDP)

Percent of GDP

1Includes onfarm investment and off-farm investment in agriculture-related infrastructure.

2Includes only on-farm investments.

Source: Reserve Bank of India, Handbook of Statistics on the Indian Economy;

Government of India, Ministry of Agriculture, 2003.

Total GFCF/total GDP

0

5

10

15

20

25

30

1970 75 80 85 90 95 2000 05

GFCF in and for agriculture/

agricultural GDP1

GFCF in agriculture/agricultural GDP2


14/43



sector have the potential to increase marketing eciency and reducecosts (Jha et al., 2007).

Oilseed and product marketing costs in India are infated because thescale, technology, and capacity utilization rates o oilseed-processingplants are well below international standards. Eciency gains associatedwith deregulation o marketing and processing o Indias major oilseedshave been estimated at 22-44 percent (World Bank, 1997).

The eciency o Indias horticultural product markets is impeded byhigh logistical costs, produce wastage rates o 20-40 percent, and lowdomestic quality standards. Indian growers, on average, receive 12-15percent o the retail price o horticultural produce, compared with 30-40percent in the United States, despite the little value that is added in theIndian market (Matoo et al., 2007).

Potential Implications o Marketing Efciency Gains

The relatively low levels o eciency in Indias agricultural marketingchains tax the output o Indian producers and the consumption o Indianconsumers, but also oer a potential source o improved agricultural compet-itiveness and income growth over the longer term. Deregulation o agricul-tural markets, such as removal o private movement and storage restrictions,can benet both producers and consumers by improving incentives orprivate investment in market inrastructure, new technology, and more e-cient horizontally or vertically integrated agribusinesses. Producers stand togain when improved marketing eciency increases demand and prices ortheir products. Consumers can gain i more ecient wholesale and retailtrade services reduce consumer prices.

The net eect o improvements in marketing eciency on agricultural

producer prices is, however, not clear cut. For producers, an eciency-induced all in the retail cost o ood should increase quantity demandedand create upward pressure on producer prices. At the same time, this priceincrease could be oset by declines in producer prices as gains in marketingeciency reduce the prices o intermediate agricultural inputs and thereorelower the costs o production. For consumers, purchase price declines associ-ated with more ecient marketing could be oset by upward price pressurerom increased quantity demanded, particularly i demand is highly respon-sive to changes in price (price-elastic demand) and/or expanding supplythrough production or trade is inhibited by technical or resource constraints(price-inelastic supply). Across commodity sectors, marketing eciencygains are likely to aect prices, output, and demand dierently, depending on

the sectors use o marketing services and potential or eciency gains.


15/43

7


The totalor economywideimpacts o increasing the eciency o agri-cultural and ood marketing may be substantial in a developing country likeIndia because the agricultural sector accounts or a large share o economicactivity and ood accounts or a large share o household budget expendi-tures. Gains in arm output associated with higher arm prices may signi-cantly aect aggregate supply o agricultural goods as well as all other goodsand services and agricultural trade. Lower ood prices or consumers can leadto potentially large economywide changes in consumption expenditures or

ood and other goods and in household savings and investment.

In addition to the potential or important impacts on aggregate produc-tion, consumption, and investment, more ecient agricultural and oodmarketing services also have the potential to benet low-income consumersin particular. The distribution o benets across income groups depends ondierences in ood budget shares across low- and high-income households,whether their ood basket is composed o commodities associated withmore or less use o marketing services, and dierences in household incomesources. A relatively large number o Indias poorest households are in ruralareas, spend a relatively large share o income on ood, and earn most otheir income as either armers or arm laborers. To the extent that improvedmarketing eciency boosts arm prices and arm output, net sellers o oodtend to gain. All net buyers o ood tend to gain as consumer prices o oodall and labor demand and wages rise. Lower income rural and urban house-holds, which tend to allocate the largest shares o income to ood and dependon wage income, stand to benet more rom marketing eciency gains thanhigher income households. However, higher income households can alsobenet rom more ecient ood marketing because they tend to purchasemore o the types o oods associated with intensive use o marketingservices, such as processed or retail convenience oods.


16/43



Estimated Costs o Marketing

Agricultural and Food Products

The costs o wholesale and retail marketing services are embedded in thepurchase prices paid by rms or intermediate inputs, by households orprivate consumption, by government and investors or purchases o agricul-tural goods, and or exports. We use multiple sources to develop a dataset

o estimated current agricultural and ood marketing costs or domestic andexport sales on a commodity basis.5 For the purpose o general equilibriumeconomic modeling, it is necessary to have comprehensive, internally consis-tent data that describe the input and output relationships o rms technolo-gies and the economywide fows o income and expenditures. In this case,we also require consistent data on the supply and demand or wholesale andretail trade services or major agricultural products. We drew primarily onthe internally consistent data on the Indian economyincluding total expen-ditures on wholesale and retail trade services by each category o demandrom the 2001 India database (version 6) o the Global Trade AnalysisProject (GTAP).6

In the GTAP data, as in the Indian national accounts on which they arebased, purchasers expenditures on each good are reported net o marketingcosts and only total trade marketing service expenditures are reported oreach category o demand (intermediate, households, government, invest-ment, and exports). These trade service data cover the ull range o marketingormats in both urban and rural areas. They include the public sectors rolein marketing some ood products; the small, private, organized sectorcomposed o relatively large-scale retail outlets, such as supermarkets; andthe marketing services provided by Indias large, private unorganizedtrading sector composed o myriad small shops, stalls, open markets,and bazaars (Government o India, Ministry o Statistics and ProgrammeImplementation, 2000, 2007). Note that, within the dataset, private-own-

consumption o agricultural goodswhich accounts or the largest share othe goods produced and consumed by Indian arm householdsis assumedto have no associated marketing costs. India has many small armersabout81 percent o all operational holdings, accounting or about 39 percent allarmland, are 2 hectares or lessand home consumption is estimated toaccount or signicant shares o total use o most crops (Government oIndia, Ministry o Agriculture, 2008).

To disaggregate total expenditures on marketing services by each purchaserinto marketing margins paid or each commodity, we relied on a combinationo judgment and ndings rom recent studies o Indias ood grain, oilseed,poultry, and horticultural product markets. A major inconsistency had to bereconciled, however, between the data on total expenditures on marketingservices reported in the GTAP data (and the Indian national accounts) and themuch larger expenditures on marketing margins reported in the commoditymarket studies. One reason or the inconsistency is that the Indian nationalaccounts data implicitly average in the zero margins or on-arm consump-tiongoods that do not enter the marketplace. A second reason may be theinaccuracy o the marketing cost data rom the various marketing studies thatare oten based on available data or a ew markets or regions that may notrepresent national average data. A third reason could be a possible tendency

5Author calculations to develop esti-

mated marketing margins are described

in urther detail in Appendix 1.

6GTAP database development or

each country or region combines cur-

rent international data on trade fows,

applied taris, agricultural subsidies,

macroeconomic indicators, and energy

use with contributed national input-

output tables to create a balanced,

internally consistent global database

or a specied base year. The GTAP v6

database used in this study incorporates

Indias 1993-94 national input-output

table contributed by Chadha and Pratap

(2006) in a balanced global database

or 2001.


17/43

9


or the Indian national accounts data to undercount the expenditures o tradeservices provided by small, rural and urban, unorganized sector rms thatmarket most agricultural products in India.

The discrepancy in the size o marketing margins between the availablesources is too large to be accommodated in the model database without asignicant revision o the Indian input-output tables, which would in turnnecessitate introducing numerous new assumptions and sources o error.

Thereore, the approach used is to maintain the internal consistency o themodel database but, to the extent possible, allocate the expenditures on agri-cultural marketing services to various commodity sectors in a manner thatrefects the relative sizes o margins ound in the marketing studies.

The data on marketing margins developed rom various market studies, alongwith the margins used or economywide model analysis, are provided intable 1. Also reported (in the nal two columns o table 1) are estimates o

Table 1Estimates o trade service costs by commodity group or India

Commoditygroup

Marketing study-based estimates Model estimates4

Marketed produce All produce

Unprocessed1Processingadjustment2

ProcessedMarketable

surplus3

Average total

costs at retail

Average total cost

At retail For export

Percent o consumer pricePercent oproduction

Percent o consumer price

Rice 0.334 1.10 0.368 0.733 0.269 0.083 0.033

Wheat 0.377 1.10 0.415 0.655 0.272 0.092 0.037

Corn 0.356 1.20 0.427 0.747 0.319 0.097 0.000

Other cereals 0.356 1.10 0.391 0.569 0.223 0.089 0.000

Pulses 0.485 1.10 0.534 0.798 0.426 0.114 0.047

Fruit & vegetables 0.700 1.20 0.840 0.859 0.721 0.126 0.052

Oilseeds 0.223 1.00 0.223 0.922 0.206 0.053 0.020

Sugar 0.386 1.40 0.540 0.985 0.532 0.118 0.047

Fibers 0.400 1.30 0.520 0.953 0.495 0.091 0.036

Crops, nec 0.435 1.40 0.608 0.802 0.488 0.113 0.048

Cattle 0.400 1.00 0.400 0.800 0.320 0.092 0.000

Poultry & hogs 0.300 1.00 0.300 0.800 0.240 0.065 0.025

Bee & mutton 0.400 1.20 0.480 0.900 0.432 0.058 0.032

Poultry meat & pork 0.200 1.20 0.240 0.900 0.216 0.014 0.000

Oil meals 0.200 1.00 0.200 0.950 0.190 0.039 0.000

Oils & ats 0.500 1.20 0.600 0.950 0.570 0.110 0.052

Dairy 0.300 1.60 0.480 0.700 0.336 0.108 0.042

Food, nec 0.800 2.00 1.600 0.850 1.360 0.118 0.061

nec = Not elsewhere classifed.1ERS estimates based on commodity market studies and 2002-04 average armgate-retail price spreads.2ERS estimates.3Government o India, Ministry o Agriculture, 2008 Agricultural Statistics at a Glance; ERS estimates.4Model estimates include the portions o the study-based commodity-specifc estimates that are consistent with the trade service expendituredata included in Indias national input-output accounts and the Global Trade Analysis Project (GTAP) database. In most cases, the expendituresimplied by the marketing cost studies exceed the corresponding trade service totals in the GTAP input-output accounts and have to be scaleddown to maintain consistency with the input-output accounts.


18/43



conversion actors and marketable surplus ratios used to adjust the marketstudy data or processing and on-arm consumption. For all commodities,the marketing margins that are consistent with the ndings o various marketstudies were scaled down to meet the constrained totals in the balancedeconomywide database. Margins are estimated separately or domestic salesand exports. Overall, the estimates or the domestic commodity marketingmargins used in this analysisconstrained by the level o total marketservice expenditures in Indian national accountsaverage about 21 percento those based on the ndings o commodity market studies. The estimatedmarketing margins or export sales, scaled to the control total in the modeldatabase, are smaller than those or domestic sales and refect that marketingmargins or export sales do not span the ull arm-to-retail supply chain.Because o the downward scaling, the model data may understate the size oIndias actual agricultural marketing margins and the impacts o improvedmarketing eciency.

The marketing margin estimates are marketing services costs as a percentageo the retail price. Estimated margins are relatively low or primary agricul-tural products, such as rice and wheat, which are oten consumed on arm

and oten minimally processed (g. 2). Estimated marketing margins ordomestic sales are highest or dairy, resh ruits and vegetables, processedoods, and sugar. The export margins are highest or resh ruits and vegeta-bles, other processed oods, pulses, and crops not elsewhere classied.7

Although the available data provide a basis or dierentiating marketingmargins across the commodity sectors, they do not permit dierentiatingmarketing costs and margins or the various categories o consumers, suchas rural or urban location and high- or low-income. We, thereore, assumea uniorm marketing margin across domestic demand categories or eachcommodity. As a result, to the extent that urban and/or higher incomeconsumers tend to purchase relatively more marketing services, the data may

understate those expenditures. Similarly, to the extent that low-income and/orrural consumers tend to buy ewer marketing services, the data may overstatetheir expenditures.

7See appendix 1 or more detail on how

these marketing costs by commodity

were estimated.

Figure 2

Estimated agricultural marketing margins in India

Marketing services/retail price

nec = Not elsewhere classified.

Source: Saluja and Yadav, 2006; Global Trade Analysis Project Version 6 database; author calculations.

ExportDomestic

0

2

4

6

8

10

12

14

Rice

Whe

atCo

rn

Othe

rgrains

Pulse

s

Fruit

and

vegetable

s

Oilse

eds

Oilse

edmeals

Oila

ndfats

Suga

r

Fibe

rs

Crop

s,ne

c

Cattle

and

sheep

Poult

ryand

hog

s

Beef

and

mutton

Poult

ryand

pork

Dairy

and

produ

cts

Othe

rfoo

ds


19/43

11


Household Income and Expenditure Patterns

In the context o the recent poor perormance o Indias rural sector and theimplications or poverty reduction, policymakers are likely to be particu-larly concerned about how various types o households, particularly ruraland low-income households, are likely to gain or lose rom a policy change.The distribution o outcomes across households is determined by the diver-sity in their sources o income and by the composition o their spending. Toexplore the impacts o marketing eciency gains across households, we usea database developed by Saluja and Yadav (2006) to disaggregate the GTAPdatawhich describe a single aggregate householdinto multiple house-holds dened by rural or urban location and by income class. The Saluja andYadav database describes 10 household types in India, 5 urban (U) and 5rural (R), each with 5 expenditure classes: abject poverty (R1, U1); poverty(R2, U2); middle income (R3, U3); upper income (R4, U4); and high income(R5, U5). We use the income classes dened by Saluja and Yadav to disag-gregate the single aggregate household in the GTAP data to the same 10household types.8

In the model database, expenditures on ood, comprised o primary agri-cultural products and processed oods and their related marketing margins,account or 47 percent o Indias private consumption expenditures on goodsand services (excluding expenditures on savings and taxes). In generalandas expected in a developing country like Indiaood accounts or a largershare o expenditure by rural households than by urban households, andpoorer households spend proportionately more on ood than do wealthierhouseholds, in both rural and urban areas (g. 3). Also as expected, thecomposition o ood expenditures varies across income. In general, poorhouseholds consume more whole grains and other unprocessed agriculturalproducts than do high-income households. At higher incomes, householdsconsume more dairy, poultry, and other processed oods, as opposed to

primary agricultural products (g. 4).

8See appendix 1 or a detailed discus-

sion o how the household data were

disaggregated.

Figure 3

Composition of Indian household consumptionexpenditures by household type

Percent of expenditures

R = Rural; U = Urban; R1, U1 = Abject poverty; R2, U2 = Poverty; R3, U3 = Middle income;

R4, U4 = Upper income; R5, U5 = High income.

Source: Saluja and Yadav, 2006; Global Trade Analysis Project Version 6 database.

Primary agriculture

Processed food

Manufactured goods

Services

0

10

20

30

40

50

60

70

80

90

100

R1 U4U3U2U1R5R4R3R2 U5

Household type


20/43



Marketing costs or agricultural and ood products in India account or asubstantial proportion o household expendituresabout 5 percent o totalIndian household spending on all goods and services. For all households,expenditures on marketing services are about 11 percent o the ood budget.Across households, the share o total household expenditures on oodmarketing services tends to all as household incomes rise, a pattern thatrefects the declining share o ood expenditures in the household budget asincomes rise (g. 5). Within the ood expenditure basket, however, the shareo spending on marketing services rises as household incomes rise (g. 6).9This pattern refects the shit toward consuming goods with higher marketingcosts as household incomes rise. Lower income households, in contrast, tend

9The exception to this pattern is the

highest income urban households (U5),

where the share o household ood

expenditures spent on ood marketing

services is relatively low. The likely

explanation is that these, generally

larger, households purchase relatively

large amounts o ood that require little

processing or other marketing services,

perhaps to meet the requirements o

large extended amilies and/or house-

hold servants.

Figure 4

Composition of Indian household expenditureson food by household type

Percent of food expenditures

nec = Not elsewhere classified; R = Rural; U = Urban; R1, U1 = Abject poverty;

R2, U2 = Poverty; R3, U3 = Middle income; R4, U4 = Upper income; R5, U5 = High income.


Food grains

Fruits and vegetablesOilseeds and products


Sugar

Meat and products

Dairy products

Crops and food, nec

0

10

20

30

40

50

60

70

80

90

100

Household type

Figure 5

Food marketing costs in India as share of total household

consumption expenditures by household type

Percent of consumption expenditures



Source: Saluja and Yadav, 2006; Global Trade Analysis Project Version 6 database;


1

2

3

4

5

6

7

Household type


21/43

13


to consume unprocessed agricultural products with relatively low marketingcosts.

Indian households also dier in their sources o incomereturns rom land,labor, and capital. Wages are the more important income source or poorerhouseholds in India, with the role o capital and land-based earnings risingwith income (g. 7). Any changes in wage incomes are thereore likely tohave the most impact on lower income rural and urban households, whereas

changes in returns to capitaland to landwill be elt mostly by higherincome households.

Figure 7

Sources of income for Indian households by household type

$U.S. billions




Land

Labor

Capital


20

40

60

80

100

120

Household type

Figure 6

Food marketing costs in India as share of household food

consumption expenditures by household type

Percent of food expenditures



Source: Saluja and Yadav, 2006; Global Trade Analysis Project Version 6 database;

author calculations.

R1 U4U3U2U1R5R4R3R2 U510.4

10.6

10.8

11.0

11.2

11.4

11.6

11.8

Household type


22/43



Household savings and tax rates dier substantially across Indian house-holds. Savings rates are either negative or negligible in lower income ruraland urban households but rise with income: The highest income rural (R5)and urban households (U5) save 33 percent and 23 percent o their respectiveater-tax incomes (g. 8). Tax expenditures are relatively low or all Indiansbut tend to be highest in middle-income rural and urban households.

Figure 8Indian household consumption, savings, and taxes by household type

$U.S. billions

R = Rural; U = Urban; R1, U1 = Abject poverty; R2, U2 = Poverty; R3, U3 = Middle income;R4, U4 = Upper income; R5, U5 = High income.


Consumption

Savings

Taxes

R1 U4U3U2U1R5R4R3R2 U5-10

10

30

50

70

90

110

Household type


23/43

15


Impacts o Potential Marketing Efciency Gains

The analytical ramework used in this study is a single-country, computablegeneral equilibrium (CGE) model developed by the International Food PolicyResearch Institute (Lgren et al., 2002). The model explicitly accounts ormarketing margins as a xed input-output quantity ratio between wholesaleand retail marketing services and the sale o domestically produced ood andagricultural goods in domestic and export markets. The purchase price orood and agricultural commodities in domestic and export markets is denedas the sum o the producer price plus the cost o the marketing service,plus any sales tax. Any change in the cost o marketing services directlyaects the commodity purchase price, and through the demand response andchanging intermediate input costs, may also aect the producer price.

The potential impacts o eciency gains are analyzed in the model byimposing a 50-percent increase in total primary actor productivity in theproduction o wholesale and retail trade services used or ood and agricul-tural commodities. This 50-percent increase implies that the same quantityo labor and capital inputs used to produce marketing services now gener-

ates 50 percent more output, thus reducing the costs o wholesale and retailmarketing services. These eciency gains are assumed to result romincreased investment, improved technology, and enhanced integration inagricultural supply chains as rising consumer demand and domestic regula-tory reorm improve the climate or agribusiness investment and productivitygrowth.

We impose a relatively large increase in marketing productivity because othe evidence that the scope or eciency gains is substantial and becausethe model data may understate existing marketing costs in the arm sector.The 50-percent productivity increase is equivalent to a compound annualproductivity growth o about 4.5 percent over about 10 years, a timerame

compatible with a medium-run outcome in which land, labor, and capitalmarkets ully adjust to the productivity gain. Although we expect the scopeor marketing eciency gains to vary by commodity, or simplicityanddue to lack o good detailed inormation by sectorwe assume the sameproductivity gain across all ood and agricultural commodities. A 50-percentincrease in productivity may not be easible in all sectors, but the anecdotalinormation derived rom studies o Indias ood grain, oilseed, poultry, andruit and vegetable industries suggests that it is a plausible overall average.

Economywide Impacts

The assumed improvement in marketing eciency raises real GDP by 1

percent, refecting the increased productivity o Indias xed aggregate actorsupplies in providing marketing services (table 2). The scenario also gener-ates a 1.4-percent increase in real household consumptiona welare indi-cator that accounts or changes in the quantities o household consumptionvalued at base period prices. Additionally, improved marketing eciencygenerates increases in investment that could increase uture economywideoutput and consumption, although these dynamic impacts are not captured inthe model. At the economywide level, the scenario has a small but positiveimpact on the producer price index, but implications or prices become moreimportant when the ocus shits to the agricultural commodity sectors.


24/43



Other economywide impacts include more government tax revenues associ-ated with increased GDP and a small increase in the aggregate demand orimports (although demand or agricultural imports alls). Import growth leadsto a small depreciation o the real exchange rate and an increase in aggregate

exports. Wages and rental rates or land and capital increase in the scenario,driven by higher actor demand and the model assumption that labor and theother primary actors are ully employed. In the India case, however, laborunemployment and underemployment are substantial. Thereore, an alterna-tive interpretation o the labor market result is that the upward pressure onwages in the labor market could increase employment and/or reduce under-employment instead o increasing wages. In this event, the increase in GDPwould be even larger as more o Indias productive actors are put to use.

Agricultural Sector Impacts

Increased eciency in wholesale and retail marketing o agricultural and

ood products signicantly aects producer and sales prices o domesticagricultural and ood products (table 3). The reduction in marketing costsbetween producers and consumers reduces consumer prices and raisesdemand or most agricultural and ood products. Greater demand increasesproduction and producer prices or most agricultural and ood products; thegains in marketing eciency, thereore, are shared between producers andconsumers. The largest production gains tend to be in such categories asoilseeds and products, sugar, dairy, and processed products (ood, not else-where classied), which mostly have relatively high marketing costs (g. 9).

Table 2

Aggregate impacts o efciency gains in agricultural and oodmarketing in India

Variable50% increase in total actor productivity

in agricultural and ood marketing

Percent change rom base

Real gross domestic product 1.0

Real household consumption 1.4

Real investment demand 0.4

Government revenue 1.0

Producer price index 0.4

Land rents 5.6

Wages 1.6

Capital rents 0.1

Exports 0.7

Agricultural exports 3.9

Imports 0.7

Agricultural imports -0.9

Exchange rate - rupees/U.S. dollar 0.7

Source: ERS model results.


25/43

17


The rise in arm output associated with marketing efciency gains urther

increases Indias normally substantial positive agricultural trade balance.

Agricultural exports, which include a broad array o grain, oilseed, horticul-

tural, and animal products, rise about 4 percent, whereas importsdominated

by edible oils and pulsesall about 1 percent. Note that these results or

trade, as well as or other variables, account only or medium-term adjust-

ments to greater marketing efciency and not or the longer term impacts o

higher, sustained economic growth on ood demand and trade.

Table 3

Price eects o efciency gains in agricultural and

ood marketing in India

Selected sectors

50% increase in total factor productivity

in agricultural and food marketing

Producer price Consumer price


Rice 1.2 -0.8

Wheat 1.8 -0.5

Corn 2.2 -0.2

Fruit and vegetables 2.3 -1.0

Oils, fats 0.4 -2.1

Sugar 1.0 -1.8

Fibers 2.4 -0.1

Poultry and pork 2.7 0.6

Dairy products 1.9 -0.7

Food products, nec -0.1 -4.0


Figure 9

Production impacts of agricultural and

food marketing efficiency gains in India




0

1

2

3

4

5

6

7

Rice

Whe

atCo

rn

Fruit

and

vegetable

s

Oilse

eds

Oilm

eal

Oiland

fats

Suga

r

Fibe

rs

Poult

ryand

hog

sDa

iry

Food

,nec

Food

retailt

rade


26/43



Household Impacts:Income and Rural/Urban Distribution

How the impacts are distributed across households hinges on how pricechanges aect the costs o the bundles o goods and services purchased byhouseholds and how changes in returns to land, labor, and capital aect earn-ings by rural and urban households at dierent income levels. The resultsindicate that all households benet rom marketing eciency gains, whichreduce consumer prices o most ood items. Lower income householdsbenet primarily because they allocate a larger share o their expenditures toood (g. 10). Higher income households also benet because they tend tospend more than lower income households on oods, such as dairy products,that use larger amounts o marketing services.

Income gains tend to avor rural households and poor households in bothurban and rural areas because wages rise more than do returns to capital.And, or rural householdswhether low- or high-incomehigher earningsare driven by increased returns to land associated with higher arm output.

The gains to rural households may be overstated and the gains to urbanhouseholds understated by the model because the underlying data onmarketing margins are an average o on-arm consumption, which doesnot incur marketing costs, and marketed consumption, which does incurmarketing costs. To the extent that lower income ruralbut not urbanhouseholds spend less than the average marketing costs on goods in theirexpenditure baskets, their estimated gains are overstated. Similarly, theimpacts on higher income households, both rural and urban, are understatedto the extent that they purchase products with higher-than-average marketingcosts or a given product category. Although it is important to realize thesepotential biases in the results, the size o the estimated gains or both ruraland urban low-income households suggests that the gains would remain

substantial even i these data deciencies could be corrected.

Figure 10

Effects on household real consumption of agricultural and

food marketing efficiency gains in India by household type






Household type

0.0

0.5

1.0

1.5

2.0


27/43

19


Comparing the Impacts o Increased

Marketing Efciency With Agricultural

Input Subsidy and Tari Reorm

Reorm o Indias agricultural subsidy and tari policies have been topicso debate by Indian policymakers and trade partners since at least the early1990s. Rising input subsidy outlays are oten criticized or causing actor

market distortions and diverting public resources rom productive invest-ments, but they have been sustained and even enlarged because o concernsor the welare o Indias many small-scale armers. Indias high agriculturaltaris are challenged in the current multilateral negotiations and elsewhereor imposing high costs on consumers and hindering longer term competi-tiveness, but liberalizing reorms are also strongly resisted largely because oconcerns with armer welare.

The goal o comparing subsidy and tari reorm with increased agriculturaland ood marketing eciency is to assess and compare their eects on theeconomy, commodity sectors, and households. These broad areas o agricul-tural policy reorm oer a general representation o the policies that might

be adopted in India, but they are not necessarily comparable in terms o themagnitude o their impacts.

Indias Agricultural Subsidies

The cost o Indias major agricultural input subsidies or electrical power,ertilizer, and irrigation water has grown about 6 percent annually in realterms since 1990, reaching nearly Rs500 billion (about $10 billion) in2002/03 (g. 11) (Mullen et al., 2005; Gulati and Narayanan, 2003). Thecost o providing ree or subsidized (depending on the state) electricity oragriculture accounts or more than two-thirds o total input subsidies, aswell as most o the growth since 1990. Irrigation is a key actor in boostingcrop yields, but the policy o subsidizing electricity is also now widelyacknowledged to be promoting ineciency in water use and overpumping ogroundwater. Fertilizer subsidies are provided to armers in the orm o pricesubsidies or domestic and imported ertilizers and to the ertilizer industryin the orm o preerential prices to oset the losses o higher cost plants. Forthis study, we include only subsidies going to armers, which amounted toabout Rs70 billion ($1.4 billion) in 2002/03 and have tended to decline sincethe mid-1990s. Subsidies or canal irrigation water were about Rs60 billion($1.2 billion) in 2002/03 but have shown little growth.

The rising outlays or input subsidies, which were equivalent to about 11

percent o total agricultural output in 2002/03, are o growing concern orIndian policymakers. By comparison, public investment in agriculture andagricultural inrastructure amounted to just 5 percent o agricultural GDP in1999-2000 (the most recent period available) and has shown relatively slug-gish real annual growth o less than 1 percent since 1990. Although reormo Indias subsidy policies would potentially make available more publicresources or productive investments in irrigation, research, market services,and inrastructure, withdrawing the subsidies has proven dicult because otheir popularity with rural voters and, in the case o power and water subsi-dies, the state governments that administer them.


28/43



The Government o India has also been incurring substantial and risingbudgetary costs or operating the system o minimum support prices (MSPs),public distribution, and storage or wheat and rice. The real cost o what iscalled the ood grain subsidy has climbed steadily since the late 1990s,reaching an average o Rs242 billion ($5.5 billion) during 2004/05-2006/07(Government o India, Ministry o Finance, 2008). Rising costs havestemmed rom increased government procurement and storage costs associ-ated with deending higher support prices and with introducing new andmore highly subsidized distribution programs or poor and disadvantagedgroups.

Despite these substantial and rising costs, the ood grain subsidy is notspecically included in the data and subsidy-reduction scenario used in thisstudy or two reasons. First, the system o supporting MSPs is combinedwith border policies, such as export restrictions, and domestic measures, suchas levies on rice millers and grain movement restrictions, that can also taxproducers. Calculations o producer subsidy equivalents (U.S. Departmento Agriculture, 1994) and producer support estimates (Mullen et al., 2005)that incorporate careul comparisons o world prices and Indian domesticprices show that these policies net out as subsidies in some years and taxes inothers. According to this evidence, there is no clear rationale or character-izing the bundle o expenditures and policies called the ood grain subsidy

as a subsidy (or tax) on producers. Second, Indias domestic marketing costsare an important actor in assessing levels o market price support, and theimpacts o these costs are already being studied separately.

Indias Agricultural Taris

India completed the elimination o quantitative restrictions on agriculturalimports in 2001 but maintains relatively high bound taris on most agricul-tural commodities (g. 12). Reductions in Indias agricultural taris have

Figure 11

Major Indian agricultural input subsidies, 2001

$U.S. billions

Source: Global Trade Analysis Project Version 6 database; author calculations.

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

Fertilizer

Power

Irrigation

Rice

Whe

atCo

rn

Othe

rgrains

Pulse

s

Fruit

and

vegetable

s

Oilse

eds

Suga

r

Fibe

rs

Othe

rcrops


29/43

21


been an important target o the United States and other trading partners inbilateral and multilateral settings, including the current Doha Round.

India remains reluctant to commit to reducing bound agricultural taris thataord protection to its large number o small-scale armers and agribusi-nesses. However, the government has made unilateral reductions in appliedrates to acilitate imports and contribute to increased price stability orcommodities in short supply, including pulses, vegetable oils, and, most

recently, wheat (Landes, 2008). There is evidence that reductions in at leastsome currently prohibitive taris could yield a signicant boost to industries,such as oilseed processing, that lack access to raw materials and to producerswho are now isolated rom world market prices (Persaud and Landes, 2006).However, in a number o cases, such as rice, oil meals, ruits and vegetables,and poultry meat, India has high applied taris on goods or which domesticprices are typically competitive in regional or global markets in order toassure adequate protection and bargaining strength with trading partners.In these cases, tari reductions would have little or no impact on importdemand. The model addresses the impacts o tari changes by treatingimported and exported products separately rom domestic goods, thuslimiting the impact o tari reductions on the Indian goods, which, despite

high taris, tend to have relatively low internal prices and import demand.

Recent studies by Ganesh-Kumar et al. (2006) and Polaski et al. (2008) o thepotential impacts o agricultural trade liberalization on the agricultural sectoras a whole have indicated that, in the short run, the benets to India romparticipating in a global agricultural reorm agenda would be positive butmodest. Polaski et al. stress the importance o setting negotiating terms thatbalance the interests o Indias poor households with the benets o improvedeciency and market opportunities.

Figure 12

Indian applied agricultural import tariffs, 2001

Percent ad valorem


Source: Global Trade Analysis Project Version 6 database.

0

20

40

60

80

100

120

140

Ric

e

Whe

at

Corn

Oth

ergrain

Puls

es

Fruits

and

veg

etable

s

Oils

eeds

Oilm

eals

Fats/oils

Suga

r

Fibe

rs

Crops

nec

Cattle

and

she

ep

Poultry

and

hog

s

Beef

and

mutto

n

Poultry

and

pork

Dairy

produ

cts

Food

sne

c

Beve

rage/toba

cco


30/43



Scenario Comparisons

The subsidy reorm scenario simulates removal o all agricultural input subsi-dies, and the tari reorm scenario simulates the removal o all agriculturaltaris. Because the eects o the subsidy and tari reorm scenarios and theincreased marketing eciency scenarios are not necessarily comparable interms o magnitude, our discussion ocuses on the patterns o their eects onthe economy and across commodities and households.

The removal o agricultural input subsidies leads to a relatively smallincrease in real GDP (g. 13), but higher producer and consumer prices orstaple ood commodities, particularly or grains and bers, which receivethe bulk o input subsidies (gs. 14 and 15), cause aggregate real house-hold consumption to all.10 Producer and consumer prices o nonsubsidizedcommodities all because o an overall decline in consumer expenditures anda shit in land use away rom crops that lose subsidies.

The decline in agricultural production reduces agricultural employment andeconomywide wages. However, the substantial government savings romeliminating subsidies boost national savings and investment, thus improvingprospects or uture growth in output, employment, and consumptiondynamic eects that are not accounted or in our static ramework. Withlower actor returns and higher prices or ood staples (wheat and rice), atleast in the near term, real consumption declines or all household types(g. 16). Consumption losses are greatest in poorer and rural households,refecting the combination o lower returns to land and labor and substan-tially higher prices or ood staples.

When agricultural taris are removed, the more ecient allocation oresources leads to growth in real GDP and increases in the returns to laborand capital and in aggregate real household consumption. But, the decline

in government tax revenue in this scenario causes aggregate savings andinvestment to all. The removal o agricultural taris has minor impacts onoverall producer and consumer prices and arm output, refecting relatively

10See appendix 3 or complete numeri-

cal results rom the scenarios.

Figure 13

Economywide effects of alternative agricultural

and food marketing policy scenarios in India


Eliminate agriculturaltariffs

Eliminate inputsubsidies

50% increase inmarketing efficiency


-6.0 -4.0 -2.0 0.0 2.0 4.0 6.0

Capital rents

Wages

Land rents

Producer price index

Real investment demand

Real household consumption

Real gross domestic product


31/43

23


competitive domestic prices and generally small amounts o trade in suchsectors as ood grains, ruits and vegetables, bers, and dairy. However,more heavily protected sectors with less competitive domestic prices,including ats and oils and poultry, ace potentially signicant producer

Figure 14

Producer price effects of alternative agricultural








-6 -4 -2 0 2 4 6 8 10

Food, nec

Dairy

Poultry and pork

Fibers

Sugar

Oils and fats

Fruits and vegetables

Wheat

Rice

Corn

Figure 15

Consumer price effects of alternative agriculturaland food marketing policy scenarios in India







Food, nec

Dairy

Poultry and pork

Fibers

Sugar

Oils and fats

Fruits and vegetables

Wheat

Rice

Corn

-35 -30 -25 -20 -15 -10 -5 0 5 10


32/43



price and output shocks when taris are removed. Higher returns to laborand capital lead to increased real consumption by all rural and urbanhouseholds. Consumption gains are largest in lower income households,driven by higher wages, relatively small increases in staple ood prices, andsignicant consumer price declines or highly protected items, particularlyats and oils.

Comparisons across the three scenarios suggest that improved agriculturaland ood marketing eciency can increase returns to land and labor andsignicantly boost aggregate real consumption relative to the other scenarios.The magnitude and distribution o household consumption gains associatedwith improved marketing eciency are similar to those in the tari removalscenario but in sharp contrast to the potentially adverse consequences osubsidy removal, at least in the near term. Unlike the subsidy and tarireorm scenarios, improvements in marketing eciency have the potentialto provide price benets to both producers and consumers, particularly lowerincome and rural households. In contrast, the results suggest that removingsubsidies could substantially raise consumer prices or ood staples, whilelowering taris could substantially reduce production in currently protectedsectors, such as oilseed products, poultry, and processed oods.

Finally, although the model does not account or dynamic changes in

savings and investment over time, the scenarios dier substantially in theirimplications or economywide savings, investment, and uture growth.The marketing eciency scenario suggests modest increases in savingsand investment, tari elimination reduces savings and investment, and theremoval o subsidies generates substantial gains in government savings andinvestment. A dynamic ramework that accounts or the eects o investmenton economic growth would likely indicate stronger growth benets romimproved marketing eciency and reduced subsidies compared with thatarising rom the removal o trade-distorting taris.

Figure 16

Household real consumption effects of alternative agricultural







Household type

-3.5

-2.5

-1.5

-0.5

0.51.5

2.5

3.5 Eliminate agriculturaltariffs




33/43

25


Conclusions

The results suggest that measures to improve agricultural marketing e-ciency in India, perhaps including strengthening public and private marketinginstitutions and improving the regulatory climate or private agribusi-ness investment, may substantially and broadly benet Indias economy.Advances in marketing eciency are shown to have the potential not only toincrease income and employment economywide, but to provide positive priceimpacts or both agricultural producers and consumers and benets to allhouseholdsparticularly or rural and low-income households.

In a broader sense, the results also suggest the potential or greater invest-ment and eciency in Indias agricultural markets to be a source o enhancedagricultural growth, trade, and competitiveness over the longer term.Whether new policies lead to rapid investment by modern retailers and othersin transorming markets or the process occurs gradually, the results suggestthat improvements in domestic marketing eciency, although boostingood demand, may also strengthen agricultural exports and reduce imports.However, these results neither account or changes in ood and eed demand

that are likely to occur as higher income growth is sustained over the longerterm nor account or constraints on crop and livestock production that mayemerge with increased land use intensity, yields, and eed demand.

Finally, this analysis highlights the dilemma o Indian policymakers as theyace domestic and international pressures or subsidy and tari reorm.Despite any economywide benets rom subsidy and tari reorm, particu-larly in a more dynamic and longer term ramework, these measures havethe potential to create adjustment costs or some commodity sectors andhouseholdsat least in the medium term. These potential costs may makeimplementing subsidy and tari reorm more dicult given the large shareo rural and low-income households in India. In contrast, the eects on the

economy and on households o increased eciency in agricultural marketingsuggest that this approach may be a valuable complement to subsidy andtari reorms because o the potential to help mitigate the adjustment costsassociated with those measures.

The modeling ramework used here does not permit ull analysis o theadjustment costs that might be incurred i Indias traditional marketingsystem were transormed into a more ecient sector. This transormationlikely would lead to losses in employment and income in some traditionalrms. But, there would be more, larger scale, vertically integrated, and moretechnologically advanced processing and marketing enterprises that wouldprovide new employment and investment opportunities. Concerns with these

adjustment costs are an important part o the current debate over whetherto permit oreign direct investment in multibrand retail marketing in India(Mukherjee and Patel, 2005). The results o this study indicate that improve-ments in agricultural marketing eciency that explicitly entail reduced laborand capital inputs in providing marketing services still put upward pressureon economywide returns to labor and capital. Although some participantsin the traditional marketing system will undoubtedly ace adjustment costs,the economywide employment and consumption impacts are shown to bepositive.


34/43



For the United States, these results suggest that the process o increasinginvestment and eciency in Indias agricultural markets could moderategrowth in Indias uture agricultural imports, but also act as an importantdriver o Indias economywide income and demand growth. The results othe analysis or various commodity sectors and households indicate whydomestic support and market access issues are sensitive or India in bilateraland multilateral negotiations. The results also suggest avenues or addressingthese concerns through cooperation and investment, which could contribute

to more ecient domestic marketing chains. Further, the gains in output andconsumption associated with introducing improvements in marketing e-ciency suggest the potential or returns to private investment in Indias agri-cultural markets.

Analytical issues that would be useul to address in uture work includeimproved data on wholesale and retail marketing costs in Indian ood andagriculture, not only by commodity, but also by purchaser. Recent sectorstudies o Indias agricultural markets conducted by the World Bank,USDAs Economic Research Service, and others indicate producer-to-retailmarketing costs that are substantially higher than those implied by the avail-able input-output data or the Indian economy used or this study. I this istrue, then the current study likely underestimates the potential economywidegains rom improved agricultural marketing eciency. In addition, as noted,the use o a dynamic ramework would better capture the potentially signi-cant, long-term growth implications o marketing eciency gains in Indianood and agriculture.


35/43

27


Reerences

Arndt, C., H.T. Jensen, S.Robinson, and F. Tarp. 1999.Marketing Marginsand Agricultural Technology in Mozambique, TMD Discussion PaperNo. 43, Trade and Macroeconomics Division, International Food PolicyResearch Institute, Washington, DC.

Chadha, R., and D. Pratap. 2006. Chapter 11-H: India, in Betina V.Dimaranan, editor, Global Trade, Assistance, and Production: The GTAP6 Data Base, Center or Global Trade Analysis, Purdue University.

Gandhi, V., and A.Koshy. 2007. Wheat Marketing In India: Systems,Efciency, and Integration, unpublished manuscript, Indian Institute oManagement, Ahmedabad, India, October.

Ganesh-Kumar, A., M. Panda, and M.E. Bursher. 2006.Reorms in IndianAgro-Processing and Agriculture Sectors in the Context o Unilateral

and Multilateral Trade Agreements, Working Paper WP-2006-011, IndiraGandhi Institute o Development Research, Mumbai.

Government o India, Ministry o Agriculture, Directorate o Economics andStatistics, 2008. Agricultural Statistics at a Glance 2008, http://dacnet.nic.in/eands/latest_2006.htm.

Government o India, Ministry o Agriculture, Directorate o Economicsand Statistics. 2003.Report o the Committee on Capital Formationin Agriculture, March, http://agricoop.nic.in/Capital%20Formation/FinalReport.doc.

Government o India, Ministry o Finance. 2008. Economic Survey, 2007/08.

Government o India, Ministry o Statistics and Programme Implementation.2005. National Sample Survey, 61st Round, http://mospi.gov.in/mospi_nsso_rept_pubn.htm.

Government o India, Ministry o Statistics and Programme Implementation.2008. National Sample Survey, 62nd Round, January, http://mospi.nic.in/rept%20_%20pubn/test.asp?rept_id=523&type=NSSO.

Government o India, Ministry o Statistics and Programme Implementation,Central Statistical Organization. 2000.Input-Output Transactions Table,1993-94, New Delhi.

Government o India, Ministry o Statistics and Programme Implementation,Central Statistical Organization. 2007.National Accounts StatisticsSources and Methods, New Delhi.

Government o India, Ministry o Statistics and Programme Implementation,National Sample Survey Organization. 2001. Consumption o SomeImportant Commodities in India, 1999-2000, Report No. 461, New Delhi.

Gulati, A., and S. Narayanan. 2003. The Subsidy Syndrome in IndianAgriculture, New Delhi: Oxord University Press.


36/43



Jha, S., P.V. Srinivasan, and M. Landes. 2007.Indian Wheat and Rice SectorPolicies and the Implications o Reorm, Economic Research Report No.41, U.S. Department o Agriculture, Economic Research Service, May.

Joseph, M., N. Soundararajan, M. Gupta, and S. Sahu. 2008.Impact oOrganized Retailing on the Unorganized Sector, Indian Council orResearch on International Economic Relations, New Delhi

Landes, M. 2008. The Environment or Agricultural and AgribusinessInvestment in India, Economic Inormation Bulletin No. 37, U.S.Department o Agriculture, Economic Research Service, July.

Landes, M., and A. Gulati. 2004. Farm Sector Perormance and ReormAgenda,Economic and Political Weekly (Mumbai) 39(32):3611-19,August 7-13.

Landes, M., S. Persaud, and J. Dyck. 2004.Indias Poultry Sector:Development and Prospects, Outlook Report No. WRS04-03, U.S.Department o Agriculture, Economic Research Service, February.

Lgren, H., R.L. Harris,

Documents

USDA Agri India