Sustainable practices and household resilience: an empirical analysis of behavior change in times of crises

The relevance of sustainable behavior change lies in its ability to deliver sustainability goals in practice. Behaviors typically under investigation include those related to reducing waste, improving energy efficiency, and conserving water and biodiversity. These behavioral changes are an integral part of the transition to a society that values sustainability in all aspects of life (Grin et al., 2010). Understanding the underlying drivers of behavior change is essential for promoting sustainability transitions. The Theory of Planned Behavior (TPB, Ajzen, 1991) provides a well-established framework, suggesting that individual actions are influenced by attitudes, subjective norms, and perceived behavioral control. This theory is central to understanding how attitudes such as environmental awareness drive sustainable lifestyle adaptations in areas such as food consumption, transport, and energy use.

However, research suggests that sustainable behavior is not only a product of individual perceptions and cognitive processes but is also significantly influenced by external socio-economic factors and institutional frameworks (Gifford, 2011). Structural barriers such as the lack of access to sustainable mobility options or economic constraints can significantly impede the adoption of pro-environmental behaviors despite high levels of environmental awareness (Steg & Vlek, 2009). The Comprehensive Action Determination Model (Klöckner & Blöbaum, 2010), that integrates core constructs of the TPB, expands on this by additionally incorporating intentional, normative, and situational influences, showing that external factors such as available infrastructure and social norms play a critical role in shaping environmentally friendly behavior like travel mode choices. Similarly, the Theory of Interpersonal Behavior (Triandis, 1977) emphasizes the importance of facilitating conditions, but also of habitual behaviors, highlighting how past behaviors and external situational influences interact with individual behavior. Moreover, factors such as cultural norms, policy incentives, and technological advances significantly influence the ability of individuals to engage in sustainable practices (Kollmuss & Agyeman, 2002). For example, the adoption of renewable energy is not simply a personal choice, but often depends on policy support and economic incentives.

Recognizing that the path to sustainability transformation is not linear or predictable and has to account for individual-level and external influences, particularly in an era of volatility, it is essential to focus on the development of sustainable behavior in times of crisis.

The role of crises in shaping sustainable behavior has emerged as a critical area of research. Global events, such as the COVID-19 pandemic and the subsequent energy crises following the the Russian invasion of Ukraine, have not only disrupted traditional patterns of consumption, but also provided a unique opportunity to accelerate the transition to more sustainable practices. Such crises can act as catalysts, prompting individuals and households to reassess their consumption habits and adopt more energy-efficient behaviors (Ergen & Suckert, 2021).

Within the context of crises, the concept of resilience is also relevant, due to its transformative capacity in fostering sustainable behavior. In the ecological sciences, resilience has been defined as the ability of a system (an individual, a community, or an ecosystem) to absorb stress, recover from disruptions and maintain its essential functions, structures, and identity (Alexander, 2013; Hosseini et al., 2016). This concept encompasses the ability to adapt to changing conditions and to transform in ways that enhance the system's sustainability, often in response to challenges and disruptions. The concept of resilience has also been used in the field of technical systems in the context of energy and logistics infrastructure. Resilience in this context is not only about the ability of these systems to withstand disruptions, but also their ability to recover quickly and even improve their functionality in the aftermath. It involves proactive planning, robust design, and adaptive management to ensure continuity of service and minimize the impact on society and the economy. Understanding and enhancing the resilience of energy and logistics infrastructure is critical to maintaining the stability and efficiency of our interconnected global systems, especially in the face of increasing uncertainty and complexity (Hamborg et al., 2020).

In the context of household behavior in crisis, this traditional notion of resilience takes on a more differentiated dimension. Households, as microunits of larger social and ecological systems, demonstrate resilience when they successfully navigate and adapt to the adverse conditions created by global crises such as the COVID-19 pandemic or energy shortages. This involves strategically adjusting resource use and actively adopting behaviors that contribute to the sustainability and well-being of the household (Ansah et al., 2019). For example, shifting to more energy-efficient behaviors, reassessing consumption habits, and adopting sustainable lifestyles are examples of household resilience in action. These adaptive and transformative responses at the household level are critical to fostering broader societal resilience and progress towards achieving the Sustainable Development Goals (SDGs) (Assarkhaniki, et al., 2023).

Recent research has begun to empirically investigate household behaviors in the context of crises and resilient behavior practices. Corbos et al. (2023) found a positive impact of knowledge of energy savings on actual energy-saving behaviors during the energy crisis. This relationship was moderated by responsible behavior, suggesting that mere awareness is not sufficient without a sense of responsibility towards sustainable practices. Liobikienė et al. (2023) investigated the factors that influence energy-saving behavior and identified concerns stemming from the war against Ukraine as the primary influencing factor. This suggests that global crises can significantly shape individual sustainability practices. Novianto et al. (2022) studied the impact of lifestyle changes on home energy consumption during recent crises. They found an increase in the use of household energy appliances, particularly among the middle-upper consumer segments. Monterde-i-Bort et al. (2022) investigated mobility patterns and mode choice preferences during the COVID-19 pandemic in 10 (mostly European) countries. During the pandemic, respondents reduced their mobility, including the use of cars, local public transport, and walking. After the lifting of mobility restrictions, however, their habits reverted almost back to the pre-COVID-19 level.

While global crises and the ability of households to adapt play an important role, individual choices and behavior within households also matter. The following section therefore explores the role of individual factors in sustainability behavior.

Individual impact factors play a crucial role in sustainable behavior, acting as both facilitators and barriers to environmental action. Sociodemographic variables such as age, gender, and income significantly influence the likelihood that an individual will engage in sustainable practices. For example, Wiernik et al. (2020) found that older people may be less likely to adopt new environmental behaviors due to deep-rooted habits, while younger people are generally more open to change and innovation.

Societal roles and structural barriers, particularly for vulnerable groups, also have a significant impact on sustainable behavior (Sovacool et al., 2022). The capacity to engage in sustainable behaviour is often dependent on financial resources and access to the required infrastructure (e.g., Axon & Morrissey, 2020). These factors include gender roles, economic disparities, and access to infrastructure, which can privilege certain individuals while disproportionately burdening others. For example, sustainable behavior can often become a responsibility assigned to women in households due to traditional divisions of labor, as seen in the management of energy consumption and waste (Räty & Carlsson-Kanyama, 2010).

Environmental awareness also plays a key role in sustainable behavior. Individuals with greater environmental knowledge are more likely to engage in sustainable behaviors such as waste reduction and energy conservation. This correlation between knowledge and action suggests that informed awareness is crucial in promoting pro-environmental behavior (Lacroix & Gifford, 2022).

In addition, risk perception, particularly in relation to climate change and environmental threats, has been shown to motivate sustainable behavior (Smith & Leiserowitz, 2022). People who perceive higher levels of environmental risk are more likely to change their behavior to reduce those risks (O'Connor et al., 1999). However, the relationship between risk perception and behavior can be complex and is often mediated by personal values and beliefs (Spence et al., 2012).

Although studies on sustainable behavioral change increasingly focus on the micro level of households and individuals and also take into account the increasing volatility caused by times of crisis, there has not yet been a detailed empirical study on the extent and nature of more sustainable behavioral patterns and individual influencing factors in the wake of past crises.

Therefore, this study aims to fill this gap by examining changes in sustainability behavior in the context of recent crises. The following research aims were pursued:

The questionnaire was modular in design and included the following sections:

The items related to attitude factors and risk perception had to be answered on a six-point Likert scale (1 = strongly disagree to 6 = strongly agree). Items that were formulated negatively were recoded before statistical analysis.

The questionnaire was pre-tested for comprehensibility with a sample of n = 3. The final version of the questionnaire was reviewed by the Faculty Ethics Committee and approved as ethically unobjectionable. The data for this study was collected in May 2023, which allowed us to capture behavioral changes occurring in response to the aftermath of both the COVID-19 pandemic and the energy crisis following the war against Ukraine.

A census-representative sample with data of n = 541 respondents (M = 48.79 years, SD = 14.6, 47.5% male, 52.5% female) was analyzed in the study (n = 1007 before data cleaning). Most participants completed secondary education (56.4%) according to the International Standard Classification of Education (ISCED, 2011), followed by those with tertiary education (42.5%). Only a small percentage (1.1%) reported primary education as the highest educational degree. Asked for their area of living, 28.1% of participants lived in the city center, 43.3% in the suburbs, and 28.7% in rural areas. In terms of income distribution, the average reported income was in the category 2000–2999€.

For a clearer presentation of the results, the information on the behavior-change-related items was summarized into three categories: a) no changes, b) reduced behavior referred to as “less climate-friendly behavior” (summarizing the responses “much less frequently, less frequently, and rather less frequently”), and increased behavior referred to as “more climate-friendly behavior” (summarizing the responses “somewhat more frequently, more frequently, much more frequently"). First, descriptive and factor analyses were applied to understand the distribution of behavioral change data and their dimensional structure. Second, cluster analyses were employed to segment respondent groups based on their behavioral patterns. Third, depending on psychometric scale properties, cluster group differences and individual factors were analyzed by MANOVAs and nonparametric tests.

A mixed pattern of behavioral changes compared to pre-crisis times was found (Fig. 1): Regarding public transport, 54.9% indicated no changes, 23.5%, reported reduced usage, and 21.6% reported an increased usage. For the use of cars, most respondents (55.1%) reported no changes, 30.9% reported reduced car usage and 14% reported increased car usage.

Full size image

Sustainable shopping behaviors also increased compared to pre-crisis times, with 48.2% of participants indicating they shopped more sustainably. No changes were reported by 40.5% and 11.3% indicated reduced sustainable shopping behaviors. Regarding energy-saving behaviors, over two-thirds (68.2%) reported increased energy-saving behavior, 27.2% reported no changes and 4.6% reported less energy-saving behavior.

After analyzing the distribution of behavioral changes, the next step focused on the analysis of their underlying dimensional structure. A factor analysis (principal component analysis with varimax rotation) was employed to determine whether the behavioral changes represented distinct domains or if they formed a singular construct. Bartlett’s test of sphericity was significant (χ ₍₆₎ = 240; p < .01), indicating that the data was appropriate for factor analysis. Two components with eigenvalues above Kaiser’s criterion of 1 explained 71.1% of the variance (Table 1). The items “use of car” and “use of public transport” loaded on the first factor, labeled as “sustainable mobility behavior” (SMB). The second factor, marked by loadings from “energy-saving behavior” and “sustainable shopping behavior”, represented “sustainable consumer practices” (SCP).

For the identified factors “sustainable mobility behavior” (SMB) and “sustainable consumer practices” (SCP) cluster analyses were employed based on the respective factor scores. First, a two-step cluster analysis was run to determine the optimal number of clusters, followed by a K-Means clustering algorithm to partition respondents into distinct groups.

Recent crises such as the COVID-19 pandemic and the energy crisis due to the war against Ukraine exerted different effects on the two domains of sustainability behavior. Although significant potential for change toward sustainable mobility practices was anticipated due to current crises (Csutora & Zsóka, 2023), our results indicate that there was rather little change—or high stability—in mobility practices. Over half of the participants reported no change in their mobility practices. We assume that external factors like lockdowns, differing work commitments (home office), public transport availability, or societal norms shaped mobility behavior (Renn et al., 2022). On the other hand, we found a significant shift in sustainable consumer practices: Most participants reported increased energy-saving behaviors and nearly half reported more sustainable shopping practices. This may either be a result of economic factors (increased food and energy prices) or an increased awareness (Liobikienė et al., 2023). Comparing the extent of changes in both sustainability domains, our findings indicate a different elasticity of behaviors. Behaviors such as energy-saving and sustainable shopping were more adaptable to change, primarily because they are less influenced by external factors and more by individual choices. This suggests accordingly that policy interventions should create external conditions to foster more sustainable mobility habits. Further, the comparative analysis of changes within the two sustainability behavior domains suggests high consistency and limited spillover effects (Puntiroli et al., 2022). This means that adopting sustainable practices in one area (e.g., energy-saving) does not necessarily lead to sustainable decisions in another area (e.g., mobility). This differentiation emphasizes the need for targeted interventions across different domains of sustainability. Finally, the shift towards sustainable consumer practices suggests an emerging recognition of the importance of (energy) resilience at the household level and underscores the potential of individual behavior in shaping a more resilient energy and consumption landscape (Bertoldi, 2022). To ensure long-term environmental sustainability, policymakers should prioritize initiatives that support and promote household energy resilience.

The varied patterns of sustainable behavior changes underscore the fact that individuals react differentially to crises. While some took it as an opportunity to adopt more sustainable behaviors (“sustainable transit converters” and “sustainable consumption champions”), others remained locked in old habits (“stable shoppers and moderate energy-savers”) or even changed their behavior to less sustainable ones (“car centric converters”, “reduced sustainable practices”). This suggests that while certain segments of society willingly shift towards more sustainable behaviors—particularly in times of crisis—others encounter substantial barriers and external constraints. These include socio-economic limitations, inadequate infrastructure, or a lack of institutional support (e.g. Sovacool et al., 2024), which may hinder their ability to engage in sustainable practices despite their willingness. Therefore, the existence of clusters that either remained static or transitioned towards less sustainable behaviors poses a critical challenge. Thus, simply raising awareness or experiencing a crisis does not automatically lead to positive behavioral shifts (e.g., Venghaus et al., 2022). Given the rising political polarization and the prevalence of climate skepticism, it is crucial to prioritize these groups displaying less sustainable behaviors to better understand and address the barriers they face, and identify potential incentives for change (e.g., Moberg et al., 2021).

The inclusion of individual factors revealed that the (non-)adoption of sustainable behaviors is influenced by several factors including socio-demographics, knowledge, attitudinal factors, and risk perception, which highlights the importance of multi-dimensional approaches in understanding and promoting sustainable behaviors. Regarding sociodemographics, the lower income of “sustainable transit converters”' indicates economic constraints leading to more sustainable choices. Similarly, higher income and education of the SCP-cluster “sustainability champions” may provide individuals the means to make more sustainable choices. For policymakers, this suggests considering the options of increased subsidies or incentives to improve affordability across all income groups. The area of living was a significant predictor in the SMB domain, underscoring the role of urban infrastructure in sustainable mobility choices. Urban areas often have better public transport infrastructure and more incentives or restrictions regarding car usage, making sustainable commuting more feasible. This highlights the importance of improving public transport and promoting sustainable mobility options also in non-urban areas. However, it is important to consider that people with lower financial capabilities may prioritize cost-efficiency over sustainability (e.g., Liobikienė et al., 2023). Economic factors become even more relevant in times of crisis, which may further exacerbate financial hardships. Regarding attitudinal factors, those with higher climate change awareness were more inclined to engage in sustainable behaviors, both in mobility and consumption. The regression analysis further validated this finding, showing a strong association between climate change awareness and positive changes in both mobility and consumer practices. Another factor, which was significant in predicting sustainable consumer practices was subjective sustainability knowledge. The better-informed individuals were about sustainability topics, the more they were engaged in sustainable shopping and energy-saving behavior. Knowledgeable consumers seem to be better equipped to make informed decisions, suggesting that communication and awareness campaigns can play a pivotal role in promoting sustainable consumer practices. Risk perception was another influential factor, i.e., those who perceived higher risks associated with climate change were more likely to adopt sustainable consumer practices (e.g., Arning et al., 2018). This implies that understanding the consequences of climate change can significantly impact sustainability behavior. Accordingly, the communication of climate change consequences should also be included in educational and communication activities. Trust in institutions was also important, especially in the context of sustainable mobility in predicting sustainability behavior changes. Clusters with higher trust levels were more inclined towards sustainable practices, suggesting that institutional trust can guide individuals towards more sustainable choices (e.g., Rayner, 2010). However, an essential point to consider is the low proportion of explained variance in the regression analyses, especially for mobility behavior. This suggests that the complexity of sustainable (mobility) behavior extends beyond these individual factors. Each cluster represents a distinct response to the crises, influenced by a combination of individual characteristics and, other, unmeasured systemic or structural factors, which underscores the need for a comprehensive research approach.

The results of our study advise targeted policy interventions to effectively promote sustainable behavior. Specifically, policymakers should recognize socio-economic differences and their implications for more diversified and targeted policy-design. Most notably, income level and climate change awareness were found to impact sustainable consumption choices and mobility behavior. Accordingly, policy interventions specifically targeting at affordability across income groups and educational or communication campaigns to increase the level of knowledge related to sustainability challenges should be prioritized. Both of the above have recently been addressed by European Union (EU) and German national policy initiatives. As part of the European Green Deal, the EU pursues significant action towards consumer empowerment with the objective of providing consumers with access to information needed to make informed sustainable consumption choices, while protecting them against corporate green washing activities (European Commission, 2024). On the German national level, following the Russian invasion of Ukraine and the consecutive surge in energy prices, the German government issued a relief package consisting of various measures including fuel tax reductions, heating cost subsidies, but also measures that specifically targeted at students or retirees (German Federal Government, 2022). In January 2024, the revised Building Energy Act (GEG) was enacted, which – for the first time – includes an income-dependent component allowing for higher subsidies for households with lower or medium income levels (BMWSB, 2024). The effectiveness and behavioral impacts of such targeted policy interventions should thus be subject of further scientific monitoring and evaluation in the future. Furthermore, energy saving behavior has been found to be most reactive to change in times of crises. Future research should therefore target the question, whether the heightened willingness to change towards energy saving behavior likewise suggests that policy interventions targeting this topic will be especially effective.

Our findings need to be contextualized within certain methodological limitations.

Our analysis was based on the use of self-reported data which may cause biases, such as the social desirability bias, which may have influenced participants to over-report socially favorable behaviors, such as sustainable actions, while underreporting less desirable behaviours. Furthermore, inaccuracies in recalling past behaviours could have affected participants' ability to accurately remember their consumption and mobility choices over time. Future studies should therefore aim to triangulate self-reported data with objective consumption data (e.g., utility bills, transportation records). Alternatively, longitudinal approaches that track behavior in real time could provide more reliable insights by reducing the reliance on participants’ memory.

In our study we focused on general behaviors such as saving energy, using public transport, and buying sustainably. For a more detailed analysis of cluster profiles, we suggest that future studies incorporate a larger and more detailed set of variables, particularly in sustainable consumption behaviors. Future research should also assess the motives or forces behind sustainable behavior changes to capture the interplay between financial constraints and sustainability-driven behaviors to better contextualize these changes. Further, future studies should include additional individual, systemic and structural factors such as social norms, flexibility to work from home, policy, and regulation to gain a deeper understanding of behavioral changes in times of crises. A further area in which more detailed data is required is in the assessment of whether expressed preferences align with actual behaviours concerning sustainable mobility and consumption. Another aspect is the timing of the data collection, which took place in May 2023. This timing allowed us to capture behavioral responses influenced by both the COVID-19 pandemic and the energy crisis following the war against Ukraine. However, focusing on a single time point may overlook ongoing fluctuations in behavior and market dynamics. Further, the study provides insights into self-reports on behaviors over three years, but the temporal stability of these changes is still unclear. It remains open whether these shifts represent short-term responses to specific crises or indicate longer-term behavioral adaptations. A longitudinal study would offer a more comprehensive understanding of how these behaviors evolve over time in response to external crises. In addition, the generalizability of our findings remains to be determined: are these patterns of behavioral change globally consistent or specific to geographical or cultural contexts? Future research might benefit from integrating mixed methods (e.g., Csutora et al., 2021) to ensure a comprehensive understanding of sustainable behavior. Finally, the study's focus on Germany limits the applicability of its findings to other regions. Variations in infrastructure, cultural norms, and policy frameworks across countries may result in different behavioral outcomes. Therefore, cross-national comparative studies should be expanded to include diverse national contexts, ensuring a more comprehensive understanding of behavioral shifts during crises.