Opportunities to build resilience of beef cattle properties in the mulga lands of south-western Queensland, Australia

Keywords: carbon farming, decision making, drought management, extensive grazing systems, farm-management economics, goats, modelling, profitability, rangeland management, rangelands, technology adoption.

The Mulga Lands bioregion in Australia covers 25.2 million ha (74% within south-western Queensland, 26% in north-western New South Wales; Thackway and Cresswell 1995), with extensive grazing by cattle and sheep being the primary land use (Commonwealth of Australia 2008; ABS 2021). Livestock producers face major challenges associated with the region’s inherently low productivity and profitability (Pressland 1984; Johnston et al. 1990; Venn et al. 2003). These challenges are exacerbated by widespread, well documented pasture degradation (Mills et al. 1989; McKeon et al. 2004; Commonwealth of Australia 2008), with this bioregion being recognised as the most extensively degraded landscape in Queensland (Wilson 1999). Additionally, similar to rangelands elsewhere in Australia, the Mulga Lands bioregion experiences substantial climate variability, including extended and extensive droughts (LongPaddock 2021), variable commodity prices and a long-term declining trend in terms of trade (ABARES 2019). To remain in business through these challenges and to build resilience, livestock producers must regularly produce a profit to build capital and equity.

Traditionally, the Merino wool sheep was the dominant production system in the Mulga Lands bioregion (Johnston et al. 1990). However, decline in the economic competitiveness of the wool industry caused a decline in sheep numbers over the past six decades, and beef cattle largely replaced wool sheep (ABS 2021; Chudleigh 2021). Livestock producers have also diversified into (1) rangeland goats through either harvesting (capture of wild/feral animals) or managed production (Heywood et al. 2000); and (2) carbon farming through carbon sequestration (Baumber et al. 2020). Few farm-level studies have been conducted during the past three decades to assess options available to mulga lands beef cattle producers to improve profitability and overall property resilience.

The value of the farm-management economics framework in assessing alternative management strategies for their impact on profit and risk, and thus to support decision-making and improve resilience of extensive livestock producers, has been demonstrated elsewhere in northern Australia (Bowen and Chudleigh 2021a, 2021b). This approach was also applied two to three decades ago to assess management options for extensive livestock production systems across Australia (e.g. Foran et al. 1990; Stockwell et al. 1991; Buxton and Smith 1996). The objective of the present study was to use the farm-management economics framework to conduct a contemporary assessment of management strategies and investment options to improve profitability and resilience of a hypothetical beef cattle property representative of the mulga lands of south-western Queensland. This region was considered as an example of low-productivity grazing lands in a semiarid environment.

A hypothetical property was constructed to be representative of the Mulga Lands bioregion and of grazing properties within 100 km of the town of Charleville in south-western Queensland. The attributes of this property and herd characteristics were derived from industry surveys and relevant research (Holmes 1980; Passmore 1990; Clarke 1991; O’Rourke et al. 1992; Bortolussi et al. 1999, 2005; McGowan et al. 2014), as well as the expert opinion of scientists, extension officers and nine local producers obtained in discussions during 2019–2020. The property comprised 20 000 ha of mulga (Acacia aneura), a native leguminous fodder tree, and associated native pastures on land types comprising soft mulga (50%), hard mulga (25%), and black soil (25%, gidgee (Acacia cambagei)/brigalow (Acacia harphylla)/yapunyah (Eucalyptus ochrophloia); The State of Queensland 2019). The black soil was scattered through the mulga land types. The property was considered deficient in pasture phosphorus (P) for cattle (4–5 mg/kg bicarbonate-extracted P (Colwell 1963) in the top 100 mm of soil; Bowen et al. (2020b)).

The initial base beef cattle enterprise had a low level of management input. The starting stocking level for the property of 600 adult equivalents (AE), (0.03 AE/ha), was considered representative of that applied by local landholders. The property relied on mechanical harvesting of mulga browse for ca. 6 months every 2 years (on average over 30 years) to provide sufficient fodder for cattle at the applied stocking rate. The beef production system was a self-replacing Bos indicus crossbred breeding herd (ca. 50% B. indicus). The management features of the initial base herd included continuous mating with two main musters per year to castrate calves, sell steers and identify cull (saleable) breeding cows. Replacement heifers were separated from the breeding herd until first mating at ca. 2 years of age. Steers were sold off their mothers at 10–12 months of age and 220 kg liveweight (i.e. without weaning). The herd received no vaccinations for animal health and no supplements other than hay for steers immediately before sale. The herd model was developed on the basis of long-term expectations of breeder reproductive performance and cattle growth paths in this environment under these low-level management inputs. Key performance and price assumptions for the initial base herd with low-level management are given in Table 1. The expected values for long-term cattle prices were estimated from 10 years of historical price data for Queensland abattoirs (MLA 2019) and at Roma, the closest saleyards to the study area. The latter prices were discounted to correct for the lower weights and condition of cattle from the Queensland mulga lands relative to other more productive areas.

Table 1.  Underlying assumptions and modelled property-level returns, expressed as the annual operating profit and the rate of return on total capital, for alternative livestock enterprises on a representative property in the mulga lands of south-western Queensland.

The farm-management economics framework (Bowen and Chudleigh 2021a) was applied to the hypothetical, representative property to assess: (1) a range of management strategies to improve the profitability of the property when operated as a beef cattle business; and (2) two alternative investment options, rangeland meat goat production and carbon farming. The farm-management economics framework applies the economic principles outlined by Makeham (1971), Makeham and Malcolm (1993) and Malcolm et al. (2005). The economic and financial effects of implementing alternative management strategies or investment options were assessed by marginal comparisons to the base property over a 30-year investment period. For each management change or alternative investment, the following were assessed: (1) additional capital and labour required, (2) effects on herd structure, (3) implementation phase, (4) timing of costs and benefits, (5) economic life of the investment, and (6) associated financial impacts and risks. The Breedcow and Dynama (BCD) herd budgeting software (Holmes et al. 2017) were used to analyse beef enterprise options. Economic and production models similar to those in the BCD software were developed to assess rangeland meat goat production and carbon farming. For each cattle and goat herd, the optimal (most profitable) age of female culling (sale), and the optimal male sale age and weight, were determined.

The economic criteria were the net present value (NPV) at the required rate of return (5%; as the real opportunity cost of funds to the producer) and the internal rate of return (IRR). The NPV represents the addition to the investors’ current wealth above or below the gain if they had invested the capital involved in an alternative that earned at the real discount rate applied. The IRR indicates the return on the extra capital invested. The NPV was calculated over the 30-year life of the investment, expressed in present day terms at the level of operating profit. The latter was calculated as Operating profit = (total receipts − variable costs = total gross margin) − overheads. The carbon farming analysis accounted for the extended nature of the contract applied (100-year life) by representing the present value of the investment for Years 31–100, for both ‘without-change’ and ‘with-change’ scenarios, as the expected sale value of the property in Year 30. The present value of the property was calculated as the discounted value of the expected future net income streams for each scenario. No real capital gain for the property was included for either the ‘without-change’ or ‘with-change’ carbon farming scenarios for the property. Additionally, the present value of the expected costs of managing the carbon farming project from Year 31 to Year 100 were deducted from Year 30 of the NPV calculation. The applied 5% discount rate sufficiently reduces the impact of costs and benefits accruing past Year 30 to make the NPVs of the alternative property management strategies broadly comparable. For all analyses, opening and salvage values for land, plant and livestock were applied at the beginning and end of the discounted cash flow analysis to capture any changes in the opening and residual asset value. Plant replacement was incurred as a capital cost less a salvage value in the year it was incurred during the investment period. An amortised NPV was calculated at the discount rate over the investment period to assist in communicating the difference in returns between the base property and the property after the management strategy was implemented. The IRR was calculated as the discount rate at which the present value of extra income equalled the present value of extra expenditure (capital and annual costs), i.e. the break-even discount rate.

Financial criteria calculated were peak deficit, number of years to peak deficit, and the payback period in years. In all analyses, the beef enterprise started debt-free, but debt was accumulated, and interest paid, as required for implementation of alternative management strategies or investment options. Peak deficit in cash flow was calculated assuming interest was paid on the deficit and compounded for each additional year in the investment period. The payback period was the number of years taken for the cumulative present value to become positive. All analyses included calculation of cumulative and net cash flow measures at the property level.

The constructed base beef production system was used to examine the benefit of implementing management strategies relevant to the region (Table 2). These strategies were assessed in three stages, each with a new base for comparison as strategies were sequentially implemented:

1. Sequential implementation of basic management strategies. The base for comparison was the property with low-level management.

2. Feeding mineral supplements containing sulfur (S), P and nitrogen (N). The base for comparison was the property that had already implemented the safe carrying capacity, weaning, pregnancy-testing, basic herd vaccinations and the optimal steer sale age of 18 months.

3. Additional strategies. The base for comparison was the property that had already implemented the safe carrying capacity, weaning, pregnancy-testing, basic herd vaccinations, the optimal steer sale age of 18 months, and S and P supplements fed during the pasture-growing period.

Table 2.  An overview of the management strategies that were modelled for the representative, example beef cattle enterprise in the mulga lands of south-western Queensland.

Key performance and price assumptions for each modified base herd are given in Table 1. Changes considered to the starting management of the base property and cattle herd were suggested by experienced industry participants as potentially positive changes for the property and herd efficiency. The estimated effect of management strategies on property carrying capacity and on cattle production indices were assigned with reference to published and unpublished research, and the expert opinion of beef producers, scientists, and extension officers with an extensive knowledge of the northern Australian cattle industry. Key parameters are in Tables 1, 2, and details of assumptions and their derivations are given in Bowen and Chudleigh (2021c).

First, the representative property was modelled to run as an established meat goat enterprise, in a steady-state analysis, for comparison with (1) the self-replacing cattle herd under various management levels, and (2) a steer turnover enterprise. For the steady-state analysis, it was assumed that wild dog exclusion fencing, internal fencing and infrastructure were in place for efficient management of rangeland goats. Second, the marginal returns were calculated for complete conversion from the self-replacing beef enterprise to the rangeland goat enterprise over a 24-month transition period. The base cattle herd was the enterprise that had implemented basic management strategies and also S and P supplements fed in the pasture-growing period. Investment in a wild dog exclusion fence (required for goats but not for cattle) was AU$500 000 (AU$10 000/km) and specialist goat handling equipment and refurbishment of internal property infrastructure required investment of AU$165 000.

The modelled meat goat enterprise was a self-replacing herd that received basic management, including (1) weaner bucks were not castrated but were separated until sale, and (2) weaner does were separated from bucks until yearling mating. The goat enterprise was based on rangeland goat genetics, with some crossbred bucks used as sires (i.e. rangeland goats crossed with Kalahari reds, Boer, or Anglo Nubian genotypes). There is a lack of published data on the biology of meat goat production in this region, so the herd characteristics were again informed primarily by expert opinion from discussions in 2019–2020. Key performance parameters and price assumptions are given in Table 1. The expected sale price for goat meat was estimated from long-term price data (MLA 2019), in accord with expectations of local property managers. Grazing pressure equivalence to 500 AE of cattle was determined according to McLennan et al. (2020), where an AE or dry sheep equivalent (DSE) rank was assigned as the ratio of the metabolisable energy (ME) requirements for a particular level of production to that of a ‘standard animal’. The standard animal for both cattle and goats was defined as having zero weight change, walking 7 km/day on level ground and consuming forage of 55% dry matter (DM) digestibility (7.75 MJ ME/kg DM). A standard bovine animal, representing one AE, was defined as a 450-kg, 2.25-year-old B. taurus steer requiring 73 MJ ME/day. A standard goat, representing one DSE, was defined as a 45-kg wether, with no fibre growth above that at maintenance, requiring 8.7 MJ ME/day. This corresponded to a ratio of AE:DSE of 1:8.4.

Potential returns to investment in carbon farming, on 50% or 75% of the property area, were examined for the property when initially fully stocked with a self-replacing herd of either (1) beef cattle or (2) rangeland meat goats. The base cattle herd was the enterprise that had implemented basic management strategies and also S and P supplements fed in the pasture-growing period. The base rangeland meat goat herd had wild dog exclusion fencing and suitable internal infrastructure already in place. A 50% conversion of the property to carbon farming was assumed to be equivalent to 50% of the long-term carrying capacity of the property, whereas 75% conversion was assumed to be equivalent to 80% of the carrying capacity. The difference was due to likely inclusion of some of the higher carrying-capacity land types in 75% of the property area. The area not allocated to the carbon contract was assumed to maintain its carrying capacity. Operating and overhead costs of the livestock enterprises were adjusted accordingly. The area contracted for carbon sequestration was set aside from tree clearing for a 100-year permanence period to align with the most common local practice (Clean Energy Regulator 2021). To simplify the analysis, and match the carbon sequestered to the estimates gained from exclosures by Witt et al. (2011), livestock grazing was excluded from the contract area for the entire period. As no data were available to model a dynamic rate of sequestration for the generic property, a constant rate of sequestration of 1.2 t CO₂-equivalent (e)/ha. annum for 25 years was applied, with a total of 30 t CO²-e/ha accumulated, in accord with published estimates and allowing for a potential negative impact on carbon sequestration of any disturbance events (Fensham and Guymer 2009; Witt et al. 2011; Fensham et al. 2012; Peters and Butler 2014; Thamo et al. 2017). The 25-year period for income (the crediting period) aligned with the conclusion of Witt et al. (2011) that mulga land carbon balances are likely to be in equilibrium after 25 years. Additionally, 25 years is the maximum crediting period currently allowed by the Clean Energy Regulator (2021). The initial gross carbon price was assumed to be AU$12.50/t CO²-e (Cockfield et al. 2019) and was maintained in real terms for the 25-year crediting period. Furthermore, following Cockfield et al. (2019) and the advice of local landholders, the carbon price achieved at auction (gross price) was reduced by 25%, to allow for project management fees and that 5% of value would be retained by the Clean Energy Regulator (2021). This provided an on-farm return of AU$9.30/t CO²-e. Operational costs to establish and maintain the areas under carbon contract were set at AU$1.5/ha. annum for the first 5 years, with annual maintenance costs being subsequently halved for the remaining 95 years.

The steady-state profit analysis of the initial, base beef cattle enterprise indicated a return of −AU$78 213, (or −2.5%), on total capital invested, with total farm income being insufficient to pay total property costs (Table 1). Of the initial basic management strategies, only targeting the optimal age of steer turnoff had a measurable (greater than ±AU$5000/annum) effect on annual enterprise profit (Table 3). However, total property returns were −1.9% (Table 1). When the property with fully implemented initial basic management strategies was used as a modified base to test the value of feeding mineral supplements, the greatest benefit was due to S and P supplements fed only during the pasture growth period. Property profit increased by AU$7100/annum (Table 3) but total property returns were −1.5% (Table 1). Implementing dry period mineral supplements decreased property returns when fed alone (−AU$2000/annum) and decreased the benefit to growing period supplements when fed in combination (AU$4100 extra profit/annum).

Table 3.  Profitability and financial risk of implementing management strategies for a representative, hypothetical beef enterprise in the mulga lands of south-western Queensland.

The effect of implementing additional strategies provided very limited opportunity to improve profitability and, hence, viability of the cattle enterprise. All strategies except ‘destocking in response to drought’ resulted in an unmeasurable (less than ±AU$5000) or large negative effect on annual profit (Table 3). Even destocking in response to drought added to profitability only when savings in fuel, oil, repairs and maintenance associated with feeding mulga browse could be reduced by at least 20%, on average, in combination with a reduction in operator’s allowance of 10%.

The steady-state profit analysis of the established rangeland meat goat enterprise indicated the property returned AU$52 223, or 1.6%, on total capital (Table 1). Where full investment in a wild dog exclusion fence around the property boundary and existing infrastructure refurbishment were required to facilitate a change from cattle to goat production, the investment added AU$48 300 profit/annum and generated an IRR of 10.8% (Table 4). However, the investment substantially increased the riskiness and indebtedness of the property, as indicated by the substantial peak deficit (−AU$876 000) and payback period (14 years).

Table 4.  Profitability and financial risk of implementing alternative investment options for a representative, hypothetical beef enterprise in the mulga lands of south-western Queensland.

Partial conversion of a beef enterprise to carbon farming substantially improved property profitability, with 75% conversion adding AU$36 800 profit/annum and 50% conversion AU$26 600 profit/annum (Table 4). However, partial conversion of a rangeland meat goat enterprise to carbon farming decreased profitability (−AU$17 400 and −AU$36 800 less profit/annum for 50 75% conversion respectively; Table 4).

In general, there was limited opportunity to improve profitability, and hence viability and resilience, of the hypothetical beef cattle enterprise. The cumulative effect of implementing a basic level of herd management and other strategies was a property with negative total returns and declining cumulative cash flow over 30 years of analysis. This finding supports earlier studies when Merino wool sheep production was the region’s dominant enterprise (Holmes 1980; Johnston et al. 1990; Passmore 1990).

Strategies likely to increase the profitability of beef cattle enterprises in the Mulga Lands bioregion were consistent with those for other geographical regions in Queensland and the Northern Territory, and included increasing the age of steer turnoff from weaners to the optimal, and provision of mineral supplements (particularly P) in the pasture growth season where deficiencies occur (Bowen et al. 2020b; Bowen and Chudleigh 2021a). In addition to improving profitability, increasing age of steer turnoff to the optimal can improve drought preparedness because of a reduced breeder herd relative to growing cattle at the same grazing pressure. However, an impediment to this strategy is the substantial peak deficit in cash flow incurred while steers are initially held for longer prior to sale. Soils of the Mulga Lands bioregion are severely deficient in available P and N (Dawson and Ahern 1973; Beale 1994; McLennan et al. 1999). Ruminant diets with high proportions of mulga browse exacerbate P and N deficiency, and S and sodium also become limiting (McMeniman et al. 1986a; Clarke 1991; Pritchard et al. 1992). Additionally, the low digestibility (ca. 45% DM digestibility) and low voluntary intake of mulga browse severely curtail performance (McMeniman et al. 1986b; Pritchard et al. 1992). Low levels of strategic mineral supplements such as P and non-protein N (urea) constitute one of the few, low-cost options for northern Australian beef producers to reduce the effects of nutritional deficiencies in pasture and improve breeder productivity (McCosker and Winks 1994; Dixon 1998). However, the present analysis only indicated measurable improvements in whole-farm profitability from supplementing with S and P in the pasture-growing period (AU$7100 extra profit/annum). The poor profitability response to dry period S, P and N supplementation (−AU$2000/annum) is in accord with analysis for the more productive Central Queensland region where, similarly, the costs relative to benefits of dry period supplementation were higher than for P supplementation in the pasture-growing period (Bowen et al. 2020b). This issue warrants more detailed investigation across northern Australia’s grazing lands.

In the present study, several management strategies that improve the profitability of beef cattle properties in more productive regions of northern Australia resulted in either unmeasurable (less than ± AU$5000 profit/annum) or negative effects on profitability. For example, the decrease in property profitability (−AU$16 100/annum) resulting from converting the mulga lands breeding herd to a steer turnover operation contrasts with results for more productive areas of Queensland (Bowen and Chudleigh 2021a). The decreasing rate of steer liveweight gain over time in the low-productivity Mulga Lands bioregion was the major cause of this difference. The rate of steer liveweight gain assumed in this analysis was primarily based on data from local producers. However, it aligns with data for steers grazing low-productivity land types in the Northern Territory (Schatz 2011; Cowley 2012).

Similarly to the results in the present study, analyses for other low-productivity regions in northern Australia have indicated that property profitability can be improved by implementing optimal stocking rates, breeder segregation, better weaner management, and, where needed, closely targeted P supplements. However, as for the current analysis, the addition of controlled mating to these management strategies generally reduced profitability (Chudleigh et al. 2017, 2019). The poor nutrition in these low-productivity regions, and consequent poor herd reproductive efficiency, make controlled mating unlikely to improve returns unless high-risk livestock trading activities are incorporated to overcome the large cash flow deficits caused by a change from continuous to controlled mating.

Despite government- and grazier-supported initiatives in the 1990s to (1) promote property amalgamation for improved enterprise efficiency, (2) control total grazing pressure, and (3) objectively assess safe livestock carrying capacities (Johnston et al. 1996a, 1996b; Rose 1998), and the more recent Queensland Vegetation Management Act 1999 (The State of Queensland 2018), the condition of the Mulga Lands bioregion as a grazing resource appears to be in continued decline (Commonwealth of Australia 2008). A stocking rate reduction to achieve the safe carrying capacity for the hypothetical property was considered the essential first step in our analysis to prevent further degradation. However, despite releasing ca. A$60 000 worth of livestock capital over the 2 years of adjustment to the lower stocking rate of 500 AE, the long-term economic and financial outlook of the property remained similar to that when running 600 AE. The lack of research data for beef cattle in this region to support assumed improvements in livestock productivity resulting from implementing a safe carrying capacity indicates that this analysis should be considered with caution. Additionally, the use of the average market price to value stock sold as part of the herd reduction may have overstated livestock value, because such stock would be unlikely to be in the same average condition as are normal sale cattle. Regardless, the outcome of this strategy is consistent with that of a similar analysis conducted for a Northern Gulf property in Queensland (Bowen et al. 2019). We did not examine further interventions to improve degraded rangeland pastures, other than reducing stocking rate, because evidence suggested that restoration would be unlikely, and because of economic constraints (Johnston et al. 1990; MacLeod and Johnston 1990; Commonwealth of Australia 2008).

Livestock enterprise and resource management in this bioregion relies on harvesting mulga browse (Page et al. 2008). However, retaining livestock through use of mulga fodder when grass biomass is limiting has been the major contributor to pasture degradation (Pritchard and Mills 1986; Mills et al. 1989; Johnston et al. 1990). Stocking rates and pasture utilisation rates in the Mulga Lands bioregion are generally higher than values estimated as ‘safe’ for maintaining pasture condition (McKeon et al. 2004; Commonwealth of Australia 2008). Additionally, high pasture utilisation levels may require more frequent and extensive mulga harvesting to maintain livestock numbers (Commonwealth of Australia 2008; Page et al. 2008). In our analysis, we examined the effect on property profit of destocking in response to extended dry periods, with a scenario of destocking half of the breeding herd for 18 months every 10 years. This analysis depended on uncertain assumptions of climatic variability and the associated impact on property carrying capacity, prices, and costs. Nevertheless, a key finding was that destocking in response to drought was likely to add to the property profitability if savings in fuel, oil, repairs and maintenance costs associated with feeding mulga browse could be reduced by at least 20%, and operator’s allowance (labour costs) reduced by 10%. The most appropriate strategy for individual producers to destock (sale or agistment of cattle on another property), then rebuild herd numbers in the recovery phase (natural increase, purchases, agistment income) will depend on the costs, livestock prices at the time, and the availability and/or demand for agistment.

Because of the limited opportunity identified in our analysis to improve the profitability and viability of the beef cattle enterprise, alternative investment options were examined. The first examined was conversion to the production of rangeland meat goats. This generated substantially more profit (AU$48 300 extra profit/annum; 11% IRR) than did any strategy examined to improve performance of the existing cattle enterprise. However, performance of this investment is heavily dependent on the assumptions that the estimated relative and absolute goat meat price will be maintained over the long term. Although profitable over 30 years, an important constraint was the level of debt required (−AU$876 000 peak deficit) and the long interval (14 years) before the property was expected to return to the same financial position as that without change. This financial risk is challenging for property managers where high costs of wild dog exclusion fencing must be funded wholly by the property. Similar results were obtained for conversion from beef to goat production for a representative property in the central-western rangelands of Queensland (Bowen and Chudleigh 2021d).

An aspect of goat production systems in the mulga lands is that goats select proportionally more browse than do other livestock species (Hacker and Alemseged 2014; Pahl 2019). Furthermore, evidence suggests that goats can digest and utilise mulga browse more effectively than do cattle or sheep because of adapted rumen micro-organisms that improve nitrogen digestion and retention (Brooker et al. 1994; Miller et al. 1995). Our estimate of the number of goats able to be run on the representative property was conservative, because we assumed grazing pressure equivalency of cattle and goat animal units, based on energy requirements. This conservative approach was taken in the absence of data to indicate substitution ratios between cattle and goats, and to reduce the risk of pasture overutilisation. Overutilisation by goats can occur on commercial properties because of high reproductive rates (124% weaning rate in this analysis), and possibly the greater drought resilience and survival of goats compared with cattle or sheep. This latter attribute is considered to be partially a consequence of the more flexible diet of goats and their better ability to utilise browse and select for diet quality. In the current analysis, the rate of sale of surplus goats was sufficient to avoid increasing grazing pressure compared with the beef cattle enterprise. Lack of reliable data for managed rangeland meat goat production in this bioregion demands caution in extrapolating these results. However, given profitability of the goat enterprise, it appears that rangeland goats could have an important role in maintaining profitable and resilient mulga land production systems. This would require ongoing demand for goat meat and improved knowledge of effective goat management systems.

Although historically dominant in the mulga lands, Merino wool sheep production is now uncommon (ABS 2021). For this reason, as well as the lack of interest by the local advisory group in examining sheep (wool or meat) enterprises for this mulga-dominant property, sheep were not included in this study. Analyses for the central-western rangelands of Queensland indicated that existing Merino wool sheep enterprises were profitable and resilient on the basis of contemporary prices (Bowen and Chudleigh 2021d). However, changing from a beef cattle to a sheep enterprise generally requires even greater capital investment than that for a change to goats, and, consequently, increased financial risk and indebtedness (Bowen and Chudleigh 2021d).

The second alternative examined in the present study, partial conversion of a beef enterprise to carbon farming, substantially improved property profitability over 30 years, with 75% land-area conversion being superior to 50% conversion. However, partial conversion of a rangeland meat goat enterprise to carbon farming decreased profitability over 30 years, despite carbon farming improving cash flows in the short to medium term. The analysis indicated that the opportunity cost of the investment will be a key factor determining whether carbon farming is attractive. With an unprofitable enterprise in place (e.g. beef production in this analysis), the opportunity costs of investing in carbon farming are low and provide a greater incentive for carbon farming. It appears likely that the widespread adoption of carbon farming in the rangelands (Baumber et al. 2020) has been predominately due to the extended droughts and low commodity prices of the past decade. These circumstances have reduced the opportunity costs, and/or increased the discount rates, of some landholders to the point that carbon farming became attractive. A sequence of more favourable seasonal conditions and continuation of higher commodity prices could slow the conversion of large parts of the mulga lands to carbon farming. Even so, the relative profitability of carbon farming indicates that its adoption on portions of properties is likely to be considered closely by many landholders. This is particularly likely if carbon prices were to increase in real terms over time. However, managers must not only apply the correct methodology when assessing the potential for carbon sequestration, but also an appropriate framework to assess the economic and financial value of carbon farming. Furthermore, the tax implications for this non-primary production income stream, and the uncertainty of impacts of long-term carbon contracts on property sale value, were not included in our analysis but should be considered by property managers.

Carbon farming and rangeland goat production provide a diversified income for a beef property, potentially stabilising income over time as well as improving average profitability. This can reduce risks from climate variability and assist with drought preparedness and resilience (Buxton and Smith 1996; Freebairn 2019; Bowen and Chudleigh 2021d). However, as identified previously (Bowen and Chudleigh 2021a, 2021b), individual properties may have characteristics different from those assumed in this analysis. Furthermore, the goals, resources, and skills of property managers will influence decision-making and adoption of alternative management strategies. We recommend that the farm-management economics framework applied in our analysis be considered at an individual property level to account for property and manager-specific factors and to better guide decision-making in rangelands generally.

The poor returns from beef cattle production have led some producers in the mulga lands to rely on non-farm income for business survival. This has been identified as important in inherently low-productivity regions with a history of subdividing large properties (Johnston et al. 1990; Fargher et al. 2003). There was considerable interest in the 1980s and 1990s in promoting property amalgamation in the mulga lands to take advantage of economies of size, and thereby improve profitability, and to reduce the economic incentive for adoption of exploitative grazing strategies (Holmes 1980; Passmore and Brown 1991). However, the disconnect between land value and production potential has limited the capacity of local landholders to achieve such an outcome. Bowen et al. (2020a) found no positive effect of purchasing additional grazing land even when more productive land was purchased, when a 5% discount rate was applied to indicate the opportunity cost of the additional capital. A similar outcome would be expected for the mulga lands property. Fargher et al. (2003) and Hamblin (2009) have argued that, in regions where environmental and social decline are endemic, more effective land management policies should be implemented to retire low-productivity land from grazing, with support given to grazing businesses to restructure.

This study provided insights into the opportunities to improve profitability and resilience of a hypothetical grazing enterprise in a low-productivity, semi-arid rangeland environment. The finding of low inherent productivity and profitability of beef cattle production in the Mulga Lands bioregion is in accord with previous studies when Merino wool sheep production was dominant (Holmes 1980; Johnston et al. 1990; Passmore 1990). Our study indicated very limited potential to improve the profitability or resilience of the existing beef cattle enterprise. However, full or partial conversion to rangeland goat production or carbon farming improved property returns and viability.

The majority of the data that support this study are available from the final project report (Bowen and Chudleigh 2021c), which is available online at: https://futurebeef.com.au/projects/improving-profitability-and-resilience-of-beef-and-sheep-businesses-in-queensland-preparing-for-responding-to-and-recovering-from-drought/. Additionally, the herd and flock models used in the analysis can be obtained from the authors upon reasonable request.

The authors declare no conflicts of interest.

This study was co-funded by the Department of Agriculture and Fisheries, Queensland, and the Queensland Government Drought and Climate Adaptation Program (DCAP 2) as part of Project DAF 6.