Assessing natural capital risks and dependencies in lending to Australian wheat farms

Developing an evidence base for assessing natural capital risks and dependencies in lending to Australian wheat farms Abstract: Farmers are highly dependent on stocks of natural capital, and lenders are in turn exposed to natural capital through their loans to farmers. Key Words: Natural capital; environmental risk; environmental credit risk assessment; responsible lending; wheat farming; Australia Acknowledgements: This work was supported by the Australian Commonwealth Department of Infrastructure and Rural Development under the Sense-T “Sensing Natural Capital” project.

Developing an evidence base for assessing natural
capital risks and dependencies in lending to Australian
wheat farms
Abstract:
Farmers are highly dependent on stocks of natural capital, and lenders are in turn exposed to
natural capital through their loans to farmers. However, the traditional process for assessing a
farmer’s credit risk relies primarily on historical financial data. Banks’ consideration of
environmental factors tends to be limited to major risks such as contaminated land liabilities,
and to large project and corporate finance, as opposed to the smaller loans typical of the
Australian agricultural sector. The relevant risks and dependencies for agriculture vary by sub-
sector and geography, and there is a lack of standardized methodologies and evidence to
support risk assessment. We provide an evidence base to support natural capital risk
assessment for a single sub-sector of Australian agriculture – wheat farming. We show that
such an assessment is possible, with a combination of quantitative and qualitative inputs, but
the complexity and interconnectedness of natural capital processes is a challenge, particularly
for soil health.
Key Words:
Natural capital; environmental risk; environmental credit risk assessment; responsible lending;
wheat farming; Australia
Acknowledgements:
This work was supported by the Australian Commonwealth Department of Infrastructure and
Rural Development under the Sense-T “Sensing Natural Capital” project. We thank the project’s
industry partners, National Australia Bank (NAB) and the Climate Disclosure Standards Board
(CDSB) for their support.

2
1 Introduction
Agriculture is a small but important part of the Australian economy, accounting for 7% (by
value) of total exports in 2014-15 (ABS 2016a), and contributing as much as 12% of Australia’s
GDP, when including the value of associated pre- and post-farm production activities (NFF
2012). Farmers own or manage 61% of Australia’s land (NFF 2012) and therefore play a key role
in managing Australia’s stocks of natural capital, or “those elements of nature which either
directly provide benefits or underpin human wellbeing” (Natural Capital Committee 2013, 11).
Farmers all over the world are highly dependent on environmental factors such as rainfall,
temperature and climate, as well as natural capital stocks such as energy, mineral and water
resources, productive soils and ecosystems. Australia’s aridity (70% of mainland Australia
receives less than 500mm of rain per year – Geoscience Australia 2016) and generally poor soils
means that Australian agriculture is particularly vulnerable to variability and extreme
conditions. For example, widespread drought in 2002-03 led to a 41% reduction in agricultural
income, reducing Australia’s GDP by around 1% in that year (Lu and Hedley 2004).
99% of the 134,000 farm businesses operating in Australia in 2012 were family owned (NFF
2012), with on average 85% of business equity being held by the operator and close relatives or
business partners (ABARES 2016). The remaining 15% takes the form of debt, including short-
term unsecured debt such as credit cards; equipment supplier finance; and formal secured
loans, usually from one of the small number of Australian banks with extensive rural networks.
Through these loans, which are typically used either to fund land or machinery purchases, or to
provide working capital to bridge temporary income-cost mismatches, these banks are exposed
to a range of natural capital and environmental risks and dependencies. The traditional process
for assessing a farmer’s credit risk, however, relies primary on historical financial data, which
only indirectly and incompletely reflects these risks and dependencies.
In 2012, around 40 international financial institutions signed the Natural Capital Declaration
(NCD), committing to integrate natural capital considerations into their financial products, and
their accounting and reporting frameworks, by 2020. 1 A pilot study (Cojoianu et al. 2015) found
that 42% of financial institutions claim to be already integrating natural capital risks in credit
risk assessments. However, the evidence to date suggests that this is limited to large (over
US$10 million) project finance deals and even larger corporate loans, and is virtually non-
existent for smaller secured loans typical of the Australian agricultural sector. A key challenge
for agriculture is the fact that the relevant risks and dependencies vary by sub-sector and in
some cases by individual crop or animal production system, as well as by geography. Other
difficulties include lack of awareness around natural capital issues, the vagueness of regulatory
1 http://www.naturalcapitalfinancealliance.org/about-the-natural-capital-declaration/ (accessed 26 October 2016).

3
requirements around natural capital issues, the challenge of relating long-term issues to short-
term materiality, and most importantly, lack of standardized industry- and geography-specific
methodologies and robust information for the quantification of natural capital risks (Cojoianu
et al. 2015). Our paper therefore aims to help address these challenges by developing an
evidence base to support natural capital credit risk assessment for a single sub-sector of
Australian agriculture – wheat farming, which contributed over A$7 billion or 13% of Australia’s
total value of agricultural production in 2015, second only to cattle grazing (ABS 2016b).
The paper is structured as follows. In section 2, we review the literature on environmental
credit risk assessment. Section 3 outlines our methodology for identifying key environmental
and natural capital risks and dependencies, and section 4 sets out our findings for the
Australian wheat sector, explaining why each risk is material, how it can be mitigated, and how
a bank could assess and monitor the risk. Finally, section 5 discusses the overall findings that
emerged from this analysis.
2 Environmental credit risk assessment
The literature on sustainability in the banking sector can be divided into two broad groups
(Zeidan et al., 2015): one dealing with external practices (which include banks’ communications
with shareholders and stakeholders around sustainability issues) and the other with internal
practices (which include how sustainability issues are incorporated into risk management
models and lending or investment decisions). Our paper contributes to the latter.
Within this strand of the literature there are two main areas of focus. The first considers how
environmental performance influences the perceived credit risk and cost of debt of companies,
with most studies finding that firms with better sustainability performance ratings experience a
lower cost of debt, and vice versa (Bauer & Hann 2010; Nandy & Lodh 2012). Most studies focus
on public companies in developed countries (although Weber, Hoque, and Ayub Islam (2015)
provide evidence linking sustainability criteria to loan default probability in a developing
country, while noting that sustainability criteria are not yet being used by lenders there), and
on major risks such as hazardous chemicals, substantial emissions and climate change
(Schneider 2011; Chava 2014). Most papers also focus on large-scale corporate loans and bond
finance, rather than smaller loans to privately held companies and SMEs, where data on
sustainability as well as financial issues is much scarcer and opaque (Berger and Udell 1998).
There is, to the best of our knowledge, no peer-reviewed research demonstrating a clear link
between environmental performance at this smaller scale, and the availability, terms or cost of
bank loans. Nevertheless, from our discussions with Australian banking officials it appears that
at least some banks are operating on the assumption that there is such a link:

4
“I guess, you know, there’s an assumption that we’ve made, that seems to be validated
by numerous case studies, that our more productive agribusiness customers are those
who manage their environmental resources better, and they seem to be one and the
same thing.” (Environmental finance professional, August 6, 2015).
The second area concerns the extent to which banks incorporate sustainability criteria in their
lending decision-making. It is generally agreed that banks have been incorporating some
environmental risks into their credit assessment processes, since the early 1990s (Weber,
Fenchel, & Scholz 2008). This was originally driven by legislation, such as the Comprehensive
Environmental Response, Compensation and Liability Act 1980 in the United States, which
imposed remediation liabilities on the owners of contaminated sites. As a result, early
environmental credit risk assessments tended to focus only on such specific liabilities, rather
than providing a comprehensive framework for the assessment of all potentially material
environmental (and social) risks, dependencies and opportunities. In the mid-2000s, Weber,
Fenchel, & Scholz (2008) identified only four classes of environmental risk being considered by
banks: contamination liabilities; impacts of mandatory environmental regulations; changes in
buyer/consumer attitudes; and reputational risk due to being associated with projects seen as
environmentally or socially damaging by stakeholders. Furthermore, they found that most
attention was given to environmental issues in the rating (or risk identification) phase of the
credit management process, least in the costing (or risk evaluation) phase and an intermediate
level in the monitoring (or risk controlling) phases.
A recent survey of 36 financial institutions and 26 financial research providers undertaken for
the NCD (Cojoianu et al. 2015) found that although 42% of respondents claim to consider
natural capital risks in their credit risk assessment, there is no evidence that this is done
systematically. Lenders cited numerous difficulties in assessing natural capital risks including
the lack of suitable contextual methodologies, data, budgets and capacity. Currently, only
project finance transactions and related services can be benchmarked to an international
standard of environmental and social due diligence: the Equator Principles, which are applied
only to projects over US$10 million, or US$100 million for corporate loans.
The aforementioned body of research on sustainability performance and perceived credit risk
of borrowers goes some way toward explaining why banks might seek to incorporate
environmental risks and dependencies in their credit assessment processes. As Weber, Hoque,
and Ayub Islam (2015, 3) observe, “The rationale for using sustainability indicators for
predicting credit risks is based on the idea that good sustainability performance mitigates risks
arising through environmental and social impacts as well as through and stakeholder pressure
and regulation.” However, while high-level sustainability indicators may be available for larger
companies, usually based on research providers’ analysis of annual and sustainability reports,

5
they are typically not applicable to smaller-scale loans such as those typical in the Australian
agricultural sector.
Very little research has been published on the incorporation of environmental risks and
dependencies into credit assessment for smaller-scale loans in any sector, or for any scale of
loans in agriculture in particular. A search across all academic databases subscribed to by the
University of Oxford using the keywords ‘environmental’, ‘credit’, ‘risk’ and ‘agriculture’ yielded
only 80 results, only a few of which were found to be relevant. For example, Georgopoulou et
al. (2015) develop a framework to assess the risk of climate change to lending across several
Greek industries, including wheat farming, using the loan portfolio of a Greek bank as a case
study. The authors use climate and agronomic modelling to make the case for materiality of
climate change to the agriculture sector, and conclude that the physical risks from climate
change to Greek crop farms are between 7.4-12.4% of their annual turnover. However, they do
not provide an evidence base for the assessment of more granular risk factors (e.g. rainfall
availability), but rather rely on a generic crop yield simulation model. The emphasis is on
calculation of portfolio-level exposure to systemic risks, as opposed to individual loan
assessment. Similarly, Do et al. (2016) find evidence that banks are considering drought risk at a
systemic level, increasing loan interest rates to large agricultural enterprises by up to 6 basis
points for every step increase in regional drought level in the 12 months prior to loan
origination. Zeidan et al. (2015) propose a sustainability credit score system for the sugar
industry in Brazil, based on six dimensions: economic growth, environmental protection, social
progress, socio-economic development, eco-efficiency and socio-environmental development.
Whilst taking a higher-level sustainability focus, this study is the most complementary to our
own in terms providing a basis for the evaluation of such risks for a specific sector and country.
In summary, there is an almost complete absence in the literature of detailed assessments of
how environmental and natural capital risks can be identified and included in the credit risk
assessment process for smaller-scale (e.g. US$0.5-2 million) bank loans. This is true across
industries, and for agriculture in particular. Our paper complements the top-down approach
taken by Zeidan et al. (2015) by developing a bottom-up, robust evidence base on why and how
different natural capital risks and dependencies are likely to affect the financial performance of
wheat farms in Australia, how these risks can be managed, and how the resultant managed risk
can be assessed by a lender.
3 Methodology
We followed a multi-stage approach to identifying, prioritizing and researching the natural
capital risks and dependencies applicable to Australian wheat farm lending, from a lender’s
perspective. First, we searched the publicly available environmental risk methodologies of 66
financial research providers and consultants identified by Cojoianu et al. (2015) for

6
environmental risk factors specifically relevant to the agricultural sector. As we found that
coverage of agriculture was limited to just nine research providers, we also conducted a desk
review of 34 publications from financial institutions related to environmental risks, published
between 2006 and 2016, to identify high level environmental risk factors which were
considered to be material across a broad range of industries. Those most applicable to
agriculture are listed in Table 1 below.
The next step was to determine which of these risks were of most importance for wheat
farming, in the Australian context in particular. To do this we triangulated evidence from three
different sources: a review of 39 relevant academic papers shortlisted from a set of 974 papers
containing the keywords ‘wheat’, ‘Australia’ and ‘yield’ in the Scopus database; a review of
online publications from Australian industry-specific bodies (such as the Grains Research and
Development Corporation, GRDC) and relevant government agencies; and a set of four
interviews with Australian agribusiness professionals, credit managers and environmental
finance professionals. Typically, industry body publications proved the most useful in
demonstrating the materiality of specific environmental risks to the wheat industry, given their
focus and access to information specific to the Australian context.
A risk was considered material if it was clearly capable of being a significant determinant of
either yields, prices or costs, either in the short- or long-term. It was not practicable to set
quantitative thresholds for significance, but where possible we have provided quantitative
evidence in support of our judgment. We consider historical average conditions to be the
baseline: a risk is therefore only significant if it results in a deviation (in a negative direction)
from this baseline, either in the short- or long-term. It should be noted that risk is the product
of the probability of occurrence and the magnitude of impact, and it is therefore appropriate to
include some factors which are low impact but high probability, and vice versa. In addition,
some factors have been included on the basis of externalities which are currently not fully
priced, but which may become more appropriately priced in the medium to long term.
Finally, to keep this paper concise, we excluded weather, climate, energy and air emissions
risks, as these are already relatively well understood. The results of these steps are summarized
in Table 1 below. Factors relevant to wheat farming are highlighted in gray, with light gray
indicating the factors that are considered material, but not included in the following discussion.
Table 1: Agriculture – Environmental and Natural Capital Risk Factors
Thematic Area Risk Factor Sub-Risk Factor
Water
Water Availability Growing Season Average Rainfall (Lower Quantity or Increased Variability
Excessive Rainfall
Water Use Efficiency Crop Water Use Efficiency
Pasture Water Use Efficiency

7
Animal Stock Watering Efficiency
Farm Water Use Efficiency
Water Rights Quantity
Price
Irrigation Technology Failure
Water Quality
Salinity Management
Water Contamination
Water Acidity
Weather and
Climate
Heat Stress
Frost Damage
Extreme Weather Events
Land Use and
Pollution
Fertiliser Use
Quantity
Price
Run-Off
Appropriate Application
Soil Organic Carbon
Soil Erosion Water Erosion
Wind Erosion
Soil Acidification
Soil Salinity
Soil Compaction
Organic Contaminants
Heavy Metals
Waste Management
Biodiversity and
Ecosystems
GMO Contamination
Pests and Diseases
Weeds
Biodiversity
Livestock Health/Animal Welfare
Pasture Cover and Composition
Energy Use On-Farm Energy Use
Fertiliser Production and Transport Energy Use
Air Emissions
Carbon Dioxide
Energy Use Emissions
Fertiliser Production and Transport Energy Emissions
Direct Land Use Change Emissions
Methane Enteric Fermentation Emissions
Manure Management Emissions
Nitrous Oxide Fertiliser Application Emissions
F-gases (Hydrofluorocarbons, Perfluorocarbons and Sulfur Hexafluoride)
The same sources of evidence were then used to explore each of these risks in further detail. In
the following section, for each key risk, we discuss why it is material, what mitigation options
are available, and what information a lender would require to assess and monitor these risks,

8
either quantitatively or qualitatively. The key risks identified for wheat can, for ease of
discussion, be grouped into the following headings: water; fertiliser use; soil health; pests,
diseases and weeds, and biodiversity.
Hoepner et al. (2014) point out that banks can adopt two different approaches to assess the
credit-worthiness of a borrower. The first is the traditional approach to credit risk assessments
or ‘transaction-based banking’, which is based on quantitative and easily verifiable information,
predominantly financial metrics. The second is ‘relationship banking’, in which lending decisions
are based on intangible factors which are harder to quantify, often collected manually and
difficult to verify (e.g. managerial competence). The authors highlight a growing body of
literature showing that the use of intangible information in tandem with financial metrics can
lead to better credit risk assessments than those based only on financial factors. Hence we have
considered both quantitative and qualitative information that could contribute to assessing
each risk.
4 Natural capital risks and dependencies in Australian wheat farming
Most of Australia’s grain production is located on a narrow belt of land (in the east, south east
and south west of the country), known as the Wheatbelt or Grainbelt, which benefits from a
temperate climate, sufficient rainfall (on average 300-600mm/year) and relatively fertile soils
(Land Commodities 2012). At 46 million hectares, the Wheatbelt comprises 6% of Australia’s
total land area (Land Commodities 2012). Despite the name, farmers in the Wheatbelt also
raise livestock and grow other crops in addition to wheat. In 2014-15, wheat was grown on
about 57% (13.8 million ha) of total Wheatbelt farmland (24.3 million ha) (Farrell 2015).
The majority of Australia’s grain, including around 90% of wheat, is grown in a single winter
growing season, with precise sowing and harvesting periods varying by region (Land
Commodities 2012). There is minimal water available for irrigation across the Wheatbelt, hence
almost all grain in Australia is produced under a dryland cropping system (i.e. crops are rain-fed,
not irrigated). Rainwater availability and efficiency of use is therefore a key natural capital
risk/dependency for Australian wheat farming.
4.1 Water
4.1.1 Overview
The level of rainfall is the biggest predictor of agricultural productivity in a given year, and long-
term averages are a key determinant of land prices (Land Commodities 2012). Rainwater
availability for any crop is a function of four inter-related dimensions (Land Commodities 2012):
• Quantity (total annual average rainfall);

9
• Timing (in particular, the amount received during the growing season, and the amount
received in excessive downpours which can cause damage during the growing season or
at harvest time);
• Reliability (the variability of both quantity and timing from year to year); and
• Water use efficiency (the proportion of rainfall which becomes available to the crop
itself, which itself is a function of various factors including soil characteristics, drainage,
topography, timing of rainfall events, climatic conditions after rainfall, nitrogen supply,
weed cover and crop characteristics).
In a relatively small study (based on rainfall data from six weather stations in New South Wales
over six years) CelsiusPro AG (2010) found a linear relationship between wheat yields and
cumulative rainfall during the late growing season (1 st August to 31 st September), with rainfall
over this period explaining 90% of the annual variation in wheat yield. This suggests that every
millimeter less rainfall in the late growing season would result in a decrease in yield of 0.0188
t/ha (compared with average yields of around 1.5 t/ha). Assuming an average farm-gate price of
A$250-300/t, this would translate into a decrease in earnings of A$4.70-5.64/ha/mm. This is
significant in the context of typical profit margins of 10-20% or roughly A$37.50-90.00/ha based
on the same yield and price assumptions. 2 While further work is required to establish whether
these implications hold for other regions and time periods, it does suggest that the quantity of
rainfall during the late growing season is a critical component of water risk for wheat.
4.1.2 Risk mitigation
Farmers can mitigate the impacts of rainfall variability in several ways. Huda (1994) concludes
that yields can be protected by reducing the area sown when rainfall is low (less than 40mm) in
the early growing season (1st April -15th June) and increasing the area sown when rainfall is high
(greater than 100mm). Timing is also important: farmers should decrease planting when the
opening of the rainy season is delayed, and increase planting when the opening is early.
Choosing to grow a longer-duration wheat cultivar in early rainfall years, or increasing sheep
stocking as an alternative to wheat during low rainfall years is another option given accurate
and early forecasting capabilities of growing season rainfall. A variety of actions can be taken to
improve crop water use efficiency, for example by increasing the depth of rooting (Turner
2004). However, this strategy is most effective on sandy soils in high-rainfall zones where
nitrogen may leach into the root zone, with neutral and sometimes even negative effects in clay
soils which experience limited wetting to root depth (Smith and Harris 1981).
4.1.3 Lender information requirements
In the short term, lenders providing finance to wheat farmers can assess likely rainfall quantity,
timing and variability according to location, by comparison with historical rainfall data. In
2 Own analysis. Indicative prices and profit margins from personal communication, Agribusiness finance manager,
August 12, 2016.

10
Australia, Bureau of Meteorology data is likely to be sufficient, but could be supplemented with
farm-specific rainfall data, particularly for remote areas. In addition, lenders could qualitatively
determine whether the farmer has the necessary knowledge and track record to deal with
variable rainfall patterns and to employ appropriate risk mitigation measures as described
above and in Table 2. In the longer term, shifting rainfall patterns could change the risk profile
of wheat farms. Assessing this risk requires inputs from state-of-the-art climatic models
producing decadal and regional projections.
Table 2: Water risks, information needs and data sources
Risk Factor Timeframe Scale Information Need Data Sources
Growing Season Rainfall
Short term Region to
farm
Region- or farm-
specific critical growing
season dates
Bureau of Meteorology or farm-
specific rainfall datasets combined
with regional- or farm-specific yield
records
Short term Region to
farm
Region-specific
historical average
growing season rainfall
and variability (e.g.
over 10 years)
Bureau of Meteorology or farm-
specific rainfall datasets
Short term Farm
Farmer’s ability to
predict and deal with
variability
Farmer questionnaire
Long term Region
Projected changes in
growing season rainfall
patterns
Regional outputs from long-term
climatic models
Harvesting Season
Excessive Rainfall Short term Region to
farm
Farm exposure to
excessive rainfall
during growing season
or harvesting season
Farmer questionnaire on historical
crop damage in growing or
harvesting season.
Bureau of Meteorology or farm-
specific rainfall datasets.
Crop Water Use
Efficiency
Short term Farm
Recent historical
average water use
efficiency, compared
with the same data for
similar farms
Farmer yield data and Bureau of
Meteorology or farm-specific
rainfall datasets
Short term Farm
Farmer’s ability to
improve water use
efficiency
Farmer questionnaire

11
4.2 Fertiliser use
4.2.1 Overview
Fertiliser use has significant environmental impacts in several ways: it represents consumption
of various input natural resources and energy; it results in GHG emissions in both production
and consumption; and it has the potential to both positively and negatively impact other
natural resources, notably soil, water and biodiversity. It is also economically very important for
farmers: fertiliser use is estimated to add $12.7 billion in increased productivity to the
Australian agriculture sector (Ryan 2010). The purchase of fertiliser is generally the largest
single variable cost for grain producers, typically amounting to 15-20% of annual cash costs
(IPNI 2013). Thus nutrient use efficiency for nitrogen (N), phosphorus (P), potassium (K) and
sulphur (S) is an indicator of more general resource use efficiency (IPNI 2014a).
Having the right amount of nutrients in the soil when the crop needs it most is crucial for both
yield optimisation and long-term soil health. While fertiliser is a major contributor to crop yield
and health, mismanaging its application can result in significant risks for both the financial
wellbeing of the farm and the environment. Key fertiliser related risks include:
• Rising fertiliser input quantity and cost;
• Energy use and associated GHG emissions across the supply chain;
• Crop over- or under-fertilisation resulting in soil degradation; and
• Off-field run-off into adjacent watercourses, water tables or land.
Farmers have an incentive to manage fertiliser efficiently to reduce costs, but there may be
trade-offs between up-front costs of efficiency investments (such as smarter fertiliser
application systems) and the longer-term benefits of efficiency improvements, as well as
between positive short-term yield improvements and negative longer-term soil, water or GHG
impacts.
Two complementary indicators are particularly relevant for nutrient use efficiency. These are
(IPNI 2014b):
• Partial Nutrient Balance (PNB) which is measured as kg nutrient removed from soil / kg
applied and can provide an insight into whether soil is under- or over-fertilised; and
• Partial Factor Productivity (PFP), or the crop yield per unit of fertiliser applied (kg yield /
kg nutrient applied).
Benchmarks are available for these indicators: for example, IPNI (2016) recommend that, for
cereal crops harvested for grain, PNB (nitrogen) should be in the range of 0.1-0.9 kg/kg with a
PNB >1.0 kg/kg implying that the soil is being ‘mined’ for short-term productivity gain at the
expense of longer-term degradation, while PFP (nitrogen) should be 40-80 kg/kg. Similar

12
benchmarks can be established for phosphorus, potassium and sulphur, and a lender could
establish their own benchmarks across a portfolio of similar farms. However, differences in soil
type, crop variety, micro-climate and other factors may make it difficult to distinguish between
best and worst performers: for example, Mahjourimajd et al. (2016) find that different wheat
genotypes have a significantly different PFP in low rainfall environments in Australia.
4.2.2 Risk mitigation
While rising fertiliser input costs, production energy use and associated GHG emissions are
outside the control of the farmer, fertiliser application rates, timing and location can all be
influenced, thereby affecting the total cost of fertiliser to the farmer, consumption-phase GHG
emissions, and impacts on soil and water.
4.2.3 Lender information requirements
Lending institutions can monitor both the risk factors which are outside the farmer’s control
(fertiliser costs, fertiliser production energy use and GHG emissions) and those which are within
the farmer’s influence such as soil nutrient balance, fertiliser productivity, timing and
placement.
Table 3: Fertiliser use risks, information needs and data sources
Risk Factor Timeframe Scale Information Need Data Sources
Fertiliser quantity
and cost
Short term Farm
Fertiliser quantity (absolute or
kg/ha) and percentage of total
annual farm cash costs.
Farmer’s ability to balance
fertiliser purchasing decisions
with weather dependent fertiliser
requirements
Farmer questionnaire
Long term National Predicted fertiliser price trends Market analysts
Fertiliser production
and transport energy
use and GHG
emissions
Short and
long term National Energy and GHG emissions per kg
fertiliser
Life-cycle assessment studies,
fertiliser manufacturer disclosure
and/or farmer questionnaire
Appropriate
application of
fertiliser
Short term Farm Farmer’s ability to optimise
fertiliser application
Farmer and peer questionnaire
(peer with similar rainfall profile)
Long term Farm Impact of fertiliser application on
soil quality (e.g. monitoring PNB)
Soil samples (for nutrient
removal rate); farmer
questionnaire (for fertiliser
applied)
Fertiliser run-off Short and Farm and
surrounding
Adjacent water and soil
contamination from fertiliser run- Farmer questionnaire;

13
long term area off environmental regulator data
4.3 Soil health
4.3.1 Overview
Agricultural activity relies heavily on the underlying health of the soil. However, agricultural
activity itself can undermine soil health through changes in nutrient balance and alkalinity, or
through increased exposure to water and wind erosion, for example as a result of tillage, loss of
soil surface cover and removal of crop residues. Soil health related risks translate directly into
decreased agricultural productivity, long-term natural capital degradation and depreciation in
land value.
Sbrocchi et al. (2015, 37) argue that soil condition indicators would ideally be a comprehensive
“measure of the soil’s capacity to function, within natural or managed ecosystem boundaries,
to sustain plant and animal productivity, maintain or enhance water and air quality, and
support human health and habitation”. The multi-dimensional nature of soil health means that,
in an ideal world, at least seven different indicators would be monitored: acidification, soil
organic carbon, water erosion, wind erosion, salinity, nutrients, physical condition and
biological condition. Physical and biological condition is very difficult to monitor at the
extensive scales relevant to Australian wheat farming, and is therefore not considered further.
In the remainder of this section we discuss the first five indicators, with nutrient balance being
already considered in the section on fertiliser use.
Soil acidification is a slowly-occurring natural process which is accelerated by agriculture,
mainly due to excessive use of ammonium-based fertilisers, and partly because the product
removed (e.g. grains or other crops) is alkaline. Nitrogen in ammonium-based fertilisers is
readily converted to nitrate and hydrogen ions in the soil. If the nitrate is in excess of what can
be taken up by the plants it can leach away, causing a build-up of hydrogen ions, or acidity,
resulting in poor root growth and restricted access to water and nutrients.
It is estimated that more than 70% of surface soils and half of subsurface soils across the
Wheatbelt are affected by soil acidity (Wheatbelt NRM 2013), which results in up to
A$500m/year in lost production. 3 The optimal pH range for wheat is around or above pH 5.5 in
the topsoil and 4.8 in the subsurface – key thresholds below which root growth is impaired
(Gazey & Andrew 2009).
Soil organic carbon (SOC), and the ability of soil to store it, is regarded as an important basis for
soil fertility, and consequently, crop yields. It is also an important component of the global
3 https://www.agric.wa.gov.au/news/media-releases/world-soil-day-wa-soil-acidity-research-boosts-yields-and-
profitability (Accessed 27 October 2016)

14
carbon cycle, with potential to provide either substantial additions or removals to atmospheric
CO2 levels, depending on how it is managed. Across Australia, the total stock of organic carbon
in the top 30 cm of soil is estimated to be 19-32 GtC (Viscarra Rossel et al. 2014). In Western
Australian soils under broadacre grain production, SOC is typically 0.8-2% of soil mass in surface
soils (the top 10cm), or the equivalent of 8-20 tC/ha (Hoyle 2016).
Increased soil carbon levels can benefit crop growth through processes such as improved
nitrogen and water supply to crops. However, this is the case only at low fertiliser application
rates (0-50kg N/ha), when the crop relies on mineral nitrogen from soil organic matter rather
than nitrogen supplied by farmers (Palmer et al. 2015). At higher rates of fertiliser application,
which is more typical of current farming practices, there was little or no productivity benefit
from increased SOC.
Dryland salinity occurs when the concentration of soluble salts near the soil surface is sufficient
to reduce plant growth. Dryland salinity develops when a supply of water and a store of salt in
the soil meets the ground surface, mainly due to clearing for agriculture and the replacement of
deep-rooted perennial vegetation with shallower rooted annual crops. Up to 17 million
hectares of mostly agricultural land in Australia is thought to be at risk of developing salinity
problems by 2050 unless effective action is taken to mitigate this risk (ABS 2013). Salinity is also
thought to have caused a 50% decrease in the numbers of wetland bird species, and is
threatening 450 plant species with extinction (ABS 2013).
Water and wind erosion can impact wheat yields by removing fine fractions of soil which
include clay, organic matter and soil nutrients, resulting in reduced capacity of the soil to retain
water and nutrients. If the top 10 mm of soil are subjected to erosion, crop yields can be
reduced by 25% (DAFWA 2013).
4.3.2 Risk mitigation
Soil acidification risk can be mitigated by using acid-tolerant cultivars of wheat, in conjunction
with liming to recover soil pH to target levels (Gazey and Davies 2009). SOC loss can be slowed,
or even reversed, by optimizing agricultural management through stubble retention,
fertilisation and conservation tillage (Wang et al. 2013). The four major strategies available to
address salinity are the planting of more salt-tolerant crops, the planting of trees, the fencing of
land from grazing and the construction of earthworks such as drains (ABS 2003). Finally, water
erosion can be minimized through stubble retention and the construction of earthworks to
reduce the velocity and volume of peak flows, 4 and wind erosion can be controlled by reducing
wind-speed at ground level (e.g. by planting windbreaks) and minimising soil disturbance.
Around 50-75% ground cover is required to minimise erosion, depending on soil type (DAFWA
2013).
4 https://www.agric.wa.gov.au/water-erosion/water-erosion-introduction (Accessed 1 December 2016)

15
4.3.3 Lender information requirements
Lending institutions would be primarily interested in the farmer’s ability to manage soil health
issues, and in monitoring long-term trends in the above indicators for evidence of soil health
degradation.
Table 4: Soil health risks, information needs and data sources
Risk Factor Timeframe Scale Information Need Data Sources
Soil acidification
Short term Farm Farmer’s ability to monitor and
manage soil acidification
Farmer questionnaire
Long term Farm Change in soil pH over time Soil samples
Soil organic carbon
Short term Farm Farmer’s ability to monitor and
manage SOC Farmer questionnaire
Long term Farm
Change in SOC over time and
estimated carbon dioxide
equivalent emissions or removals
Soil samples plus
published conversion
factors
Soil salinity
Short term Farm Farmer’s ability to monitor and
manage salinity Farmer questionnaire
Long term Farm Change in salinity over time
APSoil database
(baseline) plus on-
going soil samples or
electromagnetic
mapping
Water and wind erosion
Short term Farm Farmer’s ability to monitor and
manage erosion Farmer questionnaire
Long term Farm Soil loss over time
Farmer observations
(e.g. photos); erosion
maps
4.4 Pests, diseases and weeds
4.4.1 Overview
Pests, diseases and weeds can be seen as ecosystem interactions that result in reduced crop
health and yields if farmers do not take appropriate action to mitigate their impacts. They can
also have significant impacts on other forms of natural capital such as biodiversity, soil or water
quality. The economic losses associated with pests, diseases and weeds in the Wheatbelt are
summarized in Table 5.

16
Table 5: Yearly economic loss related to pests, diseases and weeds in agriculture. Data sources: GRDC (2013) and Australian
Government (2016)
Pests Diseases Weeds
Economic loss A$789m/year
or A$60/ha
A$913m/year
or A$76.64/ha
A$1.5bn/year (weed control)
+ A$2.5bn/year (lost
agricultural production)
The risks to wheat crops from pests, diseases and weeds are material to not only short- and
long-term crop health, but also to the surrounding environment and its ecosystems.
4.4.2 Risk mitigation
Management of pests and diseases is achieved by the use of resistant cultivars, paddock
preparation and management, and application of pesticides (Murray and Brennan 2009). Weed
control relies on effective monitoring plus interventions such as winter cleaning of pastures or
chemical fallowing.
4.4.3 Lender information requirements
Lenders would be interested in the historical frequency and severity of pests, diseases and
weeds affecting a particular region or farm, the farmer’s ability to monitor and mitigate those
risks and projected changes in risk incidence over time.
Table 6: Pests, diseases and weeds risks, information needs and data sources
Risk factor Timeframe Scale Information Need Data Sources
Pests and
diseases
Short term Farm to region
Frequency and severity of pests and
diseases outbreaks, and farmer’s
ability to monitor and manage
Grain Research and
Development Corporation
(GRDC) reports and datasets
and farmer questionnaire
Long term Farm to region
and national
Projected changes in frequency of
occurrence of pests and diseases
outbreaks
GRDC reports and datasets
Weeds
Short term Farm to region
Time and resources spent on weed
management and farmer’s ability to
monitor and manage
Farmer questionnaire
Long term Farm to region
and national
Projected changes in distribution of
weeds GRDC reports and datasets

17
4.5 Biodiversity
4.5.1 Overview
Australia’s biodiversity is in decline, with more than 1,700 species and ecological communities
known to be at risk.5 The key threats to biodiversity include agricultural practices which lead to
loss, degradation or fragmentation of habitats, changes to water flows and quality, altered fire
regimes and the introduction of invasive pests, diseases and weeds.
Biodiversity has many different values. These include regulation of bee pollination,
recreational/amenity values, and cultural/spiritual values among others. However, only some of
these values are currently priced in some way (e.g. Australia’s commercial fisheries are valued
at $2.2 billion) (ABS 2010). Threatened species and ecological communities are protected in
Australia under the Environment Protection and Biodiversity Conservation Act 1999, which
controls actions likely to cause significant impacts.
4.5.2 Risk mitigation
A variety of actions can be taken to avoid negative impacts and/or to positively enhance
biodiversity. These include: mapping, conserving and regenerating remnant patches of native
vegetation; creating buffers and setting land aside for biodiversity; creating corridors between
areas of relatively undisturbed habitat; and controlling invasive species.
4.5.3 Lender information requirements
Lenders would be interested in understanding the baseline condition of biodiversity on the
farm, the projected long-term trend for biodiversity in the area, and the farmer’s ability to
monitor and manage biodiversity. Whether the farm is in a biodiversity hotspot or close to the
range of a threatened species could be indicators of high quality risk.
Table 7: Biodiversity risk - information needs and data sources
Risk factor Timeframe Scale Information Need Data Sources
Biodiversity
Short term Farm Extent of remnant native
vegetation (ha) Farmer questionnaire
Short term Region Quality of biodiversity
Biodiversity hotspots
map; Atlas of Living
Australia; region-specific
studies
Long term Region Projected changes in
biodiversity
CSIRO and other research
on climate change and
other long-term impacts
on biodiversity
5 http://www.environment.gov.au/biodiversity/threatened/species (accessed 23 March 2017).

18
Short term Farm
Farmer’s ability to
monitor and manage
biodiversity
Farmer questionnaire
5 Conclusions
Developing a detailed, context-specific understanding of potential environmental risks and
dependencies applicable to smaller-scale borrowers could not only help overcome the
inevitable information asymmetry between borrower and lender (Akerlof 1970), but also help
identify and price risks that the borrower themselves may not be aware of. A bank’s ability to
demand information from its customers (Goss & Roberts 2011) and to analyse this across a
portfolio in combination with other sources of information has the potential to detect systemic
opportunities or risks in any area of business, and may be particularly applicable to
sustainability risks, which are often long-term, large-scale processes, where systemic
understanding is still emergent.
There is a trade-off, however, between the cost of obtaining and analysing information, and the
associated benefit. It is no surprise that the principal environmental risks which research
providers currently analyse are those which can be relatively easily measured through the use
of publicly available datasets: greenhouse gas (GHG) and other air pollutant emissions, water
and climate risks, with only a few research providers offering services related to agricultural or
natural resource depletion risks. One reason for this may be that understanding such risks
requires the ability to analyse large volumes of spatial data, for example using Geographical
Information Systems (GIS), which traditional research providers typically do not use (Cojoianu
et al. 2015).
We have demonstrated that it is possible to derive a suitable evidence base to support a
comprehensive, industry- and geography-specific assessment of natural capital risks for smaller-
scale loans in agriculture, using wheat farming in Australia as a case study. In principle, this
suggests that the same generic approach could be followed to develop similar frameworks
suitable for other sub-sectors and geographies.
However, the exercise has not been without challenges. Natural processes are often
characterised by complexity and interconnectedness. While it is relatively easy to identify high-
level natural capital risk categories, we discovered that most of these are in fact multi-
dimensional: for example, rainwater availability is a function of quantity, timing, reliability and
water use efficiency, each of which can be associated with distinct risks. These sub-factors can
combine in various ways, each entailing different risks: for example, the risk of insufficient
water availability during the critical growing stages involves a different combination of quantity

19
and timing issues to the risk of excessive rainfall at harvest. The multi-dimensionality of
environmental risks can be observed at multiple levels and often involves cross-linkages with
other environmental risks: for example, water use efficiency is itself a function of various
factors including soil characteristics, nitrogen supply, weed cover and so forth. An assessment
framework based on dissection of multi-dimensional risks in order to evaluate individual risks
may overlook some systemic cross-linked issues.
A related challenge has been to bridge the gap between the financial sector’s desire for an
unambiguous answer to what seems a simple question – is this loan an acceptable risk? – and
the complexity, uncertainty and heterogeneity of the reality on the ground. We have had to use
our own judgement in some cases to distil what seem to be the key risk factors from a mass of
detailed, sometimes contradictory, scientific and industry statements. Our presentation of any
given risk may well be regarded as a gross over-simplification by an expert in that particular
area. Whilst the evidence base we have presented can undoubtedly be refined and improved, a
balance must be struck between exactitude and practicality. No metric will ever perfectly
capture any given risk – to be useful, it need only perform better than the current alternative,
which for many of these factors is no assessment whatsoever.
Some risks are less challenging to evaluate than others. For example, energy use and GHG
emissions are relatively easy to monitor and to price with a shadow cost of carbon, and
therefore we have not discussed these risks further in this paper. Soil health risk factors are at
the other end of the spectrum, being highly complex, interconnected, and long-term issues.
Finally, we have shown that many sources of useful quantitative data to assess natural capital
risk factors exist – at least in a developed country such as Australia – but also that qualitative
data will almost always also be required to obtain a view of how able the borrower is to
manage a particular risk. The ‘managed risk’ situation is what is most relevant to the lender, as
opposed to only the underlying physical risk on its own. Furthermore, the fact that most risks
can be managed proactively suggests that there is a role for natural capital risk assessment in
ongoing monitoring and farmer engagement, and not only at the initial credit assessment stage.
We have only been able to infer financial materiality at the level of the farmer from evidence of
the magnitude of physical risks and generic financial evidence such as sector-level costs.
Further research to evaluate the impact on farm financial performance of any individual risk
factor, or ideally all of them in combination, would be extremely useful.

20
References
ABARES. 2016. “Australian Farm Survey Results 2013–14 to 2015–16.” Canberra: Department of
Agriculture and Water Resources.
http://data.daff.gov.au/data/warehouse/9aas/FarmSurveyResults/2016/FarmSurveyResul
ts2016_V1.1.0.pdf.
ABS. 2003. “Salinity and Land Management on Western Australian Farms.” Canberra: Australian
Bureau of Statistics.
http://www.abs.gov.au/ausstats/abs@.nsf/0/f59529c371a21f55ca256db800783a4f/$FILE/
ATT34B9V/Salinity and land management on Western Australian farms_1.pdf.
———. 2010. “Year Book Australia 2009–10.” Canberra: Australian Bureau of Statistics.
http://www.abs.gov.au/ausstats/abs@.nsf/Previousproducts/1301.0Feature
Article12009–
10?opendocument&tabname=Summary&prodno=1301.0&issue=2009%9610&num=&view
=.
———. 2013. “Measures of Australia’s Progress - Salinity.” Canberra: Australian Bureau of
Statistics. http://www.abs.gov.au/ausstats/abs@.nsf/Lookup/by
Subject/1370.0~2010~Chapter~Salinity (6.2.4.4).
———. 2016a. “Characteristics of Australian Exporters, 2014-15 - Table 4: Value of Goods
Exports, by Industry of Exporter and Industry of Origin of Exported Commodity.” Canberra:
Australian Bureau of Statistics.
http://www.abs.gov.au/AUSSTATS/abs@.nsf/DetailsPage/5368.0.55.0062014-
15?OpenDocument.
———. 2016b. “Agricultural Commodities, Australia, 2014-15.” Canberra: Australian Bureau of
Statistics.
http://www.abs.gov.au/AUSSTATS/abs@.nsf/Lookup/7121.0Main+Features102014-
15?OpenDocument.
Akerlof, George A. 1970. “The Market for ‘Lemons’: Quality Uncertainty and the Market
Mechanism.” The Quarterly Journal of Economics 84 (3) (August): 488.
doi:10.2307/1879431. http://qje.oxfordjournals.org/lookup/doi/10.2307/1879431.
Bauer, Rob, and Daniel Hann. 2010. “Corporate Environmental Management and Credit Risk.”
ECCE Working Paper: 1–44. Available at SSRN: https://ssrn.com/abstract=1660470 or
http://dx.doi.org/10.2139/ssrn.1660470.
Berger, Allen N, and Gregory F Udell. 1998. “The Economics of Small Business Finance: The
Roles of Private Equity and Debt Markets in the Financial Growth Cycle.” Journal of

21
Banking & Finance 22 (6-8): 613–673. doi:10.1016/S0378-4266(98)00038-7.
http://www.sciencedirect.com/science/article/pii/S0378426698000387.
CelsiusPro AG. 2010. “Professional Weather Protection Weather Dependency of Wheat Yields in
NSW, Australia Hedging Wheat Yields with Weather Derivatives.” Zurich: CesiusPro AG.
http://www.celsiuspro.com/portals/0/downloads/White
Paper_Hedging_Wheat_Yields_with_Weather_Derivatives.pdf.
Chava, Sudheer. 2014. “Environmental Externalities and Cost of Capital.” Management Science
60 (9): 2223–2247. doi:10.1287/mnsc.2013.1863.
http://pubsonline.informs.org/doi/abs/10.1287/mnsc.2013.1863.
Cojoianu, Theodor Florian, Andreas G. F. Hoepner, Rupini Deepa Rajagopalan, and Damian S.
Borth. 2015. “NCD Scoping Study: Towards Including Natural Resource Risks in Cost of
Capital.” Oxford: Natural Capital Declaration.
http://www.naturalcapitaldeclaration.org/scoping-study-on-risk/.
DAFWA. 2013. “Report Card on Sustainable Natural Resource Use in Agriculture.” Western
Australia: Department of Agriculture and Food.
https://www.agric.wa.gov.au/sites/gateway/files/2.2 Wind erosion.pdf.
Do, Viet Minh, Thu-Ha Nguyen, Cameron Truong, and Tram Vu. 2016. “Drought Risk in Private
Debt Contracts.” Available at SSRN: https://ssrn.com/abstract=2888208.
Farrell, Roger. 2015. “Australia - Grain and Feed Update.” Canberra.
http://gain.fas.usda.gov/Recent GAIN Publications/Grain and Feed
Update_Canberra_Australia_1-27-2015.pdf.
Gazey, Chris, and Joel Andrew. 2009. “Soil pH in Northern and Southern Areas of the WA
Wheatbelt.” Department of Agriculture and Food, Bulletin 4761.
https://www.wheatbeltnrm.org.au/sites/default/files/knowledge_hub/documents/North_
and_south_pH_report_151208.pdf.
Gazey, Chris, and Stephen Davies. 2009. “Soil Acidity: A Giude for WA Farmers and
Consultants.” Western Australia: Department of Agriculture and Food, Bulletin 4784, 48p.
http://researchlibrary.agric.wa.gov.au/cgi/viewcontent.cgi?article=1031&context=bulletin
s.
Georgopoulou, Elena, Sebastian Mirasgedis, Yannis Sarafidis, Vassiliki Hontou, Nikos Gakis,
Dimitri Lalas, Foteini Xenoyianni, Nikos Kakavoulis, Dimitris Dimopoulos, and Vrassidas
Zavras. 2015. “A Methodological Framework and Tool for Assessing the Climate Change
Related Risks in the Banking Sector.” Journal of Environmental Planning and Management
58 (5) (May 4): 874–897. doi:10.1080/09640568.2014.899489.
http://www.tandfonline.com/doi/abs/10.1080/09640568.2014.899489.

22
Geoscience Australia. 2016. “Deserts.” Symondson. http://www.ga.gov.au/scientific-
topics/national-location-information/landforms/deserts.
Goss, Allen, and Gordon S. Roberts. 2011. “The Impact of Corporate Social Responsibility on the
Cost of Bank Loans.” Journal of Banking & Finance 35 (7): 1794–1810.
doi:10.1016/j.jbankfin.2010.12.002.
http://www.sciencedirect.com/science/article/pii/S0378426610004498.
GRDC. 2013. “The Current and Potential Costs of Invertebrate Pests in Grain Crops.” Grain
Research and Development Corporation.
https://grdc.com.au/~/media/916F66E1F2054DE5ACFC36ED959F9EB9.pdf.
Hoepner, Andreas, Ioannis Oikonomou, Bert Scholtens, and Michael Schröder. 2014. “The
Effects of Corporate and Country Sustainability Characteristics on the Cost of Debt: An
International Investigation.” ICMA Discussion Paper. Reading.
Hoyle, Frances. 2016. “Soil Organic Matter - Frequently Asked Questions (FAQS).” Department
of Primary Industries and Regional Development - Western Australia.
https://www.agric.wa.gov.au/soil-carbon/soil-organic-matter-frequently-asked-questions-
faqs.
Huda, A. 1994. “Management Strategies to Minimise Climatic Risk to Wheat Production in Low
Rainfall Areas of Southern Australia.” Agricultural and Forest Meteorology 69 (1-2) (June):
125–146. doi:10.1016/0168-1923(94)90085-X.
http://linkinghub.elsevier.com/retrieve/pii/016819239490085X.
IPNI. 2013. “Fertilizer Expenditure in the Australian Grains Industry.” International Plant
Nutrition Institute. http://anz.ipni.net/article/ANZ-3133.
———. 2014a. “Nutrient Use Efficiency on Grain Farms.” International Plant Nutrition Institute.
http://anz.ipni.net/article/ANZ-3246.
———. 2014b. “Nutrient Performance Indicators:The Importance of Farm Scale Assessments,
Linked to Soil Fertility, Productivity, Environmental Impact and the Adoption of Grower
Best Management Practices.” Georgia: International Plant Nutrition Institute.
http://anz.ipni.net/ipniweb/region/anz.nsf/0/9312A2172A0B917CCA257E8E007219B4/$FI
LE/Issue Review Perf Ind 081114.pdf.
———. 2016. “Decision-Making Guide Nitrogen.” Georgia, USA: International Plant Nutrition
Institute.
http://www.ipni.net/publication/4rguides.nsf/0/DF7D0BCDBF63D2D685257F72004B746F/
$FILE/4RGuide-Nitrogen.pdf.
Land Commodities. 2012. “The Australian Wheatbelt - An Introduction for Investors.” Baar:
Land Commodities. http://www.landcommodities.com/downloads/.

23
Lu, Lan, and David Hedley. 2004. “The Impact of the 2002-03 Drought on the Economy and
Agricultural Employment.” Australian Government.
http://archive.treasury.gov.au/documents/817/HTML/docshell.asp?URL=03_article_2.asp.
Mahjourimajd, S, H Kuchel, P Langridge, and M Okamoto. 2016. “Evaluation of Australian
Wheat Genotypes for Response to Variable Nitrogen Application.” Plant and Soil 399 (1-2):
247–255. doi:10.1007/s11104-015-2694-z.
https://www.scopus.com/inward/record.uri?eid=2-s2.0-
84956691752&partnerID=40&md5=9f78a10cbfdfbfb90593db5ca8d6dba4.
Murray, Gordon M.;, and John P. Brennan. 2009. “The Current and Potential Costs from
Diseases of Wheat in Australia.” Grain Research and Development Corporation.
https://grdc.com.au/~/media/B4063ED6F63C4A968B3D7601E9E3FA38.pdf.
Natural Capital Committee. 2013. “The State of Natural Capital: Towards a Framework for
Measurement and Valuation.” London, UK: Department for the Environment, Food and
Rural Affairs.
NFF. 2012. “Farm Facts.” Canberra: National Farmers’ Federation. http://www.nff.org.au/farm-
facts.html.
Palmer, Jeda, Peter Thorburn, Jody Biggs, Merv Probert, Neil Huth, and Joshua Larsen. 2015.
“Quantifying the Effect of Soil Organic Carbon on Wheat Yield: A Simulation Study.”
Proceedings of the 17th ASA Conference, 20 – 24 September 2015, Hobart, Australia.: 20–
24.
Ryan, Terry. 2010. “The Australian Fertilizer Industry - Values and Issues.” Newcastle, Australia:
Meta Economics Consulting Group. https://www.iama.org.au/sites/default/files/Australian
Fertilizer Industry Value and Issues August 2010.pdf.
Sbrocchi, C, R Davis, M Grundy, R Harding, T Hillman, R Mount, H Possingham, et al. 2015.
“Accounting for Nature Australian Regional Environmental Accounts Trial.” Sydney:
Accounting for Nature Australian Regional Environmental Accounts Trial.
Schneider, TE. 2011. “Is Environmental Performance a Determinant of Bond Pricing? Evidence
from the US Pulp and Paper and Chemical Industries*.” Contemporary Accounting
Research. http://onlinelibrary.wiley.com/doi/10.1111/j.1911-3846.2010.01064.x/full.
Smith, Richard C G, and Hazel C Harris. 1981. “Environmental Resources and Restraints to
Agricultural Production in a Mediterranean-Type Environment.” Plant and Soil 58 (1): 31–
57. doi:10.1007/BF02180048. http://dx.doi.org/10.1007/BF02180048.
Turner, NC. 2004. “Agronomic Options for Improving Rainfall-Use Efficiency of Crops in Dryland
Farming Systems.” Journal of Experimental Botany 55 (407): 2413–2425.
http://jxb.oxfordjournals.org/content/55/407/2413.short.

24
Viscarra Rossel, Raphael A., Richard Webster, Elisabeth N. Bui, and Jeff A. Baldock. 2014.
“Baseline Map of Organic Carbon in Australian Soil to Support National Carbon Accounting
and Monitoring under Climate Change.” Global Change Biology 20 (9): 2953–2970.
doi:10.1111/gcb.12569.
Wang, G, Y Huang, E Wang, Y Yu, and W Zhang. 2013. “Modeling Soil Organic Carbon Change
across Australian Wheat Growing Areas, 1960–2010.” PloS One.
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0063324.
Weber, Olaf, Marcus Fenchel, and Roland W. Scholz. 2008. “Empirical Analysis of the
Integration of Environmental Risks into the Credit Risk Management Process of European
Banks.” Business Strategy and the Environment 17 (3): 149–159. doi:10.1002/bse.507.
Weber, Olaf, Asadul Hoque, and Mohammad Ayub Islam. 2015. “Incorporating Environmental
Criteria into Credit Risk Management in Bangladeshi Banks.” Journal of Sustainable Finance
& Investment 5 (1-2) (January 2): 1–15. doi:10.1080/20430795.2015.1008736.
http://www.tandfonline.com/doi/full/10.1080/20430795.2015.1008736.
Wheatbelt NRM. 2013. “Soil Acidity.” Northam: Wheatbelt Natural Resource Management.
https://www.wheatbeltnrm.org.au/sites/default/files/basic_page/files/Acid soils.pdf.
Zeidan, Rodrigo, Claudio Boechat, and Angela Fleury. 2015. “Developing a Sustainability Credit
Score System.” Journal of Business Ethics 127 (2) (March 7): 283–296. doi:10.1007/s10551-
013-2034-2. http://link.springer.com/10.1007/s10551-013-2034-2.