From Crisis to Preparedness: Enhancing Disaster Resilience for At-risk Communities with Anticipatory Cash Transfers

By Julian Garratt, UNSW Co-op, 2023 IMF/WB Fellow

See the PDF Policy Paper here

Executive Summary

Disasters have cost the Australian government billions in recovery and exposed homeowners to thousands of dollars in losses. Recent literature has proven the efficacy of anticipatory cash transfers based on Machine Learning forecasts to alleviate the strain on conventional government recovery interventions. Early action resulted in double the amount of social benefit compared to a late response and is especially beneficial to low-income households.

This paper recommends the creation of the Australian Government Anticipatory Action Fund (AGAAF) to finance anticipatory cash transfers. Compared to current funding mechanisms such as the Disaster Recovery Funding Arrangements, the AGAAF would prioritise disaster preparedness, aligning with recommendations from the 2020 Royal Commission into National Natural Disaster Arrangements. The AGAAF is expected to cost $260 million annually from 2024 and save approximately $3 billion per year in recovery costs (based on an 11:1 ROI estimate from the CSIRO). An additional $3 million is required to implement the software infrastructure to support a national anticipatory action program with oversight from the Department of Home Affairs via the National Emergency Management Agency.

Problem Identification

The cost of disasters triggered by natural hazards in Australia is projected to rise to at least $73 billion per year by 2060, assuming a low emissions scenario (Australian Business Roundtable for Disaster Resilience & Safer Communities, 2021). This is exacerbated by Australia’s high exposure to disasters relative to the Oceania region and the world (Bündnis Entwicklung Hilft & IFHV, 2023). Australia’s exposure to disasters is expected to cost homeowners $2,500 on average every year by 2050 (Mckell Institute, 2022), motivating the need for investment in disaster resilience. 

Although the Australian government currently focuses on disaster recovery, international models have demonstrated the potential to finance households in anticipation of a disaster. Cost-benefit analysis estimates vary, however, Atkinson (2018) estimates that every pound spent on early action results in 2.58 pounds in social value (as measured by the nutrition of children and livestock health for example) gained compared to a late response. This is also in conjunction with the widely accepted stance from the Commonwealth Scientific and Industrial Research Organisation (CSIRO), which states that a $1 investment in disaster risk reduction saves between $2 and $11 in post-disaster recovery and reconstruction (CSIRO, 2020). The potential to fund at-risk households is especially evident in historically low socio-economic areas. Although a dynamic situation, in flood-prone regions such as Lismore, 60% of residents live in disadvantaged neighbourhoods (Rolfe et al., 2020). Anticipatory financing supports poorer households to purchase necessities such as non-perishables or petrol, which may become unaffordable after a price rise from a disaster-induced shock. 

Context

Literature Review

Anticipatory action (AA) is defined as a set of planned and pre-financed measures taken before a disaster (Wilkinson et al, 2020). AA is also commonly referred to as Forecast-based Early Action (FbEA), Forecast-based Financing (FbF) and Early Warning Early Action (EWEA). Compared to investment after the peak of a disaster, AA has numerous theoretical advantages. In theory, AA widens the set of coping strategies for households such as evacuating or stocking up on food which may be hindered after a disaster that imposes price inflation and personal trauma (Pople et al., 2021).

Current literature exhibits an overall consensus towards the efficacy of AA (Weingärter et al., 2020). Studies in Bangladesh indicate that households that received funding in the days before a disaster experienced less debt accrual, higher food consumption, a higher chance of evacuating and were less susceptible to psychological damage (Gros et al., 2019; Pople et al., 2021). Where households received cash, nearly 90% was spent before the hazard peak indicating additional resources were used immediately in support of preparatory actions and coping strategies (Gros et al., 2019). According to Cabot Venton et al. (2013), it is estimated that early action in response to an incorrectly forecasted crisis can be made two to six times before the total cost would be equivalent to a single late response. In Nepal, the World Food Programme estimates for every dollar invested in AA, US$34.39 is saved in asset damage (WFP, 2019). Overall, AA promotes a shift in mentality from reaction to anticipation (Poterie et al., 2023).

Current Australian Policy

The Australian policy landscape does not currently enable anticipatory cash transfer arrangements. Current funding measures are perceived to be uncoordinated, under-resourced and untimely (Commonwealth of Australia, 2023). A 2022 survey found only 35% of community leaders felt funding had sufficiently supported their efforts to address emergencies confronting their communities (Cortis & Blaxland, 2022). Vulnerable communities, especially those with low financial capacity and increasing remoteness, bear the highest cost of the shortfalls in the current policy landscape (National Resilience Taskforce, 2018; Parsons et al., 2021).

There is a significant perception in Australia that disaster mitigation is consistently neglected relative to disaster response and recovery (Bradshaw et al., 2023; Parliament of Australia, 2023; Productivity Commission, 2014) and subsequent climate disasters are inundating long-term recovery measures. A recent example of this is demonstrated in those who were made homeless by the 2019/20 Black Summer Fires and subsequently had their temporary accommodation flooded in 2020 (Climate Council of Australia, 2022). Alternative recovery mechanisms such as the ‘recovery centre model’ exclude large groups of disaster-affected individuals by assuming that those who need assistance will go and seek it. Following a disaster, individuals may be reluctant or unable to travel to recovery centres for a range of reasons including trauma, loss of vehicles, a lack of financial resources to pay for fuel, or social isolation (Commonwealth of Australia, 2020).

Australia’s emergency financing arrangements are operated by The Australian National Emergency Management Agency (NEMA) under the Australian Department of Home Affairs. This is complemented by the National Disaster Risk Reduction Framework which guides Australian policy efforts on disaster risk reduction. The framework provides a comprehensive approach to proactively reduce disaster risk and aims to establish a long-term vision aligned with the 2030 Sustainable Development Goals (Department of Home Affairs, 2018b). 

The Disaster Ready Fund (DRF) is a disaster resilience and risk reduction initiative to deliver projects to support Australia’s management of the physical and social impacts of disasters. From July 2023, up to $1 billion has been allocated to the DRF for infrastructure and systemic risk reduction projects. In June 2023, $1.5 million was awarded to Emergency Management Victoria to integrate satellite camera technology and machine learning to provide rapid identification of bushfire ignitions. The fund is active federally, but projects are state and territory-based and it does not provide funding for disaster resilience to households directly. (National Emergency Management Agency, n.d.-a)

The Disaster Risk Reduction Package (DRRP) is jointly funded by the Australian Government and state and territory governments that support projects aligned to the National Disaster Risk Reduction Framework. The five-year agreement activated in 2019 comprises $104.4 million to support disaster risk reduction in the states and territories and $26.1 million for disaster risk reduction projects of national significance. Projects identified as ‘nationally significant’ are selected in consultation with Australian state and territory agencies (National Emergency Management Agency, n.d.-b).

The Disaster Recovery Funding Arrangements (DRFA) allows states to activate relief and recovery assistance immediately following a disaster such as emergency food or temporary accommodation for individuals (Department of Home Affairs, 2018b). Once the fund has been activated, the Australian Government may reimburse states and territories in some circumstances up to 75 per cent of the financial assistance provided. The 2020 Royal Commission into National Natural Disaster Arrangements outlines recommendations to facilitate resilience but admits that the DRFA does not currently incentivise or prioritise resilience measures (Commonwealth of Australia, 2020).Aside from financing AA, building a nationwide AA scheme will require infrastructure to support forecasting and a reliable method to contact at-risk households based on forecasting triggers. The Australian government has committed millions of dollars to disaster monitoring and forecasting through the DRRP and DRF, including a national bushfire intelligence capability project sponsored by the DRRP, a state-wide disaster-ready imagery project in Queensland and a project to standardise bushfire predictions in Western Australia financed in round one of the DRF (National Emergency Management Agency, n.d.-a, n.d.-b). This funding supplements the already established national warning system for all vectors of disaster (Australian Institute for Disaster Resilience, n.d.). By late 2024, the government will also deploy the National Messaging System (NMS) to target real-time messaging during emergencies (National Emergency Management Agency, n.d.-c). The NMS is a response to recommendations from the Royal Commission on the government's ability to deliver emergency messages swiftly (Commonwealth of Australia, 2020). 

Options

To establish an AA initiative in Australia, a policy mechanism must be established to operationalise and finance anticipatory cash transfers. Policy options for pre-defined financing are listed below:

1. Amend the Disaster Recovery Funding Arrangements to include a new category of measures relating to anticipatory actionImplementation requires appending a new category for subsidising AA, specifically, anticipatory cash transfers. The estimated budget impact of this option per year from 2024 is $195 million. Consolidating the DRFA centralises all intergovernmental funding agreements relating to disaster relief into a single policy mechanism.

2. Develop the “Australian Government Anticipatory Action Fund” to subsidise anticipatory cash transfers fullyAs opposed to an intergovernmental financial agreement, the Australian Government Anticipatory Action Fund (AGAAF) finances households directly through AA. Option 2 provides federal funding consistency for households, but incurs a higher budgetary impact due to covering 100 per cent of the payment. The estimated budget impact of this option per year from 2024 is $260 million.

3. Develop a voucher-based system in collaboration with a coalition of industry partnersTo mitigate inflation concerns arising from a pure cash transfer, a voucher-based system should be established with a coalition of major industry partners, specifically grocers and petrol providers. However, option 3 does not account for a diversity of essentials since households would be restricted to a certain basket of goods. The estimated budget impact of this option per year from 2024 is $260 million.

4. Develop a scheme to deliver supplies to households in anticipation of a disasterTo further avoid the risks imposed by a direct cash transfer, high-demand goods such as food, batteries or water can be delivered to disaster-prone households. Option 4 offers the greatest control and security to the government, however, it imposes the lowest range of choices available to households. The estimated budget impact of this option per year from 2024 is $260 million.

Policy Cost Estimation

From Hutley et al. (2020), 520,950 properties will be categorised as “high risk” by 2030. Given a subsidy of $500 (see “Policy Recommendation” section) and assuming all “high-risk” households experience at least one disaster shock that triggers AA per year, the budget impact estimate for each option is calculated below:

Option 1: 520,950  (500  0.75) = $195,356,250 (assume 75% rebate)Option 2: 520,950  500 = $260,475,000Option 3: 520,950  500 = $260,475,000Option 4: 520,950 * 500 = $260,475,000

Policy Recommendation 

Since AA targets Australian households directly, it would be most efficient for the federal government to finance the initiative directly. Hence, Option 2, “Develop the “Australian Government Anticipatory Action Fund” to subsidise anticipatory cash transfers fully” is most suitable. The AGAAF is a federally backed “trigger-based” fund for AA-related projects. The fund is activated and cash is transferred to eligible households where forecasts predict a statistically significant risk of a high-impact disaster triggered by natural hazards affecting the household. 

Eligibility is defined as low-income households in areas vulnerable to a high risk of natural hazards.  Low income is defined by the Australian Bureau of Statistics as a household in the lowest equivalised disposable income quintile, adjusted to exclude the first and second percentiles. 

The size of the cash transfer is $500 based on the average monthly grocery spend from consumer surveys (Bradney-George, 2024) and approximately covers emergency essentials such as petrol, food, water, portable radio, torch and/or first aid kit. Grocery spending is used as a proxy for the true allocation of cash before a disaster. This is congruent with evidence from Gros et al. (2019) demonstrating that cash is most frequently allocated to food. However, further consultation with a diverse range of disaster-affected households will be required to discover a suitably sized cash transfer for the Australian context. 

Implementation

In conjunction with the AGAAF, it is recommended to use the DRRP to finance the development of the software infrastructure to support the AA initiative. This includes a portal to onboard at-risk households into the AA program and to develop a trigger-based forecasting system to activate the fund. A portal should be established to pre-verify non-fraudulent households to minimise the delay in cash transfers once the fund is activated. The portal includes a form to collect relevant data such as the household's preferred bank details and phone number. Other relevant data includes household income to verify that the household meets the eligibility criteria.

The criteria to activate the AGAAF are defined by probabilistic forecast triggers. For example, if a bushfire is predicted to impact an eligible household in two days, the AGAAF should be immediately triggered. This would initiate the process for the household to be notified and cash transferred to their preferred bank account with additional emergency instructions.

Defining triggers and forecasts requires the most consideration. The Australian Warning System can act as a trigger for the AGAAF. More specifically, the “Watch and Act (Orange)” warning level.  However, although the Australian Warning System provides national consistency, it does not forecast disasters. Forecasts will likely derive from state-level forecasting systems. For example, the NSW Rural Fire Service provides fire danger rating predictions. Similarly, the Bureau of Meteorology issues flood warnings over 6 hours in advance of minor to major flooding. Trigger specification and the exact forecasting systems used require greater consultation with subject matter experts.

This implementation likely requires a national education initiative to educate at-risk Australians on AA, particularly, how they can register for funding from the AGAAF to receive funds preemptively. Fraudulence is also a large risk which can be offset by increasing the amount of verification for households to enter the program.

Funding

The Department of Home Affairs, via NEMA, is best suited to implement and take responsibility for both the AGAAF and the AA program. Developing the AA program is estimated to cost $3 million (based on similarly sized projects from the DRRP) using funding from the DRRP. The AGAAF is estimated to cost $260 million per year from 2024.

Success Criteria

The success of AA is measured in comparison to a regular government intervention during and after the peak of a disaster.

The policy recommendation is considered successful if there is a measurable statistical difference from regular government intervention on the following criteria (based on “The evidence base on anticipatory action” by Weingärter et al., 2020):

• Protected lives and livelihoods

• Avoided household losses

• Reduced emergency response costs

• Avoided or mitigated physical and psychological suffering

• Protected food security and nutrition

Evaluating the AGAAF and AA program would be conducted by the Australian National Audit Office (ANAO). 

Risks

Providing funding during and after days of a disaster is guaranteed to provide aid to affected households. However, there is an associated risk towards providing funds to households where it is uncertain whether they will be affected by a disaster resulting in an inefficient allocation of the AGAAF and an unnecessary psychological shock to households.

Anticipatory cash transfers to households assume the incentive will be spent entirely in the short term before the disaster peak. However, there is a risk that households do not use the cash for disaster preparedness either by purchasing subjectively irrelevant goods or by saving the cash for a later date post-disaster. Both outcomes result in an inefficient allocation of the AGAAF and could pose a minor inflationary risk to the Australian economy.

References

Australian Institute for Disaster Resilience. (n.d.). Australian Warning System. Retrieved January 14, 2024, from https://knowledge.aidr.org.au/resources/australian-warning-system/ 

Atkinson, E. (2018). Social Cost Benefit Analysis of the Early Action Fund.

Bradney-George, A. (2024, February 6). Average cost of groceries per month. finder.com.au. https://www.finder.com.au/budgeting/average-grocery-bill 

Bradshaw, S., Gardner, J., Gergis, J., & Blashki, G. (2023). Climate Trauma: The growing toll of climate change on the mental health of Australians. Climate Council of Australia.

Bündnis Entwicklung Hilft & IFHV. (2023). The WorldRiskReport 2023. 

Cabot Venton, C., Shitarek, T., Coulter, L., & Dooley, O. (2013). The Economics of Early Response and Resilience.

CSIRO. (2020). Climate and Disaster Resilience.

Climate Council of Australia. (2022). The Great Deluge: Australia’s new era of unnatural disasters.

Commonwealth of Australia. (2020). Royal Commission into National Natural Disaster Arrangements.

Commonwealth of Australia. (2023). Select Committee on Australia’s Disaster Resilience. https://parlinfo.aph.gov.au/parlInfo/download/committees/reportsen/RB000194/toc_pdf/InterimReport.pdf 

Cortis, N., & Blaxland, M. (2022). Carrying the costs of the crisis: Australia’s community sector through the Delta outbreak. Sydney: ACOSS.

Deloitte. (2021). Special report: Update to the economic costs of natural disasters in Australia. http://australianbusinessroundtable.com.au/assets/documents/Special%20report%3A%20Update%20to%20the%20economic%20costs%20of%20natural%20disasters%20in%20Australia/Special%20report%20_Update%20to%20the%20economic%20costs%20of%20natural%20disasters%20in%20Australia.pdf 

Department of Home Affairs. (2018a). Disaster Recovery Funding Arrangements 2018.

https://www.disasterassist.gov.au/Documents/Natural-Disaster-Relief-and-Recovery-Arrangements/disaster-recovery-funding-arrangements-2018.pdf 

Department of Home Affairs. (2018b). National Disaster Risk Reduction Framework.  Commonwealth of Australia.

Gros, C., Bailey, M., Schwager, S., Hassan, A., Zingg, R., Uddin, M. M., . . . Coughlan de Perez, E. (2019). Household-level effects of providing forecast-based cash in anticipation of extreme weather events: Quasi-experimental evidence from humanitarian interventions in the 2017 floods in Bangladesh. International Journal of Disaster Risk Reduction, 41, 101275. https://doi.org/https://doi.org/10.1016/j.ijdrr.2019.101275 

Hutley, N., Dean, A., Hart, N., & Daley, J. (2022). Uninsurable Nation: Australia's most climate-vulnerable places. Climate Council of Australia.

Lefebvre, M., & Reinhard, J. (2022). The Cost of Extreme Weather.

National Emergency Management Agency. (n.d.-a). Disaster Ready Fund (DRF) Guidelines. Retrieved January 14, 2024, from https://nema.gov.au/sites/default/files/inline-files/2023%20DRF%20Guidelines.pdf 

National Emergency Management Agency. (n.d.-b). Disaster Risk Reduction Package. Retrieved January 14, 2024, from https://nema.gov.au/programs/disaster-risk-reduction-package  

National Emergency Management Agency. (n.d.-c). National Messaging System. Retrieved January 14, 2024, from https://nema.gov.au/about-us/budget-2023-24/National-Messaging-System 

National Resilience Taskforce. (2018). Profiling Australia’s Vulnerability. Commonwealth of Australia.

Parsons, M., Reeve, I., McGregor, J., Hastings, P., Marshall, G. R., McNeill, J., ... & Glavac, S. (2021). Disaster resilience in Australia: A geographic assessment using an index of coping and adaptive capacity. International Journal of Disaster Risk Reduction, 62, 102422.

Pople, A., Hill, R., Dercon, S., & Brunckhorst, B. (2021). Anticipatory Cash Transfers In Climate Disaster Response. https://www.disasterprotection.org/publications-centre/anticipatory-cash-transfers-in-climate-disaster-response 

Productivity Commission. (2014). Natural Disaster Funding Arrangements: Inquiry Report No. 74. Canberra.

Rolfe, M. I., Pit, S. W., McKenzie, J. W., Longman, J., Matthews, V., Bailie, R., & Morgan, G. G. (2020). Social vulnerability in a high-risk flood-affected rural region of NSW, Australia. Natural Hazards, 101(3), 631-650. https://doi.org/10.1007/s11069-020-03887-z 

Tozier de la Poterie, A., Castro, E., Rahaman, H., Heinrich, D., Clatworthy, Y., & Mundorega, L. (2023). Anticipatory action to manage climate risks: Lessons from the Red Cross Red Crescent in Southern Africa, Bangladesh, and beyond. Climate Risk Management, 39, 100476. https://doi.org/https://doi.org/10.1016/j.crm.2023.100476

Weingärtner, L., Pforr, T., & Wilkinson, E. (2020). The evidence base on anticipatory action.

Wilkinson, E., Pforr, T., & Weingärtner, L. (2020). Integrating ‘anticipatory action’ in disaster risk management.

World Food Programme. (2020). Forecast-based Financing in Nepal - A Return on Investment Study.