Climate worry reduces farmer well-being

Abstract

Climate anxiety and worry about a changing climate have the potential to reduce individual well-being. We test for this possibility using a national survey sample of farmers, foresters, and growers in Aotearoa New Zealand. This group is of particular interest because changing climate has the potential to radically change their commercial operations. We find that survey respondents who express climate worry report substantially lower subjective well-being. Our estimates are robust to the inclusion of a wide range of controls. Our findings point to the importance of mitigating and adapting to climate change for well-being.

KEYWORDS:

• Climate change
• climate anxiety
• agriculture
• subjective well-being
• Aotearoa New Zealand

1. Introduction

Primary industry is critically important to Aotearoa New Zealand (NZ). In 2023, for example, the primary sector was responsible for over $54 billion in export revenue, 10.5% of national GDP, and 13.1% of total employment (Ministry for Primary Industries, 2023).

NZ’s primary industry largely consists of dairy and sheep and beef farming, forestry, and horticulture. Changing climate potentially affects the productivity and profitability of these sectors via changing precipitation and temperature and increased incidence of extreme weather events (e.g. Kurukulasuriya & Rosenthal, 2013; Mearns, Easterling, Hays, & Marx, 2001; Motha, 2011; Wreford, Moran, & Adgler, 2010). Changing rainfall patterns will increase drought risk across much of the country (Tait, Baisden, Wratt, Mullan, & Stroombergen, 2008), resulting in greater price volatility and other economic stressors (Pourzand, Noy, & Sağlam, 2020). Changing temperatures may affect the suitability of land for growing crops such as grapes for wine production, leading to changes in cultivation (Aussiel et al., 2021) and potentially increased land-use change (e.g. Anastasiadis, Kerr, Zhang, Allan, & Power, 2014). While these changes may increase the productivity and profitability of primary industries in some areas, potential negative impacts and uncertainty around those impacts may also cause worry among NZ’s farmers, foresters, and growers (Booth, Brown, & Walsh, 2018).

Fritze, Blashki, Burke, and Wiseman (2008) and McBride, Hammond, Sibley, and Milfont (2021) describe pathways by which the global environmental threat of climate change may create emotional distress and worry, lowering well-being and potentially even impairing everyday functioning. Ogunbode et al. (2022) have quantified the negative impact of climate anxiety on well-being among the general population of 32 countries using the World Health Organisations ‘WHO-5’ inventory of well-being; they find that climate anxiety lowers well-being in 31 of 32 countries studied, with an average reduction of 1.2 points on the 0–25 scale. The effect of climate anxiety was most pronounced in Chile and Palestine and weakest in Japan (where it did not have a measurable effect on well-being), Norway, Romania, Spain, and the US. The only study to focus on climate anxiety and well-being in NZ of which we are aware uses a different but related outcome (specifically, the Kessler-6 measure of psychological distress), finding that climate anxiety raises psychological distress among the general population by 2.4% of the mean score (McBride et al., 2021).

Several recent papers have documented the relationship between extreme weather events and well-being among primary producers. For example, Xu, Wheeler, and Zuo (2023) show that droughts in the Murray-Darling Basin in Australia resulted in elevated suicide rates among farmers. Hrabok, Delorme, and Agyapong (2020) extends the analysis to climate worry, qualitatively describing how Canadian farmers who are reliant on climate-impacted resources face lower well-being. However, to our knowledge, no existing studies have quantitatively evaluated the relationship between climate worry and well-being among farming populations, for whom climate is most closely linked to business performance.

This paper seeks to fill this knowledge gap using a survey of 1,682 commercial farmers, foresters, and growers from across NZ. We find that climate worry has a strong negative impact on well-being, reducing subjective well-being scores by 4.8% at the mean. This result is robust across empirical specifications.

This research note is organised as follows. Section 2 describes our survey data and Section 3 describes our econometric approach. Section 4 presents our evidence that climate worry negatively impacts well-being. Section 5 concludes.

2. Data

The Survey of Rural Decision Makers (SRDM) collects detailed information on topical issues in NZ’s primary industry. The 2021 wave of this biennial survey emphasised land use and land-use change, management practices, personal values, well-being, and future climate. The survey was open from 1 June until 15 August. Respondents represent the entirety of NZ’s agricultural sector from Cape Reinga to Half Moon Bay. Demographics are representative of the primary sector as a whole (Stahlmann-Brown, 2021).

To analyse how climate worry impacts well-being within the primary sector, we follow Knook, Dorner, and Stahlmann-Brown’s (2022a) measure of climate worry among farmers, in which survey respondents were asked ‘What impact do you anticipate changing climate will have on your operation between now and 2050, on balance, if any?’ Respondents could choose ‘very negative impact’, ‘somewhat negative impact’, ‘positive and negative impacts balance’, ‘somewhat positive impact’, ‘very positive impact’, or ‘unsure’. We classify respondents who responded ‘very negative impact’ or ‘somewhat negative impact’ as having ‘climate worry’.

Well-being is measured via the World Health Organisation’s WHO-5 inventory, a short questionnaire of five simple, non-invasive questions. The WHO-5 has been among the most widely used measures of well-being since its initial publication, and Topp, Østergaard, Søndergaard, and Bech’s (2015) systematic review demonstrates its high validity across a wide range of study fields.

Specifically, the WHO-5 inventory asks how often respondents felt each of the following over the preceding two weeks: ‘I have felt cheerful and in good spirits’; ‘I have felt calm and relaxed’; ‘I have felt active and vigorous’; ‘I woke up feeling fresh and relaxed’; and ‘my daily life has been filled with things that interest me’. Each statement is scored on a scale of 0 (‘at no time’) to 5 (‘all of the time’), and an aggregated score is calculated. Aggregate scores of 12 and below reflect ‘poor well-being’ while those above 21 reflect ‘high well-being’.

Table 1 provides descriptive statistics from the survey data. Among the 1,682 respondents who answered the WHO-5 questions (i.e. 99.4% of the entire sample of commercial farmers, foresters, and growers), the average WHO-5 subjective well-being score was 15.5, although the range included both extreme low well-being (a score of 0) and extreme high well-being (a score of 25). This average score is in line with the score of 15.9 for the general population of NZ (IPSOS, 2021). Approximately one-third of survey respondents reported that they felt that changing climate would have a ‘very negative’ or ‘somewhat negative’ impact on their operations over the next three decades.

Table 1. Descriptive statistics.

	Obs.	Mean	Std. Dev.	Min	Max
WHO-5 subjective well-being	1,682	15.52	5.24	0	25
Climate worry	1,692	0.34	0.47	0	1
Female	1,692	0.25	0.43	0	1
Education	1,692
Secondary or less		0.33
Certificate or diploma		0.17
University or higher		0.50
Age	1,692	58.88	11.97	20	90
Industry	1,692
Sheep and beef		0.48
Dairy		0.28
Other livestock		0.04
Arable		0.03
Horticulture		0.10
Forestry		0.07
Property size in hectares	1,692	558.69	2133.80	0.10	45,000
Profitability	1,692	0.55	0.40	0	1

The sample may further be described by gender (with 25% of respondents identifying as female), education (with 33% of respondents having secondary educations or less, 17% holding post-graduate certificates or diplomas, and 50% holding university educations or higher degrees), and age (with a mean of 59). Nearly half of respondents are commercial sheep and beef farmers, 28% are commercial dairy farmers, 10% are commercial horticulturalists, and 7% are commercial foresters. Arable farmers and farmers specialising in other types of livestock make up the rest of the sample. The average farm size is large at 559 hectares, although there is considerable variation as commercially viable kiwifruit farms are often a few acres or smaller and large sheep runs may comprise tens of thousands of hectares. Some 55% of respondents described their operations as being profitable.

3. Methods

We use Ordinary Least Squares¹ to estimate. $\mathit{y}=\mathit{X}\beta +ϵ$where $\mathit{X}$ is a vector of variables including the gender, education, and age of the respondent, a series of industry dummies, the log of property size, a dummy indicating whether the operation is profitable, and regional fixed effects in addition to a dummy indicating climate worry. Controls are added iteratively to the model. Errors are clustered at the region level.

4. Results

Table 2 reports the average marginal effects for all covariates from the OLS model. Survey respondents who expressed climate worry score 0.68–0.76 points lower on the WHO-5 measure of well-being, statistically significant at the 1% level. This reduction in subjective well-being associated with climate worry is meaningfully large at approximately 4.8% of the mean well-being score in our sample.

Table 2. Estimation results.

	(1)	(2)	(3)	(4)
Climate worry	−0.711***	−0.678***	−0.724***	−0.759***
	(0.226)	(0.207)	(0.218)	(0.195)
Female		−1.328***	−1.220***	−1.145***
		(0.290)	(0.284)	(0.300)
Education: Cert.		0.718*	0.802**	0.789**
or diploma		(0.357)	(0.332)	(0.315)
Edu: University		0.993**	0.877**	0.887**
Or higher		(0.386)	(0.385)	(0.355)
Age		−0.121	−0.128	−0.113
		(0.0985)	(0.0988)	(0.102)
Age squared		0.00154*	0.00153*	0.00134
		(0.000860)	(0.000870)	(0.000900)
Industry: Dairy			−1.129***	−1.449***
			(0.334)	(0.244)
Industry: Other			0.802*	0.490
livestock			(0.402)	(0.468)
Industry: Arable			−0.594	−0.463
			(0.774)	(0.657)
Industry:			−0.601	−1.475**
Horticulture			(0.553)	(0.603)
Industry: Forestry			1.415***	1.261***
			(0.403)	(0.392)
Property size				−0.358***
in hectares (log)				(0.0743)
Profitability				1.625***
				(0.366)
Constant	16.68***	17.79***	18.48***	19.52***
	(0.0836)	(2.643)	(2.591)	(2.551)
Regional FE	YES	YES	YES	YES
Observations	1,682	1,682	1,682	1,682

Notes: ***p < 0.01; **p < 0.05; *p < 0.10.

We additionally find that women have significantly lower subjective well-being than men. Higher education is associated with higher well-being and that the relationship between age and well-being is weakly convex. Dairy farmers and horticulturalists exhibit lower subjective well-being than sheep and beef farmers (the omitted industry group) while foresters and those with pigs, deer, and other forms of livestock exhibit higher subjective well-being. Profitability is positively associated with well-being while property size is negatively associated with well-being.

5. Discussion and conclusion

Changing climate may provoke worry or anxiety, potentially reducing psychological well-being, as has been demonstrated for the general population in more than 30 countries (Ogunbode et al., 2022), including NZ (McBride et al., 2021). Given that climate has direct impacts on the livelihoods of people working in the primary sector, this paper focuses on farmers, foresters, and growers. This is particularly relevant in NZ, given the centrality of primary industry to NZ’s identity and economy.

We find that farmers, foresters, and growers who believe that changing climate will adversely affect their commercial operations have lower well-being than those who believe that climate change will have no effect or a positive effect on their operations. These effects are sizable, with climate worry reducing well-being scores by 0.68–0.76 points on the 0–25 WHO-5 scale, roughly 4.8% of the mean score.

Our estimates are smaller than those reported in Ogunbode et al. (2022), who find a 1.2 point (approximately 8.8% of the mean score) reduction in well-being on the 0–25 point WHO-5 scale, on average, among the general population in 31 countries under study. However, our estimates are larger than the estimated effect of climate anxiety on psychological distress among the general population of NZ (2.4% of the mean score) reported in McBride et al. (2021). We note that both comparisons must be made with caution because our measure of climate worry focuses narrowly on the impact of changing climate on respondents’ businesses, whereas the other studies use a wider definition; for example, the Ogunbode et al. (2022) asked respondents to report their emotional response to climate change, noting the extent to which they felt ‘tense’, ‘anxious’, ‘worried’, ‘terrified’, ‘calm’, ‘relaxed’, and ‘peaceful’.

Hrabok et al. (2020) observe that both worry about future climate and the impact of contemporaneous climate-related disasters may impact well-being. Indeed, during the time that the SRDM was open, NZ experienced three major adverse weather events – severe flooding in Canterbury at the end of May and beginning of June 2021; major storms and a rare tornado in Auckland on 18 June 2021; and extreme rain in Nelson, Tasman, and Marlborough between 15 and 18 July 2023 (the first and third of these resulted in states of emergency). Therefore, to help distinguish the effects of future climate worry from the effects of increasingly common climate-related disasters, we replace the region fixed effects in our previous estimates with a dummy indicating that the respondent completed the survey in an affected region after the date of these disasters²; just over 18% of our sample meet these criteria.

Results are presented in Table 3. Respondents completing the survey after a major climate-related disaster in their region express lower well-being than other respondents, on average. This effect is statistically significant at the 0.05 level in every specification apart from the final specification, when it falls just short of significance at the 0.10 level. Regardless, climate worry continues to have a sizeable negative effect on well-being, statistically significant at the 0.01 level across specifications That is, the impact of long-term climate worry on well-being functions independently of recent climate-related disasters.

Table 3. Estimation results with an indicator of exposure to extreme weather events.

	(1)	(2)	(3)	(4)
Climate worry	−0.663***	−0.644***	−0.693***	−0.726***
	(−0.209)	(0.193)	(0.210)	(0.181)
Affected by weather	−0.555*	−0.531**	−0.542*	−0.450
	(0.267)	(0.246)	(0.280)	(0.287)
Female		−1.313***	−1.217***	−1.154***
		(0.294)	(0.287)	(0.303)
Education: Cert.		0.773**	0.838**	0.823**
or diploma		(0.354)	(0.345)	(0.333)
Edu: University		1.006***	0.892**	0.915***
Or higher		(0.386)	(0.388)	(0.360)
Age		−0.120	−0.123	−0.111
		(0.0986)	(0.101)	(0.103)
Age squared		0.00156*	0.00153	0.00136
		(0.000854)	(0.000882)	(0.000911)
Industry: Dairy			−0.851**	−1.247***
			(0.291)	(0.242)
Industry: Other			0.764*	0.380
livestock			(0.387)	(0.432)
Industry: Arable			−0.503	−0.422
			(0.813)	(0.676)
Industry:			−0.339	−1.315***
Horticulture			(0.593)	(0.593)
Industry: Forestry			1.160***	0.975**
			(0.357)	(0.346)
Property size				−0.399***
in hectares (log)				(0.0668)
Profitability				1.552***
				(0.369)
Constant	15.84***	16.97***	17.50***	18.72***
	(0.174)	(2.745)	(2.679)	(2.537)
Observations	1,682	1,682	1,682	1,682

Notes: ***p < 0.01; **p < 0.05; *p < 0.10.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This research was supported by Strategic Science Investment Funding (SSIF) for Crown Research Institutes from the Ministry of Business, Innovation and Employment’s Science and Innovation Group.

Notes

1 Results obtained from a tobit model that allows for censoring are substantively identical to those obtained from OLS, so we opt to present the simpler OLS results.

2 We are grateful to an anonymous referee for raising this suggestion.