Speaker: Barbara Harriss-White Authors: Barbara Harriss-White, Alfy Gathorne Hardy. D. Narasimha Reddy, Deepak Mishra and R Hema Date: November 18, 2013 Venue: Cornell University, Ithaca New York Sponsors: SRI-Rice and International Programs (CALS), Cornell Subject: Climate Change, Materiality and Rice - A Research Project in Andhra Pradesh in India

1. CLIMATE CHANGE, MATERIALITY AND
RICE – A RESEARCH PROJECT - ideas,
methods and results
Barbara Harriss-White and Alfy
Gathorne Hardy with D. Narasimha
Reddy, Deepak Mishra and R Hema
1

2. ORIGINS
http://www2.lse.ac.uk/researchAndExpertise/units/mackinder/theHar
twellPaper/Home.aspx
• Thought-process of HARTWELL PAPER (2009) - but see
Dubash et al (June 2013 EPW) and Bidwai (2012)
• critique of structural flaws of policy for climate change
response:
• Fails test of history - global goal of emissions reduction not
possible; 2013 – energy majors retreat from renewables
• Cap and trade has major problems: carbon price s.t. political
control (overallocation of permits: lobbies/recession);
• Clean devt mechanism also struggling ((non)counterfactuals,
measurement, corruption + rent seeking.)
2

3. WHAT IS THE ALTERNATIVE?
• ‘Relentless pragmatism’ : decarbonifying the
econ desirable as a benefit in itself and also
contingent on co-benefits and other goals.
• Hartwell’s goals: ren en / en efficiency and access (reduce
waste) / respect for eco-systems / protn from risk.
• Dubash et al 2013: growth, inclusiveness, environmental
gains
• Policy instruments not confined to ‘markets’
but include bans, taxes - what about
technology? organisation of society?
3

4. CO BENEFITS APPROACH BY NGO/
POLITICAL PARTIES/ TRADE UNIONS/
EXPERT GROUPS IN THE UK

5. INDIAN DEBATE
Majority view:
• Stock argument : justice
• Flow argument : right to remove poverty (with 70% thermal)
VERSUS minority view / conclusion / starting point :
• FF based development is luxury India can’t afford
rate of degradation of natural resources
rate of addition to workforce (jobless growth)
So leapfrog to clean devt – low C transition - TO WHAT?
new development models / new indl revn AC+DC
5

6. IDEAS AND METHODS

7. OUR PROJECT
• Global responses to climate change neglect
informal economy / ILO’s and NCEUS’project
of decent work /standard work
• 90% jobs 60% GDP inf econ
• 43% workforce has half a year’s work
SO?
7

8. Develop methods to
1.think of the
economy as a system of capital and labour
using materials to produce commodities
and physical – solid, liquid and gaseous waste
2. measure parameters of informal
economy
8

9. Objectives - developing methods contd
• 3. to scope
technological
alternatives according
to several
incommensurable
criteria (environmental
/ social / economic)
which generate tradeoffs between criteria.
• 4. to mainstream social
relations of work –
quantity and quality of
labour

10. Big problems of separate knowledge
fields – at least a dozen
climate change – climate policy – life cycle assessment –
informal economy – value / supply chain analysis –
science and technology studies – policy studies – labour
studies – agriculture – rice - agricultural markets
For 1. and 2. – the materiality of the economy - fuse
life cycle assessment (LCA) with value / supply chain
analysis (VCA ) - politicise the frontier between formal
regulation/ policy and social regulation
For 3. the evaluation of incommensurable dimensions
of technological choices - use multi-criteria mapping
(MCM) and
For 4. the study of livelihoods and labour – start with
Decent Work / Standard Work (DW)
10

11. Innovative Learning Workshops
11

12. 1. LCA and 2. VCA
• LCA measures GHGs at all stages of a productionconsumption process from raw materials
procurement to the waste disposal at the end of
a consumption process
• VCA has a similar concept of a ladder/chain and
seeks to compute value added, rents and
potential for technological / managerial
upgrading
• OUR PROJECT SEEKS TO DEVELOP A FUSION
(AGH-Hema) that is sensitive to the political
space at the frontiers of formal regulation (AP)
12

13. 3.ALTERNATIVE TECHNOLOGY/ POLICY
• Research on energy and materials efficiency leads
to normative questions: how to scope and
analyse technologies / policies reducing
environmental impact in the informal economy
• LCA leads to MCM experiment (AGH-BHW)
• Mitigation technology options (4) assessed by
stakeholders (30-40) on incommensurable criteria
(min = 3 (costs, GHGs, and labour quantity and
quality))

14. 4. WORK AND WORK ALTERNATIVES
• Research on the quality of work and
production relations leads to nightmare of
indicators - 125 in the ILO’s Decent Work
• Trades Unions, frequently airbrushed out as
‘stakeholders’, are actively involved here –
NTUI – VCA -> labourist supply chain analysis
(MM-GM-MS)
• LCA develops new measures of work for trade
offs with GHGs/energy and costs (AGH)
14

15. CLIMATE CHANGE AND AGRICULTURE
– WHY AGRICULTURE?
• Agriculture innocent? Small proportion of global
GHG emissions? 10-14%
versus
• USDA (2011) estimate that entire food-system
production-distribution-(transport-processingstorage)-consumption-waste
PLUS land use change (burning forests for
agriculture) = +/- 33 - 45% global GHGs
AND agriculture as an emissions floor is ‘poorly
understood’ (Anderson, 2011)
15

16. CLIMATE CHANGE AND RICE
• Vulnerability – temperature, weeds and pests; rain and
crop failure -> yield declines expected -> critical impact
on food /feed prices and food security/ availability esp if
no change in access and utilisation (S Asian prodn/nutn
paradox)
• Agenda of adaptation in rice: crop-livestock stress
research; irrigation management; biotech innovations
(inc. GM and hybrids) ;collective action/farmers’ groups
(info – dissemination)
• -> funds in new revolutionary high-tech frontier rather
than farming systems; focus on adaptation rather than
mitigation
• Sources: Nelson et al, IFPRI, 2009; Alagh, 2013
16

17. RICE AS A CASE STUDY – HERE -1
• NOT because a big polluter [though food
system and all land-based activity thought to
account for up to 45% GHG]
BUT
• Rice is bio-physically complex – emitting
various GHGs as well as sequestering them –
so scientifically interesting
• Rice is socio-technically complex – 4
production systems 4-5 marketing systems –>
social scientific interest
17

18. RICE AS A CASE STUDY - 2
• Resources, employment and poverty are
entwined in production-distribution systems –
>policy interest
• Production and distribution weave in and out of
the informal economy –> theoretical and policy
interest
• Food is generally exempt from the scenarios
lowering emissions (Anderson/Royal Society
2011) i.e. something of a political special case –
but how special ?

19. LCA-VCA-MCM generate cross-cutting
themes
• Socio-political limits to the reach of state
regulative policy, the interface and forms of
regulation of the informal economy (A.P.)
• The means whereby UN-organised labour
makes gains in the informal economy (BHW/
NTUI/CWM)
• Innovation and technical change in the
informal economy (BHW/GR)
19

20. 20

21. 21

22. 22

23. 23

24. 24

25. 25

26. 26

27. 27

28. 28

30. 30

31. 31

32. 32

33. OBJECTIVES:
Through the perspective of different production
methods, measure variation in the entire rice supply
chain - production and distribution - to understand
where, how and why GHGs, labour and money interact.
Criteria:
physical : water, fossil energy, animal and human
energy, greenhouse gases
economic/social: costs and returns, quantity and
quality of labour

34. Sub-categories of the chain:
•
•
•
•
Four different production systems in 3 states
Different transport distances
Different milling types
2 different retail types + PDS
• extreme complexity…

36. A stylised model
Farm
Paddy
Transport
Paddy
Mill
Rice
Transport
Rice
Retail

37. Embodied water
Human and
animal energy
Fossil
energy
Electricity
energy
Embodied
energy
GHG
emissions
Embodied labour
Embodied human/animal energy
Farm
Labour
costs
Paddy
Capital
costs
Transport
Paddy
Loans and
loan
repayments
Mill
Embodied GHG
emissions
Rice
Ground
water
Transport
Quantity of
employment
Rice
Cost of carbon
Cost of
electricity
Retail
Quality of
employment
Cost of family
labour

38. RESULTS
38

39. The role of samples – treat the supply
small each stage in with
cautionfor each criteria
chain
Yes.
12
No. 11
10
Does not
9
include GHG
8
Ground water
management
7
Fossil energy
6
Total energy
5
Labour quantity
4
M labour quality
F labour quality
3
Costs
2
Profit
1
0
0
HYV farm Transport Large3Mill Small retail Large Retail
1
2
4
5

40. The role of each stage in the supply chain
for each criteria
Yes.
12
No.
11
10
9
GHG
8
Ground water
7
Fossil energy
6
Total energy
5
Labour quantity
M labour quality
4
F labour quality
3
Costs
2
Profit
1
0
0
1
HYV farm
2
3
4
5
Transport Large Mill Small retail Large Retail

41. PRODUCTION- variation in paddy
yields

42. EMISSIONS
ONSEMIE

43. GHG EMISSIONS / KG PADDY - No significant
differences at relatively shallow water tables
• THESE RESULTS ARE SIMULATED FOR A 7M WATER
TABLE

44. Approximate constituent emissions
1.4
kg GHG / kg paddy
1.2
1
0.8
N2O
0.6
CH4
0.4
CO2
0.2
0
HYV
Rainfed
SRI
Organic
• Methane from HYV, SRI and Organic from soils. Rainfed methane from draught
animals.
• Very little CO2 from rainfed as no irrigation and minimal use of tractors (some tractors
were used for manure transport in rainfed, but most cultivation and harvesting did
not use tractors/harvesters)

45. Activity related emissions –
conceptualisation of production through
activity
kg CO2 eq / kg paddy
1.5
1.3
1.0
0.8
0.5
0.3
0.0
-0.3

46. Activity related
emissions, HYV/intensive
kg CO2 eq / kg paddy
1.0
0.8
0.5
0.3
0.0
-0.3

47. Activity related emissions –
HYV/intensive
kg CO2 eq / kg paddy
1.0
0.8
0.5
0.3
0.0
-0.3
• Irrigation and soil derived methane = 87% of total.
Remember that this is at a quite shallow water table

48. kg CO2-eq /kg paddy
1.0
0.8
Constituent emissions SRI
(compared to HYV TN in blue)
HYV-TN
SRI -TNS
0.6
0.4
0.2
0.0
-0.2
• High methane emissions due to very high use of manure.
• Much lower irrigation based emissions.

49. kg CO2-eq /kg paddy
1.4
Constituent emissions organic
(compared to HYV TN in blue)
1.2
1.0
HYV-TN
Organic -TNS
0.8
0.6
0.4
0.2
0.0
-0.2
• Same pattern as HYV, except no synthetic
fertiliser emissions and more soil methane
due to more FYM.

50. kg CO2-eq /kg paddy
1.0
0.8
Constituent emissions rainfed
(compared to HYV TN in blue)
HYV-TN
Rainfed
0.6
0.4
0.2
0.0
-0.2
• Dominated by bullock based emissions and nitrous
oxide. If you do not include bullocks, then overall
emissions reduce to 50%

51. Nitrogen Use Efficiency
• SRI has low N efficiency due to the massive
over-supply of nitrogen. SRI uses more
synthetic N than HYV/ha, and x 2.5 the
amount of manure than HYV
HYV in TN
37%
Rainfed
45%
SRI
31%
51

52. WATER
• INTENSIVE / HYV in
both states use < 5 t /
kg paddy
• SRI used about 1.9 t / kg
/paddy
less than half

53. First problem – water measurement
• 1. Measure the field and model the field depth day to
day - very difficult
• HP of motor known, but pumping is hard to calculate:
– “How much electricity to you use?”
• Electricity is free, supply is highly erratic, no meters
• Pump use often constant when electricity is on, but
that varies considerably day by day.
• 2. SO Build profiles from farmers who keep watch on
well depth / water depth, hours of motor use, HP, and
efficiency loss factor
53

54. 2nd problem - water baseline
• We’ve used N TN water table as a ‘baseline’, at
7m (since that was the average level for
intensive (HYV)rice).
• In fact the water table for:
– Organic (also Tamil Nadu) was 12 m
– SRI (in Andhra Pradesh) was 27 m

55. The impact of simulating an AP water table
• With a 27m water table, SRI and rainfed are significantly
lower than organic and HYV

56. ENERGY

57. Diesel is not important
kg CO2 eq / kg paddy
1.0
0.8
0.5
0.3
0.0
-0.3
Including embodied C of
tractor/power tiller

58. Fossil energy
6.0
•
Energy (MJ/kg)
5.0
4.0
3.0
2.0
1.0
0.0
HYV
Rainfed
SRI
Organic
• HYV is higher than SRI due to more irrigation (although comparable
fertiliser use).
• HYV is higher than organic due to no synthetic fertiliser use in organic
(although comparable irrigation energy demand)

59. Fossil energy (HYV)
3.5
3.0
MJ kg paddy-1
2.5
2.0
1.5
1.0
0.5
0.0

60. All energy used in production / kg
(including animal and human)
Energy (MJ/kg)
25
20
15
10
5
0
HYV
Rainfed
SRI
Organic

61. Fossil Energy Return on Investment
350
300
EROI
250
200
150
100
50
0
HYV
Rainfed
SRI
Organic

62. LABOUR

63. For most systems labour demand is dominated
by transplanting and weeding

64. But rainfed labour requirements are higher
and different…

65. This difference is exaggerated by the low
yield of rainfed – 7.5 min /kg HYV paddy vs
50 min/kg rain-fed paddy
• Demonstrates the labour-displacing impact of technology – in this case
tractors and harvesters - - transplanters are displacing labour /
responding to labour shortages in some parts of TN.
• Weeding – still marked labour use in SRI, even with the use of weeders –
more weeds in moist semi-aerobic soil.

66. If we assuming (a) more labour is better and
(b) more pay is better,
is rainfed rice (RR) the best form of production from the
labour perspective?
(Is there an alternative? Is some work better than no work?)
• Per hectare little difference – per kg high
labour demand for RR
• Family labour matters for RR; pay = very low
for wage lb (Rs 100/d for men Rs 40/day for
women)
• RR provides Rs 7,740 of employment / ha,
compared to the max of Rs 24,802 / ha for
HYV TN.

67. Hours/ha
Overall gender balance –
intensification and feminisation
1000
900
800
700
600
500
400
300
200
100
0
• Women do absolutely more in
rainfed, mainly because of their
high input to weeding
• Women have lost out from
dominating weeding to sharing it
Female
equally with men in SRI, due to
shift to male push-weeding
machines.
Male
HYV Rainfed SRI
67

68. INTERACTIONS

69. Interaction of rainfed labour and GHGs

70. Interaction of HYV labour and GHGs

71. COST AND RETURNS - SIMULATIONS

72. IN PRODUCTION

73. 1. Increase wages to labour by 25%
HYV TN
Rainfed
SRI TNS
HYV AP
Organic
TNS
Cost
increase (%)
6
5
5
9
Profit
increase (%)
-10
-29
-3
-16
Actual profit
/kg
2.7
0.8
7.6
3.4
Actual profit
/ha
19,009
4,778
75,729
21,784
11
-8
6.8
31,463
All still profitable

74. 12,000
10,000
2. Charging for electricity
Rs/ha
8,000
6,000
4,000
2,000
0
HYV
Rainfed
SRI
SRI control
74

75. 2. SIMULATING ELECTRICITY CHARGES
• Charging for electricity substantially alters profitability of
farming
• 2Rs/unit) for electricity. Widely stated as reasonable price.
• -> An important increase in costs.
• Increases costs by 6% HYV in TN
• 12% SRI in AP
• 15% SRI control in AP
• Note : this simulation takes into account the water table
depth – even though SRI uses half the water it pumps from
greater depth.
• If electricity is costed at Rs 2/unit, all farms are still profitable
AT THE TN WATER TABLE LEVEL.

76. 3. Charging for GHG emissions
• Rs 0.74 / kg CO2-eq
• Taken from Certified Emission Reduction before carbon
price crash (ie €12.87 / t)
• The figures are identical for kg paddy and ha.
• Don’t include family labour costs, or electricity costs.
% change in
costs with C
price
% change in
profits with C
price
HYV TN
13
-25
Rainfed
12
-103
SRI TNS
18
-10
HYV TNS
25
-48
Organic TNS
19
-16

77. IN THE POST HARVEST MARKETING
SYSTEM / SUPPLY CHAIN

78. Lorry transport (750km)
Costs / kg
0.2
0.1
0
0.12
GHG emissions / kg / 750km
0.3
0.1
0.08
0.06
0.04
Diesel
Embodied
emissions
0.02
0
Labour: 2.8minutes/kg
Pay driver: Rs 600/day driver
Pay loader: Rs3.5 sack loading,
Rs3/sack unloading (75kg)
Profit: 0.01Rs/journey/kg
Charge for GHG, diesel increases by 5%.
 Some no longer make profits

79. Large mills

80. Large mills
Costs (Rs kg rice-1)
12.0
10.0
Costs
8.0
6.0
4.0
2.0
0.0
0.09
0.08
GHG
Increase labour rates by 25%,
profits decrease by 1.7%.
Charge for GHG emissions,
profits decrease by 1.4%
kg GHG / kg rice
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0.00
Electricity Biomass use Embodied
use
emissions
Diesel
generator

81. RETAIL – LARGE AND SMALL

82. RETAIL (field survey not complete yet)
ENERGY AND COSTS
GHGs

83. Storage/losses

84. Conclusions. Production
• Agricultural production GHGs, energy, water
dominate the supply chain – distribution is far
less important.
• GHG emissions are highly dependent upon the
water table. At shallow water table, there is no
significant difference in GHG emissions. At
deeper water tables, SRI and rainfed show
significant GHG savings.
• Labour use / ha is highest in SRI and Rain-fed rice

85. Conclusions. Production. 1. GHGs
GHGs
1. GHG emissions from three main sources –
methane, electricity, nitrous oxide.
2. Mitigate methane through reduced irrigation, and
appropriate use and timing of organic inputs (not more
than needed, not just before flooding)
3. Reduce embodied electricity GHGs through less water
and less technological transmission losses.
4. Reduce N2O emissions through appropriate use of
nitrogen – many farms putting far more on than
necessary. Reduce N2O emissions through total flooding
– but trade-off with methane

86. Conclusions. Production. 2: Water
SRI uses significantly less water per kg, but
not a significantly less compared to HYV on an
area basis.

87. Conclusion . Production 3: Fossil
Energy
• Irrigation dominates and synthetic fertilisers
also very important (mainly
nitrogen). Diesel not important (although
dominates rainfed as no ground water
irrigation or synthetic fertilisers). Embodied
energy in machinery is not important.
• Energy return on investment (EROI) (fossil
energy). Positive for all. V positive for
rainfed

88. Conclusions. Transport.
• Transport, even simulated for 750km, plays a
very small role in the overall chain.
• GHG emissions and costs are not well
matched.
• C price would have big impact on profitability
due to low margins

89. Conclusions. Mills
• Modern rice mills highly mechanised, so
capital a key component
• Biomass emissions likely to be more important
than normally recorded.
• Due to the progressive labour displacement
increasing wages to labour makes minimal
impact on profits
• High degree of variability in profit.

90. Conclusions. Retail.
• Large retail is surprisingly carbon intensive,
although sample size is low.
• Labour use / quantity is important in retail.
Labour quality (Y) appears to be relatively high in
large retail compared to the rest of the chain but
hours are long and pay in small retail is higher.
• Profitability appears to be lower in large rather
than small retail – deliberate loss leading strategy
• Energy use is substantial in large, but not small
retail.

91. Cautions
• Data results from a range of ways of knowing:
measurement, recall, calculation and extrapolation as careful as possible
• Alone they cannot be used as a prescriptive tool for
policy makers – sustainability is much more
complicated than just GHG emissions and labour use
(biodiversity loss; immediate water pollution etc)

92. UNIQUE SELLING POINT!
•
•
•
•
•
But our approach is a novel combination of
1. first hand fieldwork;
2. practical alternatives;
3. multidisciplinary analysis of criteria
4. the integration of measurement methods with a
consultation comparing expert and general public
evaluations of policy outcomes from the measurement
research.
• 5. multiple elements of sustainability from a whole
supply chain with many elements in the informal sector.
The research raises issues of trade-offs/synergies in a
novel manner and often for the first time

93. MORE ON OUR WEBSITE
• http://www.southasia.ox.ac.uk/resourcesgreenhouse-gases-technology-and-jobs-indiasinformal-economy-case-rice
93