Presentation made to Ministry of Agriculture, Government of India

1
Towards Improving Understanding of Agricultural
Markets in India

• Presentation made to Ministry of Agriculture,
  Government of India
• May 24th 2011

2
Objective

• Prices Price Behaviour, Volatility Price
  Transmission Farm gate ?Wholesale ? Retail,
  Domestic ? International, Thinness of
  international market
• Consumption Monetization of economy, Changes in
  diet, calorie and nutritional content
• Output Composition of Output, Yield, Efficiency
  (NFSM), Ground Truthing
• Data Issues and Taxonomy for Agriculture
  Statistics
• Policy Responses to Volatility and Mitigation
  Short, Medium and Long Run Supply and Demand
  Side, Reliance on market mechanism, Trade policy

3
Characterising Price Behavior

• Cyclic pattern i.e. swings from trough to peak
• Right skewness Upward spikes not matched by
  similar price decline
• Excess kurtosis Tails of the price distribution
  fatter than the normal
• Time varying volatility Unstable variance across
  time
• Stochastic trend Random movements across an
  average price
• Positive autocorrelation due to storing of
  commodities from the harvest to post-harvest
  season

4
(No Transcript)
5
(No Transcript)
6
(No Transcript)
7
(No Transcript)
8
(No Transcript)
9
Characterising Price Behavior

• Variation of prices over intra harvest period and
  distribution of prices across regions
• Volatility - by type of commodity
  (characteristics)
• Seasonal
• Storage, Warehouse Receipts
• Futures Market (Why markets function well in some
  commodities)
• Market Integration
• Objective 1 Identify Cycles, Temporary /
  Permanent Shocks, Structural Factors - Group
  commodities based on their price behavior

Literature
10
Price Transmission Farm gate ?Wholesale ?
Retail

• Imperfect price transmission
• Incomplete transmission
• Lags in price adjustment between respective
  stages in the marketing chain
• Asymmetric responses to positive and negative
  price changes
• Market structure matters - Number of market
  intermediaries, Differences in volumes transacted
  in Mandis, Market integration

11
Price Transmission Farm gate ?Wholesale ?
Retail

• Asymmetry in transmission from wholesale to
  retail Increase in the wholesale prices is
  passed on quickly (no. of days) to consumers as
  compared to a decrease
• Size and speed of transmission is crucial from
  policy perspective
• Frequency of price change and quantum of price
  change
• Objective 2 Understanding transmission mechanism

Literature
12
(No Transcript)
13
(No Transcript)
14
Price Transmission Domestic ? International

• Objective 3 To quantify thinness of market and
  understand its implications for price
  transmission and trade policy
• International trade in agricultural commodities
  by country commodity pairs. What is the duration
  and volume of trade by country commodity pair?
• Composition of Indias (exports and imports)
  trade basket role of tariffs
• Studying impacts (such as trade, welfare and
  revenue effects) associated with alternative
  trade policy scenarios can be analyzed using the
  SMART model

15
(No Transcript)
16
(No Transcript)
17
Consumption

• Looking at
• Trends in Consumption
• Producer Monetization of the rural economy,
  vulnerable to price volatility
• Consumer Prices, Affordability, Dietary habits

18
Consumption

• Inferring food security from consumption patterns
  (NSSO Data) Calorific content of Indias
  agricultural output, Food Security Macro ?
  Household (?) (Markets, Prices, Affordability)
• Technology fortification
• Changing patterns in domestic consumption
• Objective 4 Secondary data analysis using NSSO
  data and NNMB data to understand linkages between
  household occupation, poverty and nutritional
  value of consumption basket

19
Output Composition and Yields

• Trends in Production and Yields
• Food grains (buffer stocks, PDS) Non Food
  grains
• Vegetables
• Fruits
• The G-20 document talks about Rice, Wheat, Maize
  such a focus is very narrow
• Objective 5 The debate on food-non food crop
  production and its impact on prices Relevance
  to India (CGE model)

20
Improving Yields

• Bridging Yield Gap - Ensure access to existing
  technology, gaps in access use of technology
  across agro climatic regions, developing new
  technology (shifts the production frontier -
  collaborate with agricultural scientists)
• Improving Yields - Investments in agriculture and
  Investments for agriculture
• Objective 6 Use the unit level cost of
  cultivation (input and output) data to understand
  the extent to which farmers are away from the
  production frontier and quantify the yield gap

21
Investments in / for Agriculture

• Objective 7 Analysis of state governments
  capital expenditure (1991-2010) on agriculture
• Objective 8 Quality of investments in and
  investments for agriculture Examine specific
  schemes for example RIDF (NABARD)
• Objective 9 Output elasticity of agricultural
  credit
• Objective 10 Relate the yield gap to quality of
  investments in and investments for agriculture
  subject to availability of data on yield gap
• Objective 11 Review paper on role of markets
  institutions

22
Factors contributing to increase in food
production

• Objective 12 Pilot Study of impact of NFSM
• Role of ATMA
• Convergence with other programmes
• Identify a cluster of villages in one or two
  districts where production has increased
• Conduct survey to identify factors

23
Validation of Crop Forecasts

• Objective 13 Ground Truthing Exercise (V.C)
• Developments in remote sensing techniques have
  enabled generation of contemporaneous estimates
  of crop area and production
• Identify a district covered under National Food
  Security Mission for ground truthing the
  estimates from remote sensing data
• IGIDR needs to collaborate with ISRO \ National
  Remote Sensing Agency in this regard

24
Online Digital Portal

• Objective 14 Online Portal
• IGIDR will collaborate with IRIS Knowledge
  Foundation (IRIS-KF) to build the online portal
  on aspects related to agriculture
• IRIS-KF has developed eSocialSciences, an online
  social science portal and Knowledge Community on
  Children in India for UNICEF

25
Taxonomy for Agricultural Statistics

• Rangarajan Committee Recommended use of XBRL.
  XBRL is now the mandated reporting standard for
  banks and companies in India
• This reporting standard can be extended to
  socio-economics statistics including agricultural
  statistics
• In order to work towards adoption of XBRL
  standard it is important to develop a taxonomy
• Objective 15 Review paper on improving
  agricultural statistics the role of XBRL

26
Deliverables

• Develop a comprehensive interlinked database
• Taxonomy for Agricultural Statistics
• Knowledge Briefs - 4 Pager
• Discussion Papers
• Analytical Papers
• Dissemination

27
Issues for Discussion

• Access to price database (available online and
  internally within the ministries), cost of
  cultivation data, NNMB data
• Collaboration Identify partner institutions
• Budget to be finalized upon clarity on scope and
  duration of the project

28
Recap of Objectives

• Group commodities based on their price behavior
• Domestic price transmission mechanism
• Frequency of price change
• Thinness of international market and its
  implications
• Linkages between household occupation, poverty
  and nutritional value of consumption basket
• Debate on food-non food crop production and its
  impact on prices
• The extent to which farmers are away from the
  production frontier and quantify the yield gap
• State governments capex on agriculture

29
Recap of Objectives

1. Quality of investments in and investments for
   agriculture
2. Output elasticity of credit
3. Relate the yield gap to quality of investments in
   and investments for agriculture
4. Review paper on role of markets institutions
5. Pilot study on National Food Security Mission
6. Ground Truthing Exercise
7. Online Portal
8. Taxonomy for Agricultural Statistics

30
(No Transcript)
31
Trend

• Deflated prices of wheat and corn show a downward
  trend from 1950 to 2001 (Wright, 2011)
• Trends in the price series studied through the
  unit root analysis (Cuddington, 1992 Ghoshray,
  2010)
• Long run trends are small as compared to the
  price variability (Cashin and McDermott, 2002)
• Decline in the trend is not gradual but takes
  place in installments i.e the price series
  have structural breaks (Zanias, 2005)

32
Cycles

• Real commodity prices are characterised by
  long-cycles (Hadri, 2010)
• Reasons for the cyclic pattern Low elasticity of
  demand and supply, Speculative bubbles
• Commodity price cycles are characterized by
  short-lived booms and sharp bursts (Deaton and
  Laroque, 1992 Deaton, 1992)
• The presence of cycles create booms and busts in
  GDP (exports) and hence the estimates of
  magnitude, duration and shape of the cycle are
  important from the policy perspective

33
Cycles contd...

• Cashin, McDermott and Scott (1999) date commodity
  prices using the Bry-Boschan business cycle
  algorithm and estimate the amplitude, duration
  and frequency of the cycle. They also examine
  whether the duration spent in either boom and
  slump affects the probability of a change in the
  state
• Price slump lasts longer than the booms (Cashin
  and McDermott, 2002)
• Labys et al. (2000) use the NBER (Moore, 1980)
  chronology to find the timing, frequency and
  amplitude of price cycle

34
Frequency of Price change

• It reflects how quickly prices adjust in response
  to changing demand and supply conditions
• Is the frequency of price change equal for both
  the wholesale and retail markets at a centre
• How synchronized is the frequency of price change
  across different markets in the country

35
Frequency of Price Change

• Frequency of price change implies the percentage
  of price quotes which changed values from their
  last month level
• Disaggregated CPI data has been used to
  understand the frequency of price change (Bils
  and Klenow (2004) Baharad and Eden(2004))
• The aim is to understand whether prices change in
  a staggered (State Dependent Pricing) or random
  (Time-Dependent Pricing) manner
• Hazard Functions are further used to measure the
  predictability of a price change

36
Frequency of Price Change and Hazard Function

• Hazard function represents the distribution of
  the length of time that elapses from the
  beginning of an event until its end (Ikeda and
  Nishioka, 2007)
• Hazard rate, an outcome of the hazard function
  predicts the chances of prices changing in the
  next period given that they have remained
  constant till the last period.