The Impact of Economic And National Identity Loss Messages, And The Moderating Effect of Political Orientation, On Climate Change Policy Support

As climate change continues to be politically divisive, developing communications that 7 align with right-leaning beliefs may increase bipartisan support for climate policy. In two 8 experimental studies (Study 1, Australia, N = 558; Study 2, United States, N = 859), we tested 9 whether an economic loss or national identity loss message would elicit greater support for 10 mitigation and adaptation policies when compared to one another and to a control message. We 11 also tested whether the direct effects of these message types were conditional on political 12 orientation (specifically, identifying as politically right-leaning). In both studies, preliminary 13 analyses indicated that the message manipulations were effective and that there was a high level 14 of support for both types of climate policy. When compared to left-wing adherents, those who 15 were politically right-leaning were less likely to support mitigation and adaptation policies in 16 either sample. Australian (Study 1) identification – although not American identity (Study 2) – 17 also uniquely predicted adaptation support (but not mitigation support). Yet, there were no 18 significant message frame or interaction effects in the Australian (Study 1) or U.S. sample (Study 19 2). This suggests that neither an economic loss nor national identity loss message frame may be 20 effective in overcoming the political polarization of climate change in Australia or the United 21 States. Nevertheless, national identity could still play a useful role in Australian climate 22 communications given its positive relationship to adaptation policy support, and therefore 23 warrants further investigation.

Running head: LOSS FRAMES, POLITICAL ORIENTATION, AND CLIMATE CHANGE POLICY SUPPORT 8 identification, including right-wing authoritarianism (RWA) and social dominance orientation 141 (SDO) (as seen in the United States; Roccas et al., 2010). This loyalty and gratitude to one's 142 country is often expressed by promoting the interests of one's nation over and above the interests 143 of those external to it. It even appears to occur in environmental contexts with those who are 144 politically right-leaning more likely to be motivated to protect the environment if it means that it 145 will also protect national interests (Sapiains, Beeton, & Walker, 2016;McCright & Dunlap, 146 2014). 147 As well as aligning with conservative values, a national identity loss frame taps into 148 conservatives' higher sensitivity to loss when something they value culturally is under threat 149 (Feygina et al., 2010;Jost et al., 2003). For instance, in one study, Feygina et al. (2010) 150 investigated a national identity loss frame with a U.S. sample and found that by presenting pro-151 environmental messages as patriotic and environmental conservation as key to protecting and 152 preserving the "American way of life", it was possible to increase conservatives support for 153 general pro-environmental behaviours, such as recycling. Similarly, Wolsko et al. (2016) 154 demonstrated right-wing adherents in the U.S. are more likely to support climate change action 155 when exposed to a frame that outlined how protecting the natural environment is a matter of 156 obeying authority, defending the purity of nature, and patriotism (all conservative values). 157 Finally, research with Australians who rejected anthropogenic climate change showed that they 158 were more likely to support climate change policies after receiving a national identity gain 159 message when compared to biodiversity or economic gain messages (Sapiains et al., 2016). 160 While none of these findings examined a national identity loss frame, its comparison to 161 economic loss message specifically, and whether its effectiveness was conditional on political 162 Running head: LOSS FRAMES, POLITICAL ORIENTATION, AND CLIMATE CHANGE POLICY SUPPORT 9 orientation, they do point to the potential usefulness of developing message frames that target 163 conservative values of economic stability and nationalism for right-wing adherents. 164

The Present Research 165
The aims of this research were to investigate whether an economic loss or national 166 identity loss message increased climate change policy support when compared to one another 167 and to a control message, and whether their potential effectiveness was conditional on political 168 orientation (specifically, right-wing political orientation). We conducted two online experimental 169 studies to test these aims: the first in Australia (Study 1) and the second in the United States 170 (Study 2). Australia presents a ripe area to test these potential effects as the topic of climate 171 change continues to be politically polarizing and the country itself rates as one of the lowest 172 performing Western nations on climate policy (Sachs et al., 2021). Furthermore, Australia relies 173 heavily on fossil fuel exports for its Gross Domestic Product (GDP) (Department of the 174 Environment and Energy, 2018), even though Australians view their natural environment as a 175 core component of their national identity (Purdie & Wilss, 2007). This suggests Australians 176 could be especially sensitive to economic and national identity loss messages. Further still, the 177 United States has also experienced significant political polarization (Finkel et al., 2020), 178 especially regarding climate change (Dunlap et al., 2016). Therefore, a United States sample also 179 provides the opportunity to potentially replicate and extend upon any potential effects found in 180 Australia. 181 In both studies, we employed a between-subjects experimental research design, with one 182 factor (message type) that had three levels (economic loss, national identity loss, control). 183 Participants were randomly assigned to one of these three message conditions (i.e., levels). As 184 one of our message types threatened national identity (i.e., national identity loss message), we 185 The market research company PureProfile advertised the study to their participant pool in 206 July 20180 F 1 , actively targeting those who had indicated previously to them that they either self-207 identified as politically left-leaning (n = 276) or politically right-leaning (n = 282). To reduce 208 demand characteristics, participants were told that the study assessed their understanding of 209 science communication, with no reference made to climate change in recruitment materials. 210 After reading the plain language statement and providing consent, all participants completed the 211 pre-message measures and then were randomly allocated to receive one of the three messages 212 using the automatic randominsation feature in Qualtrics. Participants then completed the post-213 message measures. The study was 10 minutes and participants were paid $AUD5.30 in 214 compensation. 215

Pre-Message Measures. 216
Participants' age, gender, education, income, and state of residence was collected. 217 Participants' strength of Australian identity was measured using a single item ('I identify with  218 Australians', 1 = strongly disagree, = 7 strongly agree) (Postmes, Haslam, & Jans, 2013). A 219 single-item measure of political orientation was also employed ('In politics people sometimes 220 talk of the left-wing and right-wing. Where would you place yourself? 1 = very left-wing, 7 = 221 very right-wing)1 F 2 2F 3 . The higher one scored on this item, the more they identified as politically 222 right-leaning. 223

Message Frames. 224
Each message was a fictitious online newspaper article which detailed a new report 225 released by leading university scientists. The first message outlined that the national economy 226 would be threatened if effective climate policy was not implemented (economic loss message). 227 The second message outlined how Australia's natural icons would be jeopardized or lost, and 228 therefore Australia's unique cultural identity and way of life, would be under threat if effective 229 climate change policy was not implemented (national identity loss message). The third message 230 served as a control condition and described how antibiotic resistance was predicted to threaten 231 Australia's health outcomes (control message). See supplementary materials for a copy of the 232 message types. 233

Attention and Manipulation Checks 235
Using two items, we checked that participants read and understood their randomly 236 allocated message (Attention check 1: What issue did the newspaper article discuss?; Attention 237 check 2 = What negative impacts did the newspaper article discuss?). Participants who failed at 238 least one of these two attention checks were excluded from final analyses (n = 121) (see 239 supplementary materials for more details). We also asked participants to indicate the likelihood 240 that climate change would impact the outcome outlined in their allocated message to ensure the 241 manipulations were successful ('How likely is that climate will influence economic prosperity 242 /Australian identity/ /antibiotic effectiveness?') (1 = Not at all, 7 = A great deal). Analyses 243 indicated that the manipulations were effective (see supplementary materials for details and 244 results of these analyses). 245

Climate Change Policy Support 246
Bateman and O'Connor's (2016) two-factor climate change policy support scale 247 measured participants support for mitigation policies (e.g., "implementing a "carbon tax" on coal 248 and other fossil fuels to reduce their use", 6-items) and adaptation policies (e.g., "Setting aside 249 land corridors to help species migrate", 5-items) (1 = strongly disagree, 7 = strongly agree). 250

251
We used R (Version 4.0.2; R Core Team, 2020) and the R-packages apaTables (Version 252 2.0.5; Stanley, 2018), psych (Version 2.0.12; Revelle, 2018), tableone (Version 0.9.3; Yoshida & 253 Bohn., 2018), and interactions (Version, 1.1.3; Long, 2020) for our all analyses. 254 Table 1 presents the descriptives for the major continuous variables in Study 1. The 256 sample overall largely identified as Australian and demonstrated moderate to high support for 257 both mitigation and adaptation policy. There was a small, positive correlation between Australian 258 identity and political orientation, suggesting that identifying as an Australian was weakly related 259 to whether one was politically right-leaning. Australian identity had no relationship with 260 mitigation policy support, and a small, positive relationship with adaptation policy support.

Descriptives. 255
Right-wing political orientation was negatively related to both mitigation and adaptation policy 262 support, and moderately so in the case of mitigation policy. 263 264 265 Despite initial data collection resulting in even participant numbers across the three F(2,555) = .28, p = .753, partial η 2 = .00. Therefore, it appeared message randomisation was still 287 successful despite participant numbers differing slightly per condition post data cleaning. 288

Direct and Interactive Effects on Climate Change Policy Support in Australia. 289
Two moderated multiple regressions were conducted, one for each dependent variable. 290 We first started by creating two planned orthogonal contrasts to compare the effect the messages 291 frames had on the dependent variables. Firstly, to compare the intervention messages to the 292 control message, we created planned contrast 1, coding: .5 = economic loss message, .5 = 293 national identity message, -1 = control message. Then, to compare the intervention messages to 294 one another (while simultaneously controlling for the control message), we created planned 295 contrast 2: -1 = economic loss message, 1 = national identity loss message, 0 = control. To 296 examine the unique effect of the message frames, planned contrasts were entered in at Step 1. In 297 Step 2, we entered the covariates of Australian identity and political orientation. Finally, in Step 298 3, the two-way interactions between political orientation and message type (i.e., contrast 1, 299 contrast 2) were included. A sensitivity power analysis conducted post data collection suggested 300 Running head: LOSS FRAMES, POLITICAL ORIENTATION, AND CLIMATE CHANGE POLICY SUPPORT 16 that with a sample of 558, the final sample size would deliver 80% power to detect an r-square 301 of .025 or greater, with six predictors, at an alpha of .05. 302 For the outcome variable of mitigation policy support (Table 2), there was no significant 303 effect of message type on mitigation policy support. However, a significant negative effect of 304 political orientation emerged, providing support for H1. When controlling for Australian identity, 305 Australians who were politically right-leaning were less likely to support policies that aim to 306 limit the magnitude or rate of long-term climate change (e.g., emissions trading scheme, carbon 307 tax). However, there was no direct effect of Australian identity. We also obtained no support for 308 H2; there was no statistically significant political orientation x message type interactions 309 observed on mitigation policy support. 310 For the outcome variable of adaptation policy support (Table 3), there was no significant 311 effect of message type on adaptation policy support, once again suggesting that the message 312 frames did not influence the outcome variable. A significant negative effect of political 313 orientation emerged, providing further support for H1. Australians who were politically right-314 leaning were less likely to support policies intended to cope with the inevitable effects of climate 315 change, when also controlling for Australian identity. Unexpectedly, we found a significant 316 unique positive effect of Australian identity, in that those who identified more strongly as an 317 Australian were more likely to support adaptation policies. H2 was also not supported as we did 318 not observe any significant political orientation x message type interactions on adaptation policy 319 support.  Note. N = 558. β = standardized regression weights. sr 2 = semi-partial correlation squared. R 2 = coefficient of determination. ΔR 2 = change in R 2 . 95% CI = 95% confidence interval. Square brackets enclose the lower and upper limits of a 95% confidence interval. ** p < .01.

Discussion
Using an Australian sample, Study 1 investigated whether there was a relationship between climate change loss messages (i.e., economic and national identity loss) and policy support, and whether this relationship was moderated by right-wing political orientation. We also measured Australian identity to control for its potential impact on the major variables of the study. As hypothesized, political orientation negatively predicted climate policy support, with those who were politically-right leaning less likely to support mitigation and adaptation policies.
Consistent with emerging research, mitigation policies appeared to be more politically divisive in the Australian context (Bateman & O'Connor, 2016;Tranter & Lester, 2017), with the negative effect of political orientation stronger in the context of mitigation policies than adaptation policies.
Identifying as Australian was weakly related to political orientation, yet it did predict support for adaptation policy (but not mitigation policy) when controlling for message type and political orientation. Therefore, it appears that national identity may have different implications in the Australian context, such as not being tied strongly to being right-wing as is the case in the United States (Osborne et al., 2017;Roccas et al., 2010). Furthermore, the association between Australian national identity and support for adaptation policy (but not mitigation policy) may be partially explained by how Australians typically view environmental icons as crucial to their national identity (Bonaiuto, Breakwell, & Cano, 1996;Purdie & Wilss, 2007). While mitigation policies are focused on reducing the severity of carbon emissions, adaptation policies are typically oriented towards protecting and preserving natural environments via technological advancements, and this could include some of these environmental icons in Australia (e.g., protecting coastal areas like the Great Barrier Reef, protecting against bushfires in iconic areas). Nevertheless, neither the economic loss nor national identity loss messages influenced policy support, and the effectiveness of these messages was not conditional on political orientation.
Therefore, presenting Australians with messages which emphasize the potential threats to the economy or the potential loss of Australia's environmental icons, relative to a control condition, did not increase climate policy support. This was the case even for adaptation policies and for participants who were more politically right-leaning.
Overall, the mean of Australian identity was high, suggesting most participants largely identified as Australian. As we actively targeted those who had indicated previously that they either self-identified as politically left-leaning (n = 276) or politically right-leaning (n = 282), our sample also did not skew towards one end of the political spectrum. Furthermore, mitigation and adaptation policy support were reasonably high across all conditions. This suggests that there may have been a potential ceiling effect, in that we did not observe any meaningful differences between groups according to message type because participants on average supported mitigation and adaptation strategies. It is also possible that our sample contained individuals who were skeptical or denied climate change, as these participants were not screened out prior to receiving one of the messages. As noted above, prior research has demonstrated that those who are politically right-leaning in Australia and the United States are more likely to deny the existence of climate change (Clarke et al., 2019;Hornsey et al. 2018), with this political polarization especially strong in the United States (Finkel et al., 2020). Therefore, it can be argued that those who deny climate change and its impacts simply do not think it is an issue worth acting upon and are therefore less likely to attend to and process information which suggests climate policy should be enacted. This may partially explain why although participants understood the content of the national identity and economic loss messages (that is, the manipulations were successful), was happening, testing the same message types and potential interactions in this specific sample.

Study 2: The United States
In Study 2, we sought to test the same aims from Study 1, but this time in a sample from the United States who believed climate change is human-induced and occurring. We had the same hypotheses for Study 2 as those in Study 1, except this time we controlled for American identification. To minimize sampling bias, we actively targeted participants who had indicated previously on TurkPrime that they self-identified as either liberal (n = 460) or conservative (n = 399). Data was collected in July 2019, took 10 minutes, and participants were paid $US1 compensation. The Study 2 procedure was the same as Study 1, except that participants' climate change belief was asked before they randomly received one of the three messages.

Method
Running head: NATIONAL IDENTITY, POLITICAL ORIENTATION, AND

Measures.
We employed the same measures in Study 2 except for the changes outlined below.
Participants' strength of American identity was measured using the single-item measure ('I identify with Americans', 1 = strongly disagree, 7 = strongly agree) (Postmes et al., 2013). A single-item measure of political orientation was employed ('In politics people sometimes talk of liberals and conservatives. Where would you place yourself?', 1 = very liberal, 7 = very conservative). The messages contained the same content as Study 1 except that we replaced all Australian information with American relevant information, and manipulation checks indicated that the manipulations were once again effective (see supplementary materials for analyses).

Climate change beliefs
Belief that climate change is occurring was measured with a categorical item before participants were presented with their randomly allocated message (Greenhill, Leviston, Leonard, & Walker, 2014). Analyses revealed that believing climate change was occurring (either naturally or human induced) was associated with a significant, positive increase in mitigation and adaptation policy support when compared to those who denied climate change or were unsure it was occurring. Therefore, participants who indicated they did not believe in climate change or were unsure were excluded from the final analyses (excluded n = 83; more details can be found in the supplementary materials).

Results
We used the same R packages as in Study 1 to conduct all analyses.  Overall, participants mostly identified as American, and support for both mitigation and adaptation policies were high in the sample. There was a moderate, positive relationship between American identity and political orientation, suggesting that identifying as an American was positively associated with being conservative. While American identity had weak, negative relationships with both policy support types, political orientation had a strong, negative relationship with mitigation policy support, and a moderate, negative relationship with adaptation policy support.

Descriptives.
Like in Study 1, removal of participants who failed attention checks resulted in uneven participant numbers in the three message frame groups (N = 859, economic loss n = 294, national identity loss n = 278, control n = 287). Therefore, we also conducted two one-way ANOVAs to = .509, partial η 2 = .00. Therefore, it appeared randomisation was still successful despite participant numbers differing slightly per condition post data cleaning.

Direct and Interactive Effects on Climate Change Policy Support in the U.S.
We used the same data analytic approach as Study 1, conducting moderated multiple regressions and using planned contrasts to compare the effectiveness of the message types on both dependent variables. We again conducted a sensitivity power analysis post data collection.
A final sample size of 859 was found to deliver 80% power to detect an r-square of .016 or greater, with six predictors, at an alpha of .05.
For the outcome variable of mitigation policy support (Table 5), there was no significant effect of message type suggesting that the message frames did not influence this type of policy support. However, a significant negative effect of political orientation emerged, once again providing support for H1. Participants who identified as conservative and believed in climate change were less likely to support mitigation policies, when controlling for American identity.
There was no direct effect of American identity on mitigation policy support. We also obtained Running head: NATIONAL IDENTITY, POLITICAL ORIENTATION, AND CLIMATE CHANGE COMMUNICATIONS 25 no support for H2; there was no statistically significant political orientation x message type interactions observed on mitigation policy support.
For the outcome variable of adaptation policy support (Table 6), we did not find a significant direct effect of message frame. However, a significant negative effect of political orientation emerged, providing further support for H1. Americans who identified as conservative and believed in climate change were less likely to support adaptation policies, when controlling for American identity. Unlike Study 1, we did not unexpectedly find a direct effect of American identity on adaptation policy. Contrary to H2, there were no significant political orientation x message frame type interactions observed on adaptation policy support. Note. N = 859. β = standardized regression weights. sr 2 = semi-partial correlation squared. R 2 = coefficient of determination. ΔR 2 = change in R 2 . 95% CI = 95% confidence interval. Square brackets enclose the lower and upper limits of a 95% confidence interval. * p < .05. ** p < .01.  Note. N = 859. β = standardized regression weights. sr 2 = semi-partial correlation squared. R 2 = coefficient of determination. ΔR 2 = change in R 2 . 95% CI = 95% confidence interval. Square brackets enclose the lower and upper limits of a 95% confidence interval. * p < .05. ** p < .01.

Discussion
In Study 2, we collected a sample of United States citizens and/or residents who believed climate change was occurring to investigate the relationship between climate change loss messages and policy support, and to test whether this relationship was moderated by conservative political orientation. We also controlled for American identity. As hypothesized and consistent with Study 1, conservatives were less likely to support mitigation and adaptation policies, and this negative direct effect was stronger in the context of mitigation policies than adaptation policies (Bateman & O'Connor, 2016). However, unlike in Study 1, there was no direct effect of American identity on support for adaptation policies. Therefore, those identifying as an American were not more likely to support adaptation policies. We also found no significant political orientation x message type interaction on mitigation or adaptation policy support.
However, it must be noted that mitigation and adaptation policy support was once again high across all message conditions (as seen in Study 1). Given we only obtained participants who believed in climate change in Study 2, thereby accounting for the impact that climate change deniers may have had in Study 1, it is possible that some potential differences were reduced or not observed in this study due to ceiling effects.

General Discussion
The aims of this present research were twofold. Our first aim was to investigate whether an economic loss or national identity loss message increased climate policy support (mitigation, adaptation) when compared to one another and to a control message in Australia (Study 1) and the United States (Study 2). Secondly, as right-wing adherents are less likely to support climate change policy in these countries, but national identity is typically related to conservatism, we also aimed to test whether the effectiveness of these climate messages was conditional on identifying as politically right-leaning (Study 1 in Australia) or conservative (Study 2 in the U.S.). As one of our message types threatened national identity (i.e., national identity loss message), we also measured participants' level of national identification to control for its potential effect on the variables of the study.
Across the two studies, we found a direct, negative effect of political orientation on climate change policy support. When controlling for Australian (Study 1) or American (Study 2) identity, right-wing adherents were less likely to support mitigation and adaptation policies. This is consistent with previous research that has shown that those who are politically left-leaning in Australia and the United States are more likely to support climate change policies and action when compared to their right-wing counterparts (Fielding et al., 2012;Hornsey et al., 2018;Dunlap et al., 2016;Unsworth & Fielding, 2014). Interestingly, this negative relationship was more pronounced with mitigation policy support in both Australia (Study 1) and the United States (Study 2), suggesting that conservative respondents were less hostile to adaptation policies than mitigation policies. Mitigation policies are focused on limiting the severity of climate change, primarily through reducing carbon emissions via governmental policy. Therefore conservatives may either be less concerned about the severity of climate change, believe reducing carbon emissions is a less appropriate course of climate action, or more resistant to governmental policies which rely on changes to the current socio-economic or tax system (e.g., the carbon tax). This may be due to the varying level of socio-economic change mitigation and adaptation strategies require. Mitigation policies typically involve substantial changes, and therefore threats, to the current socio-economic and cultural system as they aim to limit the existing impacts of climate change through governmental intervention, which right-wing adherents are particularly sensitive and reactive to (Clarke et al., 2019;Feygina et al., 2012).  (Chinn & Hart, 2021) or can even lead to some form of reactance, with right-wing adherents viewing climate change policy as less important post-message (Singh & Swanson, 2017). It is also important to note that the source of our message were university scientists, and while generally trusted, these may not be trusted by right-wing adherents when it comes to climate change. While not the same source employed in our study, recent research has demonstrated left-leaning sources result in more positive attitudes towards climate policies for left-wing adherents, yet this does not occur when right-leaning sources present the same information to right-wing adherents (Fielding et al., 2020). Therefore, future research should further investigate whether loss messages that tap into traditional conservative economic (economic growth) or cultural (nationalism) values may work in specific countries, with policy types, and with certain sources (e.g., political in-group members).
An unexpected finding of the current research was that we found a positive, unique effect of participant national identity on support for adaptation policy in our Australian sample (Study 1). While this effect was smaller than political orientation in both studies, those who more strongly identified as Australian were more likely to support adaptation policies, but not mitigation policies. Firstly, this suggests that it is still possible to utilize national identity as a potential driver of adaptation policy support, especially in Australia, which is less politically polarized than the United States (Finkel et al., 2020). This is because while prior research shows that nationalism is positively related to right-wing ideologies in the United States (Roccas et al., 2010), there was only a weak, positive relationship between Australian identity and right-wing orientation in our research. Further still, prior research suggests that Australians do view 'caring for the environment' as an important part of their national identity (Purdie & Wilss, 2007). Secondly, this unexpected finding suggests that adaptation policies, which are focused on adapting to the negative environmental effects of climate change, may have different psychological antecedents to those policies which are focused on limiting the severity of climate change. As noted above, mitigation policies tend to result in system change (Clarke et al., 2019), while adaptation policies typically result in changes within an existing system. Therefore, it is possible that those who strongly identify with their national social group find small changes in a system they already value (i.e., their country) more palatable. Future research should explore the utility of national identity as an antecedent to adaptation policy support and see whether it could be more effectively employed in climate change communication, especially in Australia.

Limitations
The current research provided an experimental test of whether economic loss or national loss messages may contribute to climate change policy support across two different countries -Australia (Study 1) and the United States (Study 2). While political orientation does contribute to climate change outcomes in these countries, it is possible that different effects may arise in other countries where climate change is not politically polarizing (Smith & Mayer, 2019).
Furthermore, the findings across the two studies did differ somewhat, suggesting potential country-level differences in how individuals respond to climate change communications.
Although prior research demonstrated that national identity is related to political conservatism in Anglosphere nations, we did not find this in the Australian context in Study 1 (there was only a weak relationship), and so its ability to accentuate the impact of political conservatism may have been weaker than in the U.S. context. Finally, while there was appreciable variability within conditions, the average endorsement of policies was reasonably high across all groups in both studies. As such, this may limit inference to those strongly opposed to adaptation and mitigation Running head: NATIONAL IDENTITY, POLITICAL ORIENTATION, AND CLIMATE CHANGE COMMUNICATIONS 33 strategies and may further have put a limit on the size of the observable effects due to measurement ceilings.

Conclusion
Across two studies, and in line with prior research, we found that identifying as politically right-leaning (in Australia, Study 1) or conservative (in the United States, Study 2) was negatively related to both mitigation and adaptation policy support, providing further evidence that governmental climate action is politically polarized in these nations. Yet, we also found that Australian (Study 1) identity was uniquely and positively predictive of adaptation policy support, but not mitigation policy support. This suggests that different psychological variables may predict different types of climate policies, and in the case of this research, those who identify as an Australian appear more likely to support policies which simply adhere to the current political environment rather than change the socio-economic and cultural system.
Findings across the two studies also showed that an economic loss and national identity loss message did not result higher mitigation and adaptation policy support when compared to one another or to a control, and this was not conditional on right-wing political attitudes. These findings suggest that if one must discuss climate change, describing the threat to cultural symbols and national environmental icons, or the potential economic losses if we do not act on climate change, may do little to increase climate policy support in right-wing adherents.