2. Materials and Methods
Data analysis is considered in the context of the voice of the distressed business operator (referred to as client in the context of this paper). Consequently, to address the research aim, data analysis was based on the responses linked to unique client ID’s as recorded in the RFCS program database (representing the total number of business operators that have entered the program). About half of the clients in the program have one record corresponding to a single start and end date. The other half of clients have multiple start and end dates within the four-year period.
Data analysis was conducted using a dataset collected by individual RFCS providers and collated and maintained by the Australian Government during the period 2012 to 2016 as part of the formal process of registering and accepting a person or business (herein referred to as clients) into the financial counselling system. The Australian Government provides oversight of the RFCS program through the Department of Agriculture, which is currently situated within the Department of Agriculture, Water and Environment established in February 2020. The dataset was drawn from an archived version of the Australian Rural Counselling (ARC) database, which is the central repository for RFCS client data [2
]. The data maintained in ARC comprise records all of the RFC interactions with clients for the period, as collected and uploaded by the RFCs. De-identified data were provided for analysis by the Department of Agriculture, Water and Environment (DoAWE), with individual clients identified by a unique number only. The provided dataset was accessed and managed in accordance with ethics approval HEC18044 granted by the La Trobe University Ethics Committee.
2.3. Data Processing
De-identified data were available for ten distinct modules related to the way the RFC captures and characterizes the interaction with the clients (Table 1
). This included the client’s postcode, industry sector, financial situation, the reasons for joining the program, and the outcome at the end of the program. Analysis focused on data in the Assessment module, which reported the reasons nominated by the client to join the program (causes of difficulty). Unique client identification numbers were cross matched with the remaining modules to determine other key variables.
The Assessment data comprised 977,933 records across eight attributes, which were processed to a final dataset of 88,148 instances attributed to 33,325 unique clients. Data processing included removal of “NULL” records (no data recorded for cause of difficulty) and duplicate client ID’s (due to multiple start and end dates in the program database). Data were further processed to remove records for which causes of difficulty were identified that were not consistent with the identified options. The ARC database was “rebuilt” in 2012 (DoAWE Personal Communication) with the result being that clients of the RFCS rolled over from the previous service provision period (i.e., pre-2012) recorded causes of difficulty that were not anticipated within the context of the 2012–2016 service. Consequently, the assessment of cause of difficulty captured for these clients was reflected in the database as “Data Migration”. This characterisation was associated with 4903 instances.
For each client ID interaction, the cause of difficulty (and hence the reason for participating in the program) was identified by the RFC from a drop-down list as follows:
Low sales/Commodity prices
Enterprise management skills
Increased operating costs
Financial management skills
Declining asset values
To further characterize the client’s profile in terms of their location and industry sector, data from the Client Details module reporting the “home address state” and “primary enterprise ANZSIC”, were merged with Assessment.
The “home state address” refers to the six federated states and two territories in Australia, namely, the Australian Capital Territory (ACT), New South Wales (NSW), Northern Territory (NT), Queensland (QLD), South Australia (SA), Tasmania (TAS), Victoria (VIC) and Western Australia (WA).
The industry sectors are categorized according to the ANZSIC (Australian and New Zealand Standard Industrial Classification) codes, of which 26 activities were included in the dataset, which included: Citrus Fruit Growing; Cotton Growing; Dairy Cattle Farming; Floriculture Production (Outdoors); Grape Growing; Horse Farming; Nursery Production (Outdoors); Nursery Production (Under Cover); Onshore Aquaculture; Other Agriculture and Fishing Support Services; Other Crop Growing n.e.c.; Other Fishing; Other Fruit and Tree Nut Growing; Other Grain Growing; Other Livestock Farming n.e.c.; Pig Farming; Poultry Farming (Eggs); Poultry Farming (Meat); Rice Growing; Sheep-Beef Cattle Farming Sheep Farming (Specialised); Small Business; Stone Fruit Growing; Sugar Cane Growing; and Vegetable Growing (Outdoors).
2.4. Data Analysis
The 33,325 unique client records were aggregated (using Excel Pivot Tables) according to the causes of difficulty reported, their location (postcode and state) and industry sector. The results obtained were compared, whenever relevant, with data available for agribusiness at the national and state level. ArcGIS 10.5 software (esri, Beijing, China) [12
] was used to map the spatial distribution of client IDs using their postcode, and the causes of difficulty reported.
Causes of difficulty were statistically compared at the national level (i.e., using all responses) using a chi-square (Χ2
) goodness of fit test to determine if the response frequency for each cause was statistically different from an expected frequency distribution consisting of all causes being reported equally. At the state level, a Χ2
association test was performed to determine if there was a relationship between causes of difficulty reported and the client’s location (state or territory). Likewise, the same test was carried to evaluate if causes of difficulty were associated or not with industry sector. These statistical tests were performed using available functions in R software (R Core Team, Vienna, Austria) [13
Articulating the drivers of rural hardship is an important step in informing future government and industry intervention to underpin the resilience and sustainability required of the farm-business sector if it is to capture the benefits of a burgeoning global population. This paper sought to understand the relative drivers of rural business hardship as interpreted by those operators experiencing the hardship. That is, through the voice of distressed farmers. The data reviewed reflected the collective voice of approximately 33,000 rural business operators seeking the assistance of RFCs across Australia during the period 2012–2016.
Reconciling the number of clients, the total number of instances and the total number of farms is a vexed issue. While the number of unique clients utilising the program throughout the five year period (2012–2016) is available, the number of instances is less clear as a consequence of the manner in which the data has been recorded. The data reviewed identified numerous entries (data not shown) in which individual clients had been recorded as appearing to exit the program on a particular date and subsequently re-engaging with the program as a new instance on the following day. This may have been a perverse outcome of service-level interpretation of a desire by the Australian Government to ensure effective and efficient management of clients through the program (DoAWE personal communication). However, it may also reflect broader concerns regarding the historic collection and maintenance of the data. This study has reviewed approximately 33,000 instances captured by RFCs. While this data is the only record of the transactions between the farm business owners and the RFCs, anecdotal evidence has previously questioned the veracity of this data, with errors arising from a limited understanding of either the reason for the collection of the ARC data or the need for its accurate capture [2
When considered in the context of the total number of farms in Australia during that period, the unique interactions reflect approximately 26% of that number (Table 6
). Furthermore, interactions are spread across all industry sectors and throughout the continent.
The estimation of total number of farms in Table 6
is given via the Rural Environment and Commodity Survey (REACS) conducted by the Australian Bureau of Statistics, which uses an internal assessment of estimated value of agricultural operations (EVAO) to derive the number farms in the survey period based on a financial threshold. In 2013–14 that threshold was >AUD 5000, which [16
] was subsequently revised in the 2015–2016 REACS to > AUD 40,000 [17
]. Applying the latter valuation to the data, which reflects of the order of 88,000 farm businesses, is beyond the scope of this paper as it requires a far more nuanced understanding of the client businesses than is provided for in the in the data available for our research. It does, however, demonstrate the challenges of providing effective, evidence-based agriculture policy to underpin a sustainable and resilient sector. Given the ambiguity regarding what constitutes a farm business, it is important that a more rigorous boundary is placed around the terminology to enable a more effective use of the data captured in the delivery of programs such as the RFCS.
A key feature of the data accessed throughout the project was that the businesses participating in the RFCS program perceive that their involvement with the program is a result of climatic variation. That is but for the intervention of the climate, or indeed a changing climate, they would not have required the agricultural welfare support provided by the service. Although not clear if this is interpreted homogenously by farmers in the program, we can assume that changes in climate patterns have somehow disrupted business as usual and prompt the farmers to seek for help. Prima facie this is consistent with the prevailing science and analysis indicating that an increasingly dynamic climate is impacting and will continue to impact farm productivity into the future [8
]. Consequently, it is reasonable to argue that in the absence of a clear strategy to enable the effective adaptation to, and mitigation of the effects of projected climate change, agricultural resilience in Australia will be markedly impacted. A key outcome of which is the likely need for ongoing agricultural welfare over a longer timeframe. Indeed, it is clear from the current cohort that a lack of capacity or incentive to implement mitigation strategies is limiting resilience. Approximately 75% of the cohort examined in this study were either transferred to alternative government support or remained in the RFCS program at the end of the 2012–2016 period (data not shown). Considering that evidence of financial hardship is a pre-requisite for participation in this program, it could equally be argued that access to such a program risks embedding financial hardship in some businesses. However, there are insufficient data to assess the relative economic performance of the businesses in this cohort, and as such, this element remains an intriguing issue for future research.
It is important to reflect on the meaning of climate variation in the context of the current study. Australia’s climate patterns are characterized by frequently occurring extreme events, namely droughts, floods, heavy rainfall and heatwaves [19
]. It is particularly difficult to identify a long-term rainfall trend as spatial and temporal rainfall variability change significantly from year to year, season to season, region to region [21
]. This determines that an extreme event felt intensely in one part of the country might not be noticed in the other part. Kiem et al. [22
] suggest that taking rainfall variability as normal has led to ignoring the importance of its spatial and temporal dimensions when making decision about farming investment or water storage and supply. What can be shown is that the spatial extent of extreme events has been steadily increasing as well as its frequency and duration. Moreover, it has been shown to negatively impact agricultural production. For example, Hochman et al. [23
] report that wheat yields declined in Australia by 27% from 1990 to 2015 and attribute that to reduced rainfall and temperature rising. A feature of this change, particularly in south eastern Australia (where most of the agricultural production is concentrated), has been a clear change in rainfall seasonality with amounts of rainfall decreasing during winter/regular growing season [21
]. Anecdotal evidence provided by RFCs from across the RFCS program (Personal Communication) indicates that the category Climatic variation was used by RFCs to recognise the impacts of a broad range of climate related events impacting business sustainability, including extreme weather events such as fires, floods and storms, as well as complications associated with drought.
Current agriculture policy in Australia, the Agricultural Competitiveness White Paper, is focused on the profitability and resilience of farming families [3
]. Although the policy recognises climate as a risk factor, the rhetoric of both the current and broader policy thinking (see, for example, the Australian Government’s Backing our Farmers and Drought Affected Communities https://www.pm.gov.au/media/backing-our-farmers-and-drought-affected-communities
, accessed 31 October 2019) is directed at the mitigation of drought impacts as opposed to a broader consideration of the dynamic influences of climate more generally. On that basis, current policy settings may not be sustainable. This policy positioning is in contrast with the broader interpretation of climate change impacts in Australian agricultural environments that relate to a suite of outcomes that are projected to limit the resilience and sustainability of the sector, including temperature increases, changes in amount and distribution of rainfall, and more extreme weather events such as fires and flood [11
While the data in this paper is instructive, it should be considered in both context of its collection and the broader operating social context of the businesses themselves and it is appropriate that the data is interpreted with some caution. The data was collected using a computer assisted personal interviewing (CAPI) technique [25
]. While the technique is recognised as being superior to paper-based survey approaches, it is subject to both operator and interviewer action that may bias results [26
]. During the initial interview to assess the key issues causing the rural business operator to present on that occasion, a pre-determined list of factors is offered for consideration. In the CAPI model used in the capture of the 2012–2016 data, the list was only presented in descending alphabetical order, with the category of Climatic variation appearing at the top of that list. Consequently, this lack of randomisation of potential response may have influenced the selection of this factor as the primary cause of difficulty experienced by the business operators [25
]. While this may be a confounding factor, it does not seem reasonable to suggest that the overwhelming response (63% of clients) noted is due to that issue.
The perception that climate variability was a key driver in their current position may have also been influenced by the availability of direct government support to rural businesses available through a range of drought support programs such as Exceptional Circumstances (EC) and Climate Change Adjustment Program (CCAP) [2
]. While this too may have contributed to the result noted, there is insufficient detail in the collated data to consider its impact.
The social context of the business owners may also be a consideration in the identification of climatic variation as a primary cause of difficulty. Recognition as a good operator, providing the economic foundation for the farm business has been identified as a key element of farmer self-actualisation [27
]. It is conceivable that individual operators may have selected the category of climatic variation as a primary cause of difficulty resulting in their participation in the program as a means of deflecting attention of the RFC from other issues such as a lack of management skills and expertise. Indeed, these results suggest that a focus on climate in isolation masks the capacity to reconcile the impact of a suite of other factors that may be impacting business performance, such as personal factors and debt, which were of notable impact in a state-based context. This has important implications for future policy development in that it reinforces the argument for a broader evidence-based, more holistic approach to agriculture policy than has been demonstrated by successive governments in Australia over the past four or so decades [1
]. It follows, therefore, that the appreciation of the complex social context of farm business failure is an important consideration for future research that seeks to provide the evidence to policy makers enabling them to break from the status-quo and develop policy that supports sustainable and resilient agricultural landscapes and economies.
Agriculture has a long history of successfully lobbying government in Australia for preferential treatment [28
]. The voice of the farmer in this study is an important addition to the ongoing debate regarding government support for rural enterprises. However, as noted above, the veracity of the data maintained within the ARC database has been challenged [2
]. It is imperative, therefore, that the data in this study are considered in a broader, more objective context. That is, that the subjective voice of the farmer is balanced by an objective voice, with significant knowledge and experience working with this cohort. As independent service providers working with multiple businesses using a case management approach, RFCs have a holistic understanding of the financial, social and environmental factors contributing to the hardship experienced by stressed rural businesses participating in the RFCS program [2
]. As such, they are a key source of knowledge that could significantly broaden the understanding of the key drivers of rural business failure and, as such, provide important intelligence for future policy development. It is unfortunate that the data management obligations accompanying the implementation of the RFCS program have been so poorly fulfilled. Greater confidence in datasets such as these could have markedly improved the effectiveness of future policy decisions. That opportunity is lost.
It should be noted, however, that the RFC network remains an important source of skills, knowledge and experience that is not reflected in the current dataset. It follows then that a key next step in understanding the core issues driving rural hardship must necessarily include integration of this resource in policy consideration. As such, accessing the extensive skills knowledge and experience of the RFC network is an important and urgent step in the development of robust agriculture policy that will underpin a resilient and sustainable future for the sector.
Despite the opportunity presented by the demand for food and fibre to support a burgeoning global population, a cohort of Australian farm businesses are struggling to capitalise on that opportunity. A number of these farm businesses are experiencing significant hardship, with the failure softened only through access to agriculture welfare. This paper has sought to understand and highlight the drivers of significant hardship for these failing businesses, as articulated by farm business operators accessing the RFCS program funded by the Australian Government. It provides an important cross-sectoral assessment of issues limiting profitability and resilience in the Australian agricultural sector, as understood and articulated by operators of stressed rural businesses.
Exposure to climate is a recognised significant risk to the resilience and sustainability of agribusiness in Australia, a fact that is reflected in the access to and provision of agricultural welfare and financial assistance. It is foreseen that frequency and duration of disruptive climate events will increase prompting further climate-related productivity reductions. Consequently, direct and indirect agricultural welfare is likely to remain a much-needed support for Australian farming businesses into the future.
Climate variation was identified as the primary driver of business hardship by the majority of client participants in the program. The inference of this finding being that, were it not for the impact of climate, they would not have required the services of the RFCS program. It is important, however, to understand that this strong voice of the distressed farmer reflects their perception and as such must necessarily be considered within a broader social and business context. While the influence of climate is particularly evident for farmers in the jurisdiction of NSW in the program, other jurisdictions also highlighted the importance of alternative causes of difficulty including personal factors, lack of management skills, low prices, and debt levels. Hardship was also found to be related to specific industry sectors. Specifically, our study shows that grape growing farmers are more likely to identify low sales and commodity prices as the main cause of difficulty whereas grain growing is related to climate variation. These findings highlight that a more nuanced, broad-based approach to policy than has been demonstrated historically is required to sustain resilient agricultural landscapes and communities in Australia.
A more nuanced approach to future agriculture policy requires substantial research to enable the appropriate recognition of both the financial and social drivers of rural business failure in the context of a dynamic and changing future climate. While existing databases such as that underpinning the RFCS program have enormous potential to support future policy development, the use of such databases should be carefully considered. Issues such as the manner in which the data used in this study were collected and, indeed, its veracity, have markedly impacted the value of the current dataset in contributing to this higher policy goal. Ensuring that these issues are resolved through effective program design and oversight is an important first step.
The focus of this study was to illuminate the key drivers of rural business failure, as understood by business operators themselves. In doing so it has highlighted numerous challenges to the development and implementation of effective policy for resilient agriculture. A key next step for this work is to pursue an objective assessment of the drivers of business failure by capturing the combined experience and insights of the RFCs to provide a reality check to balance the subjective farmer-perceptions expressed here.