Project: FREE policy brief
From Integration to Reconstruction: Standing with Ukraine by Supporting Ukrainians in Sweden
Sweden has strongly supported Ukraine through both public opinion and government actions, yet there has been little discussion about the needs of Ukrainian displaced people in Sweden. The ongoing war and the rapidly shifting geopolitical landscape have created uncertainty – geopolitical, institutional, and individual. Ukrainian displaced people in Sweden face an unclear future regarding their rights, long-term status, and opportunities, making future planning or investing in relevant skills difficult. This uncertainty also weakens the effectiveness of integration policies and limits the range of policy tools that can be deployed, which hinders participation in the labor market, affecting both displaced and employers. Addressing these challenges is essential, not only for the well-being of Ukrainians in Sweden, but also for Sweden’s broader role in supporting Ukraine. Helping displaced Ukrainians rebuild their lives also strengthens their ability to contribute both to Swedish society and to Ukraine’s future reconstruction and integration into Europe.
The Swedish Approach to Displaced Ukrainians
In response to the Russian full-scale invasion of Ukraine, the Temporary Protection Directive (2001/55/EC) (commonly referred to as collective temporary protection) was activated in March 2022, granting Ukrainians seeking refuge temporary protection in EU countries, including Sweden. This directive provides residence permits, access to work, education, and limited social benefits without requiring individuals to go through the standard asylum process.
However, the practicalities of the Directive’s use differed significantly between countries. Sweden, despite its, until recent, reputation of being relatively liberal in its migration policies, has at times, lagged behind its Scandinavian neighbors in supporting Ukrainian displaced people. To illustrate this, it is useful to compare the Swedish approach to that of other Nordic states, as well as Poland.
Comparison to Other Nordic States
The Nordic countries have implemented the directive in different ways, adopting varying policies toward Ukrainians demonstrating different degrees of flexibility and support. Despite its generally restrictive immigration policy, Denmark introduced some housing and self-settlement policies for Ukrainians that were more liberal than its usual approach. Norway also initially introduced liberal measures but later tightened regulations, banning temporary visits to Ukraine and reducing financial benefits. Finland, meanwhile, has taken a relatively proactive stance, granting temporary protection to over 64,000 Ukrainians – one of the highest per capita rates in the region. Its strong intake reflects a more flexible and effective implementation of the directive, particularly from late 2022, when it surpassed Sweden and Denmark in number of arrivals.
In Sweden the so-called “massflyktsdirektivet“ grants Ukrainians temporary protection until at least March 2025. Its future beyond that, however, remains uncertain, adding to the challenges faced by refugees and policymakers alike. Sweden – considered liberal in migration policies (at least, up until 2016) – has been criticized for offering limited rights and financial support to displaced Ukrainians, making it one of the least attractive destinations among the Nordic countries (Hernes & Danielsen, 2024). Under “massflyktsdirektivet”, displaced Ukrainians were entitled to lower financial benefits and limited access to healthcare compared to refugees or residents with temporary permits. It was only in July 2023 that they became eligible for Swedish language training, and only in November 2024 could they apply for residence permits under Sweden’s regular migration laws – a pathway that can eventually lead to permanent residence.
Figure 1 illustrates significant fluctuations in the number of individuals granted collective temporary protection in the Nordic countries over the first two years following Russia’s full-scale invasion. As Hernes and Danielsen (2024) show in a recent report, all Nordic countries experienced a peak in arrivals in March-April 2022, followed by a decline in May-June. Sweden initially received the most, but aside from this early peak, inflows have remained relatively low despite its larger population (Table 1). Since August 2022, Finland and Norway have generally recorded higher arrivals than Denmark and Sweden. By August 2023, Norway’s share increased significantly, accounting for over 60 percent of total Nordic arrivals between September and November 2023.
Figure 1. Total number of individuals granted collective temporary protection in the Nordic countries
Source: Hernes & Danielsen, 2024, data from Eurostat.
Table 1. Total number of registered temporary protection permits and percent of population as of December 2023
Source: Hernes & Danielsen, 2024, data from Eurostat.
Comparison to Poland
Sweden’s policies and their outcomes compare rather poorly to those of Poland, one of the European countries that received the largest influx of Ukrainian migrants due to its geographic and cultural proximity. A key factor behind Poland’s relatively better performance is that pre-existing Ukrainian communities and linguistic similarities have facilitated a smoother integration. Ukrainians themselves played a crucial role in this regard, with many volunteering in Polish schools to support Ukrainian children. Sweden also had a community of Ukrainians who arrived to the country over time, partly fleeing the 2014 annexation of Donetsk and Crimea. Since these individuals were never eligible for refugee status or integration support, they had to rely on their own efforts to settle. In doing so, they built informal networks and accumulated valuable local knowledge. Nevertheless, after the full-scale invasion in 2022, they were not recognized as a resource for integrating newly arrived Ukrainian refugees – unlike in Poland.
However, Poland’s approach was shaped not only by these favorable preconditions but also by deliberate policy choices. As described in a recent brief (Myck, Król, & Oczkowska, 2025), a key factor was the immediate legal integration of displaced Ukrainians, granting them extensive residency rights and access to social services, along with a clearer pathway to permanent residence and eventual naturalization.
Barriers to Labor Market Integration
Despite a strong unanimous support for Ukraine across the political spectrum, there is less public debate and fewer policy processes in Sweden regarding displaced Ukrainians, most likely attributable to the general shift towards more restrictive immigration policies. The immigration policy debate in Sweden has increasingly emphasized a more “selective” migration, i.e. attracting migrants based on specific criteria, such as employability, skills, or economic self-sufficiency. This makes it puzzling that displaced Ukrainians, who largely meet these standards, have not been better accommodated. Before the full-scale invasion, Sweden was a particularly attractive destination among those who wanted to migrate permanently, especially for highly educated individuals and families (Elinder et al., 2023), indicating a positive self-selection process.
When large numbers of displaced Ukrainians arrived after the full-scale invasion, many had higher education and recent work experience, which distinguished them from previous refugee waves that Sweden had received from other countries. Despite a strong labor market in 2022, their integration was hindered by restrictions imposed under the Temporary Protection Directive, which limited access to social benefits and housing. At the same time, Sweden explicitly sought to reduce its attractiveness as a destination for migrants in general, contributing to a sharp decline in its popularity among Ukrainians after the war escalated.
In addition to the restrictiveness and numerous policy shifts over time, the temporary nature of the directive governing displaced Ukrainians – rather than the standard asylum process – creates significant policy uncertainty. This uncertainty makes it difficult for Ukrainians to decide whether to invest in Sweden-specific skills or prepare for a potential return to Ukraine, whether voluntary or forced, complicating their long-term planning. It also hinders labor market integration, increasing the risk of exploitation in the informal economy. Another key challenge is the unequal distribution of rights, as entitlements vary depending on registration timelines, further exacerbating the precarious situation many displaced Ukrainians face in Sweden.
A survey of 2,800 displaced Ukrainians conducted by the Ukrainian NGO in Sweden “Hej Ukraine!” in February 2025 provides key insights into their labor market integration (Hej Ukraine!, 2025). Survey results show that, currently, 40 percent of respondents are employed, with 42 percent of them holding permanent contracts while the rest work in temporary positions and 6 percent being engaged in formal studies. Employment is concentrated in low-skilled sectors, with 26 percent working in cleaning services, 14 percent in construction, and 12 percent in hospitality and restaurants. Other notable sectors include IT (11 percent), education (8 percent), warehousing (7 percent), elderly care (5 percent), forestry (3 percent), and healthcare (3 percent). The lack of stable permits, access to language courses (until September 2024), and financial incentives for hiring displaced persons have complicated their integration.
As mentioned above, the Swedish government has over time introduced several initiatives to facilitate the integration of displaced Ukrainians. However, assessing their effectiveness is crucial to identify persistent challenges and to formulate targeted policy solutions.
The Role of the Private Sector and Civil Society
The business sector, civil society and NGOs have also played a role in supporting displaced Ukrainians, filling gaps left by the public sector. This includes initiatives aimed at creating job opportunities that encourage voluntary return. However, broader systemic support, including simplified diploma recognition and targeted re-skilling programs, is needed to enhance labor market participation.
Moreover, there is a lack of information among displaced, potential employers and public institutions (municipality level) about the tools and programs available. For example, a community sponsorship program funded by UNHCR, which demonstrated positive effects on integration by offering mentorship and support networks, was only applied by five municipalities (UNHCR, 2025). Similar programs could be expanded to address structural barriers, particularly in the labor market. Another example is the Ukrainian Professional Support Center established to help displaced Ukrainians find jobs through building networks and matching job seekers with employers (UPSC, 2024). The center was funded by the European Social Fund, and staffed to 50 percent by Ukrainian nationals, either newcomers or previously established in Sweden, to facilitate communication. Experiences from this initiative, shared during a recent roundtable discussion – Integration and Inclusion of Ukrainian Displaced People in Sweden, highlighted that between 2022 and 2024, about 1,400 Ukrainians participated in the project, but only one-third of participants found jobs, mostly in entry-level positions in care, hospitality, and construction. Restrictions under the temporary protection directive, along with the absence of clear mechanisms for further integration, posed significant challenges; the lack of a personal ID, bank account, and access to housing were considered major obstacles. The uncertainty of their future in Sweden was also reported as a significant source of stress for participants.
Implications and Policy Recommendations
The lack of clarity surrounding the future of the EU Temporary Protection Directive, as well as its specific implementation in Sweden, leaves displaced Ukrainians in a precarious situation. Many do not know whether they will be allowed to stay or if they should prepare for a forced return. This uncertainty discourages long-term investment in skills, housing, and integration efforts.
Uncertainty also affects Swedish institutions, making it difficult to implement long-term policies that effectively integrate Ukrainians into society. To address these issues, the following policy recommendations are proposed.
- Extend Temporary Protection Status Beyond 2025: Clear guidelines on the duration of protection are necessary to provide stability for displaced Ukrainians
- Improve Labor Market Access: Introduce targeted programs for skill recognition, language training, and financial incentives for businesses hiring displaced Ukrainians
- Enhance Civil Society and Private Sector Collaboration: Support mentorship and community sponsorship programs that facilitate integration
- Acknowledge and Utilize displaced Ukrainians as a Resource: Recognizing displaced Ukrainians as potential assets in rebuilding Ukraine and strengthening European ties should be a priority.
- Increase Public and Policy Debate: There is a need for greater discussion on how to integrate Ukrainians in Sweden, as an important complement to the policy priority of providing aid to Ukraine.
By implementing these measures, Sweden can provide displaced Ukrainians with greater stability, enabling them to engage in the formal labour market rather than being pushed into informal or precarious employment. This not only benefits Ukrainians by ensuring fair wages and legal protection, but also strengthens Sweden’s economy through increased tax revenues and a more sustainable labour force.
As Sweden continues to support Ukraine in its fight for sovereignty, it should also recognize the value of displaced Ukrainians within its borders, fostering their contribution to both Swedish society and Ukraine’s eventual reconstruction.
References
- Hernes, V., & Danielsen, Å. Ø. (2024). Reception and integration policies for displaced persons from Ukraine in the Nordic countries – A comparative analysis. NIBR Policy Brief 2024:01. https://oda.oslom et.no/oda-xmlui/handle/11250/3125012
- Hej Ukraine! (2025). Telegram channel. https://t.me/hejukrainechat
- Elinder, M., Erixson, O., & Hammar, O. (2023). Where Would Ukrainian Refugees Go if They Could Go Anywhere? International Migration Review, 57(2), 587-602. https://doi.org/10.1177/01979183221131559
- EUROSTAT. Decisions granting temporary protection by citizenship, age and sex – monthly data. Dataset. https://ec.europa.eu/eurostat/databrowser/view/migr_asytpfm__custom_15634298/default/map?lang=en
- Myck, M., Król, A., & Oczkowska, M. (2025, February 21). Three years on – Ukrainians in Poland after Russia’s 2022 invasion. FREE Policy Brief. Centre for Economic Analysis (CenEA). https://freepolicybriefs.org/2025/02/21/ukrainians-in-poland/
- Ukrainian Professional Support Center (UPSC). (2024). https://professionalcenter.se/omoss/
- United Nations High Commissioner for Refugees (UNHCR). (2025). Community sponsorship. UNHCR Northern Europe. Retrieved [March 6, 2025] from https://www.unhcr.org/neu/list/our-work/community-sponsorship
Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.
Human Capital Loss Among Belarusian and Ukrainian Migrants to the EU
This policy brief examines the underutilization of human capital among involuntary migrants from Ukraine and Belarus in Poland and Lithuania. Focusing on those who migrated after 2020 (Belarus) and 2022 (Ukraine), the brief investigates the factors influencing the conversion of their pre-migration skills into gainful employment in their host countries. Our findings show that despite many migrants possessing high levels of education and professional qualifications, structural barriers and low convertibility of their skills, hinder their full labor market integration. This skill underutilization not only limits migrants’ professional growth and earning potential but also deprives the host countries of valuable skills and potential economic gains.
Effective labor market integration substantially benefits both host and sending countries and migrants themselves. For host nations, successful integration can alleviate critical skill shortages, boost productivity, and drive economic growth (Boubtane, Dumont, & Rault, 2016; Boubtane, 2019; Engler, Giesing, & Kraehnert, 2023; Bernstein et al., 2022). Conversely, inadequate integration leads to underemployment, diminished potential, and economic inefficiency. Countries of origin can benefit from remittances, the return of migrants with enhanced skills, and strengthened international economic ties. However, poor integration risks an uncompensated “brain drain” (Reinhold & Thom, 2009; Barrett & O’Connell, 2001; Iara, 2006; Barrett & Goggin, 2010; Co, Gang, & Yun, 2000). For migrants, the ability to continue their careers means higher earnings and less stress from the acquisition of a new profession, while the non-utilization of existing skills results in their depreciation, potentially causing permanent wage reductions even upon return to the home country (Bowman & Myers, 1967).
Migrants can be broadly categorized into voluntary migrants or forced migrants. Voluntary migrants assess labor market prospects beforehand and often possess convertible human capital – one that can be used in a new labor market. This group often includes professionals like IT specialists and scientists and those in low-skilled but highly transferable professions. Forced migrants, on the contrary, may be utterly unprepared for changes in jurisdiction and possess skills of limited transferability. For example, even highly specialized professions requiring extensive training and substantial human capital, such as lawyers, officials, and teachers, often prove “non-convertible“ (Duleep & Regets, 1999). These individuals’ skills are frequently country specific.
Low convertibility of skills generates significant negative consequences. Highly educated professionals, for instance, may find themselves relegated to low-paying, unskilled jobs, unable to leverage their expertise. This hinders their professional development and deprives host countries of valuable skills and potential contributions to economic growth. Addressing these mismatches is crucial for maximizing the benefits of migration for stakeholders in both home and host countries.
Forced Migration from Belarus and Ukraine
The political crisis in Belarus, starting with the contested 2020 presidential elections, led to widespread repression and significant forced migration. Belarus’s role in supporting Russia’s 2022 invasion of Ukraine exacerbated this situation, resulting in approximately 300,000 Belarusians seeking refuge in the European Union (Eurostat). This number accounts for a substantial proportion of the country’s 9 million population and its approximately 5 million-strong labor force (Belstat).
Russia’s full-scale invasion of Ukraine triggered the most significant wave of migration in Ukrainian history, with over 6 million of the pre-war 44 million population fleeing to the EU (UNHCR). About 90 percent of the initial refugees were women and children due to a mobilization law preventing most men aged 18 to 60 from leaving (UNHCR).
Online Survey and Migrant Differences
To better understand the situation of migrants, their integration into the EU labor market, and to develop data-driven recommendations for improving their conditions, the CIVITTA agency, in partnership with BEROC, conducted an online survey in the summer of 2024. This brief is based on the survey results. The survey includes responses from 616 Ukrainian nationals who migrated to Poland or Lithuania after Russia’s full-scale invasion of Ukraine in 2022, as well as 173 Belarusian migrants who left their home country after 2020. The research focuses on individuals aged 28 to 42, providing insights into their experiences and challenges in the labor market in their host countries. While we acknowledge the sample’s limitations in terms of representativeness, we believe the findings provide valuable insights into the specific challenges faced by involuntary migrants and their adaptation strategies in the new labor market.
Key differences characterize these migration waves. Ukrainian migration comprises of more women, while Belarusian migrants show a more balanced gender distribution, with 47 percent women in our sample versus 62 percent for Ukrainians. Family separation is also notable, as 91 percent of married Belarusians live with their spouses, compared to only 75 percent of Ukrainians (due to the mobilization law).
Survey respondents from both groups possess high levels of human capital with 60 percent of Ukrainians and 90 percent of Belarusians holding higher education degrees. Among Belarusians, 94 percent had over five years of work experience before migration, with and 79 percent of Ukrainians stating the same.
Ukrainian return intentions are split: 38 percent plan to return, 19 percent will not, and the rest are undecided. An end to the war and changes in Russian foreign policy would increase return rates to 70 percent. For Belarusians, 35 percent plan to return, 38 percent will not, and the rest are undecided. Education level is key, as less-educated Belarusians are more likely to stay abroad. An end to repression would increase the share of those Belarusians who want to return to 70 percent, and a regime change would increase this percentage to 82 percent.
Factors Conditioning Human Capital Loss
As expected, due to the involuntary nature of migration of the two groups in focus, a large fraction of survey participants reported losing their profession after migration. As Figure one shows, 48 percent of Belarusians and 63 percent of Ukrainians in our sample reported full loss of their prior careers. The lower percentage of Ukrainians fully retaining their careers (23 percent) compared to Belarusians (44 percent) could be attributed to several factors, including the more recent and disruptive nature of the Russo-Ukrainian war leading to more significant displacement and challenges in finding comparable work. The higher percentage of Ukrainians starting their careers from scratch (49 percent compared to 29 percent among Belarusians) also supports this idea.
Figure 1. Preservation of careers in the EU
Source: Authors’ computations based on survey data.
To foster an evidence-based discussions on the smooth integration of migrants into the EU labor market and the prevention of human capital loss, it is crucial to examine the individual factors that influence career continuity for Belarusian and Ukrainian migrants. We therefore utilize a logistic regression model to identify key predictors that increase the likelihood of migrants remaining in their profession after relocating to Poland and Lithuania.
In our quantitative analysis, an outcome binary variable for staying in the profession is equal to 1 if an individual either “continued career started in a home country (in the same position)” or “remained in the same profession but started working in a position lower than the one held before emigration.” As predictors, we consider a set of sociodemographic variables reasonably related to the probability of staying in the profession and dummy variables for the most common spheres of employment (see Table 1).
Table 1. Overview of model variables
Who Maintains Their Career After Emigration?
Based on the regression coefficients in Table 2, we can identify characteristics related to losing career-specific human capital. In our regression, we control for both home and host country factors. One noteworthy finding is that, while Ukrainian migrants in our sample report significantly higher rates of career loss than Belarusian migrants, nationality itself does not emerge as a significant predictor of career loss once other characteristics are accounted for.
Our results also show that the probability of staying in a profession is higher among men, those with more extended work experience and higher income before emigration, and those who were invited to a host country by an employer. The same holds for entrepreneurs, those who do not plan to return, and those employed in the fields of Architecture & Engineering and Information and Communication Technologies.
Table 2. Results of regression analysis
Note: *** Significant at the .001 level. ** Significant at the .01 level. * Significant at the .05 level.
Conclusion
Several conclusions and policy advice can be derived from the survey results.
The higher likelihood of entrepreneurs staying in their profession suggests that supporting migrant entrepreneurship can be a valuable strategy to retain human capital. This can be done, for example, by:
- Providing access to resources, mentorship, and funding for migrant entrepreneurs.
- Streamlining the procedures for migrants to start and operate businesses.
- Facilitating access to capital for migrant-owned businesses.
The research highlights the disproportionate impact of human capital loss on women. Therefore, policies should include gender-specific programs that address women’s unique challenges in integrating into new labor markets. This could include:
- Skills retraining and certification programs: Designed to align women’s existing skills with the demands of the host country’s labor market, with consideration for childcare needs and other barriers women may face.
- Connecting women migrants with established professionals in their fields to facilitate knowledge transfer and career guidance.
- Language training programs: Tailored to the specific needs of women, potentially incorporating childcare support to enable participation.
The study highlights the positive role of international companies in supporting employee relocation. Respondents who were invited by an employer demonstrated the most successful integration into the new labor market. To enhance and strengthen these networks, policies may focus on:
- Encouraging corporations to hire and train migrant workers, potentially through tax breaks or other incentives. This could include partnerships with migrant-serving organizations to connect companies with qualified candidates.
- Developing digital platforms that connect migrants with diaspora networks, potential employers, and relevant resources.
In addition, policies should address the non-recognition of foreign qualifications, simplifying and expediting the procedures for recognizing foreign degrees and professional certifications. Initiatives to create targeted training programs could complement such policies and allow migrants to quickly acquire any missing skills or certifications required by the host country’s professional bodies. These policy measures would enhance the utilization of migrants’ human capital, benefiting both migrants and host countries while also supporting sending countries. This could be achieved by fostering a successful diaspora or facilitating productive reintegration in the case of return migration.
References
- Barrett, A., & Goggin, J. (2010). Returning to the question of a wage premium for returning migrants. National Institute Economic Review, 213, R43–R51. https://doi.org/10.1177/0027950110389752
- Barrett, A., & O’Connell, P. J. (2001). Does training generally work? The returns to in-company training. ILR Review, 54(3), 647–662. https://doi.org/10.1177/001979390105400403
- Bernstein, S., Diamond, R., McQuade, T. J., & Pousada, B. (2022). The contribution of high-skilled immigrants to innovation in the United States (No. w30797). National Bureau of Economic Research. https://doi.org/10.3386/w30797
- Boubtane, E. (2019). The economic effects of immigration for host countries. L’Economie politique, 84(4), 72–83. https://doi.org/10.3917/leco.084.0072
- Boubtane, E., Dumont, J.-C., & Rault, C. (2016). Immigration and economic growth in the OECD countries 1986–2006. Oxford Economic Papers, 68(2), 340–360. https://doi.org/10.1093/oep/gpv024
- Bowman, M. J., & Myers, R. G. (1967). Schooling, experience, and gains and losses in human capital through migration. Journal of the American Statistical Association, 62(319), 875–898. https://doi.org/10.2307/2283723
- Co, C. Y., Gang, I. N., & Yun, M.-S. (2000). Returns to returning. Journal of Population Economics, 13, 57–79. https://doi.org/10.1007/s001480050121
- Duleep, H. O., & Regets, M. C. (1999). Immigrants and human-capital investment. American Economic Review, 89(2), 186–191. https://doi.org/10.1257/aer.89.2.186
- Engler, P., Giesing, Y., & Kraehnert, K. (2023). The macroeconomic effects of large immigration waves. IAB-Discussion Paper. https://doi.org/10.5167/uzh-239271
- Iara, A. (2006). Skill diffusion in temporary migration? Returns to Western European working experience in the EU accession countries (Development Studies Working Paper No. 210). Centro Studi Luca d’Agliano. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=921492
- Reinhold, S., & Thom, K. (2009). Temporary migration and skill upgrading: Evidence from Mexican migrants. University of Mannheim, unpublished manuscript.
- UNHCR. (n.d.). Operational Data Portal. https://data.unhcr.org/
Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.
Should the $60 Price Cap on Russian Oil Exports be Lowered?
Western governments have imposed a $60 price cap on Russian seaborne oil exports using Western services. To evade the policy, Russia has developed a “shadow fleet” which uses no such services. In this policy brief, we claim that the resulting segmentation of Russian oil exports dramatically modifies the conventional analysis of a price cap. Our research shows that lowering the cap would not hurt Russia as intended unless a robust expansion in non-Russian oil supply was to limit the induced increase in the world oil price. If this price increase is not limited, lowering the cap could even moderately increase Russian profits because shadow fleet sales would be more profitable. By contrast, policies that reduce some shadow fleet capacity would reduce Russian profits if undertaken while Russia still relies on some Western services.
In response to Russia’s invasion of Ukraine in February 2022, the EU, the U.S., and other G7 countries (hereafter the West) ceased their imports of Russian oil, leading Russia to export more to India, Turkey, and China instead. In addition, the West imposed sanctions on oil exports from Russia, whose profits are instrumental in supporting its war.
Since more than 80 percent of Russia’s seaborne oil exports relied on the provision of Western services (CREA, 2023) (financial, operational, and commercial) the EU suggested banning the use of these Western services for all Russian seaborne exports. However, governments feared that this would cause a spike in the world oil price. As an alternative, the U.S. suggested a price cap, which the West ultimately imposed in December 2022, limiting Russian revenues from oil shipped using Western services to $60 per barrel.
Oil transported without Western services is exempt from the cap. Therefore, Russia has gradually assembled a “shadow fleet” that uses non-Western services in order to sell oil at prices above the cap.
The price cap on Russian oil is a new, insofar untested economic sanction, currently a subject of active public discussion, with experts recommending potential adjustments and application to more countries, and policymakers currently considering to tighten the price cap – see for example the January 2025 call by Sweden, Denmark, Finland, Latvia, Lithuania and Estonia to lower the price cap below $60. The policy quickly piqued the interest of economists – see for example Spiro, Wachtmeister, and Gars’ (2024) comprehensive review of policy options to limit Russia’s ability to finance the war.
In their pioneering contribution to the literature, Johnson, Rachel, and Wolfram (2025) provide a rich analysis of the effects of the price cap, albeit under the assumption that the shadow fleet has a fixed capacity. In a recent working paper (Cardoso, Salant, and Daubanes, 2025), we present a new dynamic economic model that accounts for the expansion of the Russian shadow fleet. The model is calibrated to reproduce observed facts and used to simulate the effects of (1) various levels of the price cap, including the extreme case of a complete ban, (2) enforcement stringency, and (3) policies targeting the shadow fleet.
Perhaps surprisingly, our analysis shows that, in the absence of any increase in non-Russian oil supply, lowering the level of the price cap below $60 would benefit Russia. This includes lowering the cap to levels so low (below $34) that the policy amounts to a ban as Russia would prefer not to use Western services at all at these cap levels. More generally, the model reveals that a lower cap would have two opposite effects on Russia: On the one hand, it would reduce Russia’s profit (i.e., revenues net of production costs) from sales at the cap. On the other hand, since a lower cap would reduce Russia’s oil exports, it would increase the oil price and, therefore, Russia’s profit from sales through its shadow fleet. Our analysis yields a testable and intuitive condition under which the latter effect dominates the former, making a lower cap counterproductive. This condition depends on the shadow fleet capacity relative to Russian sales at the ceiling price.
Application of this condition shows that when sanctions were imposed, Russia’s shadow fleet capacity was already sufficiently high for Russia to benefit from a reduction in the price ceiling. Russia would even have benefited from a reduction in the cap if the West had prevented any expansion in Russia’s shadow fleet beyond its initial level. With no such limitation, Russia would continue to expand its fleet size regardless of the size of the cap reduction. This leads us to conclude that Russia would also benefit if an unanticipated reduction in the cap (or a complete ban) occurred subsequently.
It should be noted that in the absence of a non-Russian supply response, caps at different levels quantitatively impact Russian total profits in a similar way. For example, the $60 cap reduces Russian profits by about 25 percent compared to a scenario without sanctions, and a complete ban would have impacted Russia only slightly less.
The following figure shows a comparison of prices, shadow fleet capacity, and profits under a price cap sanction (solid lines), a service ban (dotted lines), and the absence of sanctions (grey dashed lines). The simulations assume no supply response from non-Russian producers (none occurred when the cap was first implemented). A lower cap cuts Russian exports and raises the global oil price, increasing Russian profits from its fleet sales. A non-Russian supply response would dampen this oil price spike and would, therefore, diminish the resulting revenue increase from Russian fleet sales.
Figure 1. Outcomes under different sanction scenarios
Source: Authors’ calculations.
Russia sometimes uses Western services to ship oil at a price above the cap, taking the risk that its shipments get sanctioned. Increasing the probability that cheating is punished lowers the price Russia expects to receive, with consequences identical to a reduction in the cap level.
By contrast, policies that reduce some capacity of the shadow fleet (“sidelining” some of its tankers) may harm Russia, even though they prompt Russia to rebuild its fleet rapidly. This happens, for example, if sidelining part of the fleet occurs while oil is also being sold at the ceiling, so that ceiling sales replace the lost fleet sales and there is no increase in the world oil price.
Conclusion
To conclude, we consider a variety of oil-market sanctions that have been have imposed on Russia to reduce the total export profits it uses to finance the war in Ukraine. As seen, tightening these sanctions is more effective if the induced increase in the world price can be significantly mitigated (if not entirely eliminated); otherwise, increased revenues from shadow fleet sales will weaken or undermine the intended effect of the tighter sanctions.
In one case we considered, no supplementary intervention is required for the sanction to be effective. Reducing Russia’s shadow fleet capacity when Russia is still selling at the ceiling price will induce an equal and offsetting increase in Russian sales at the ceiling, resulting in no increase in the world price.
However, other sanctions – lowering the ceiling, increasing its enforcement, or even reducing the shadow fleet capacity after Russian sales at the ceiling have ceased – will induce an increase in the world price sufficient to undermine the sanctions’ intended effect unless accompanied by a simultaneous expansion of non-Russian supply (presumably from the U.S. or OPEC) to dampen the increase in the world price. Supplemented in this way, the potency of each of these sanctions would be restored.
Overall, our results call attention to the need for complementary energy policies that would facilitate the response of non-Russian oil production to higher global prices.
References
- Cardoso, D. S., S. W. Salant, and J. Daubanes. (2025). The Dynamics of Evasion: The Price Cap on Russian Oil Exports and the Amassing of the Shadow Fleet. MIT CEEPR Working Paper 2025-05.
- Centre for Research on Energy and Clean Air. (2023). December 2023 Monthly Analysis on Russian Fossil Fuel Exports and Sanctions.
- Johnson, S., L. Rachel, and C. Wolfram. (2025). A Theory of Price Caps on Non-Renewable Resources. NBER Working Paper No. 31347.
- Spiro, D., H. Wachtmeister, and J. Gars. (2024). Assessing the Impact of Oil Sanctions on Russia. SSRN Working Paper.
Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.
Agricultural Subsidies: The Case of Georgia
This brief explores the role of government subsidy programs in Georgia’s agricultural sector, with a focus on grapes, apples, and hazelnuts. These subsidies play a significant role in providing social assistance to the sector and in supporting farmers; however, their long-term impact on industry growth remains a subject of discussion. Key challenges include ensuring product quality, enhancing productivity, and expanding market opportunities, particularly regarding export market concentration and infrastructure constraints.
Introduction
Governments have historically intervened in agricultural markets under the pretext of promoting food security. At first, interventions aimed to provide affordable food for rapidly growing urban populations, afterwards more emphasis was put on enhancing agricultural productivity. Nowadays, agriculture remains a priority for policymakers due to its role in promoting inclusive growth and reducing poverty. Additionally, renewed concerns about food security have further driven these policy efforts (Gautam, 2015).
One of the key instruments of these interventions are subsidies in different forms – such as various input subsidies, price supports, and trade interventions. While their use has been widespread, the economic effectiveness of subsidies continues to be heavily debated. Economic theory suggests that subsidies are useful in resolving market failures; however, even in this case, the actual effect of subsidies is highly dependent on the specific implementation. Further, in many other cases, subsidies have led to distortions and have been detrimental to countries’ own economic interests (Gautam, 2015).
Another important concern arises from the political economy of subsidies use. Widening rural-urban income disparities create political pressure to implement measures that support the livelihoods of the large agricultural population. Subsidies, due to their visibility, are a convenient instrument to increase political support from this population group. Further, subsidies offer immediate or near-immediate gains to recipients, whereas public capital investments take longer to deliver results, therefore subsidies are often used as a political instrument. Since political decision-making is typically driven by short-term considerations, often aligned with electoral cycles, long-term investments do not always align with political incentives (Gautam, 2015).
|
Box 1. Subsidies Subsidies are financial assistance provided by governments to support or promote specific sectors, industries, or activities within the economy. They can take various forms, including direct cash payments, tax relief, low-interest loans, and in-kind support, such as the provision of goods and services at below-market prices. Subsidies play a significant role as a tool in government expenditure policy. They influence resource allocation decisions, income distribution, and expenditure efficiency (Schwartz & Clements, 1999). |
In the case of Georgia, subsidies are the main instrument for support to the agricultural sector, with direct subsidies accounting on average 45 percent of total government expenditure in the sector (2014-2024). The government provides subsidies for most of the country’s main crops, including wheat, grapes, hazelnut, tangerines and apples.
Given the scope of this policy brief, only subsidies for major perennial crops – grapes, hazelnuts and apples – are discussed. This as as the wheat sector involves additional food security considerations and due to lack of data for tangerines. Among perennial crops grapes have the highest share of total production (46 percent, including both white and red grapes), followed by tangerines at 14 percent, apples at 10 percent, and hazelnuts at 8 percent (2023, Geostat).
This policy brief firstly explains the Georgian context in more detail, followed by sub-sections discussing each major perennial crop sector, ending with conclusions and policy recommendations.
The Georgian Context
Agriculture plays a crucial role in Georgia’s economy. As of 2024, 39 percent of the population resides in rural areas (Geostat, 2024), where agriculture serves as the primary source of income. The sector employs the largest share of the country’s workforce—17 percent (Geostat, 2023)—yet it contributes to only 7 percent of Georgia’s GDP (Geostat, 2023). At the same time, the disparity in income, and other major socio-economic indicators between the rural (agricultural) and urban population is large. For example, in Tbilisi, the capital of Georgia, the average monthly nominal earnings are 78 percent higher than the average for the rest of Georgia. Additionally, Tbilisi accounts for 70 percent of the total value added generated in the country (Geostat, 2023).
In recent year, the country has undertaken significant efforts to modernize and improve the agricultural sector, yet significant challenges remain. Georgian agriculture is largely characterized by small, fragmented family farms focused on subsistence farming with restricted market access. They are highly vulnerable to weather conditions, yet there is little awareness of or adoption of insurance and risk mitigation measures (State Audit Office of Georgia, 2023). Traditional farming methods remain dominant, with limited use of modern technology. Additionally, most farmers operate on a small scale and lack cooperation and coordination, further hindering efficiency and competitiveness. As a result, they often struggle with low productivity and have difficulty producing high-quality products in stable quantities. Lastly, a high dependency on the Russian market for most agricultural products poses significant risks, as Russia is not a stable trade partner.
Given this context, agricultural subsidies are a highly important topic in Georgia. The Georgian government implements various subsidy programs to support agricultural sectors such as fruit production, viticulture, hazelnut farming, and wheat production. These initiatives aim to promote the sales of grapes, non-standard apples, and tangerines, enhance hazelnut production, and ensure food security by subsidizing essential staples like wheat, particularly during the Covid-19 pandemic.
Starting from 2014 to 2024 (Figure 1), the share of subsidies in total agricultural expenditure has followed an increasing trend, ranging from 21.4 percent in 2014 to peaking at 67.5 percent in 2021. In 2024 the respective share is 54.1 percent. A decline occurred in 2022–2023, following the stabilization of the Covid-19 pandemic. Apart from this, the share of subsidies within agricultural expenditures has been increasing over the last ten years.
Figure 1. Total and subsidy expenditures on agriculture, million GEL (2014-2024)
Source: Geostat, 2025.
While these programs are designed to assist farmers and increase sales, how these subsidies support in addressing the mentioned structural challenges – therefore advancing the effectiveness of the sector – is under question.
The Grape Subsidy Programs
The grape subsidy programs in Georgia are primarily aimed at supporting viticulture in key wine-producing regions, such as Kakheti, Racha-Lechkhumi, and Kvemo Svaneti. These subsidies were designed to stabilize farmers’ incomes and ensure smooth harvests, to guarantee that even lower-quality grapes will be sold, particularly for grape varieties used in wine production. In general, the government uses two types of subsidies: direct and indirect. Direct subsidies involve paying farmers a certain amount of money per kilogram of grapes. Indirect subsidies are implemented through state-owned companies that are responsible for purchasing grapes from farmers.
Georgia’s grape subsidy program (direct subsidies) was introduced in 2008 and has been implemented every year except for in 2018 and 2019. Starting from 2014, the government provided substantial direct financial support to grape producers. However, starting in 2017, direct subsidies began to decline sharply, and by 2018–2019, the government announced that it would no longer directly subsidize the grape harvest. However, during this period, the state’s grape purchasing program remained in place, purchasing any surplus grapes left on the market after private acquisitions.
The Covid-19 pandemic in 2020 prompted a renewed surge in subsidies, with financial support reaching its highest levels in years. This elevated support continued until 2022 but was significantly reduced again in 2023 (by 63 percent), following a decline in production (Figure 2).
Figure 2. Grape production, subsidies and wine exports (2014-2023)
Source: Geostat, 2025.
Grape production has generally followed an upward trend, with record harvests in 2019 and 2020. Given the absence of direct subsidies in 2017 and 2018, the effect of subsidies on production levels is questionable. In more recent years, production has become more volatile, displaying a noticeable decline by 2023.
Wine exports, a crucial part of Georgia’s economy, have grown steadily, with volumes peaking in 2022, and persisting at high levels ever since. Export revenues have also increased consistently, reaching an all-time high in 2024, according to preliminary data.
The main destination for the Georgian wine sector is CIS countries. Russia accounts for the largest share among the CIS, with an average of 75.4 percent, between 2014-2024. Russia’s share has been increasing in recent years, reaching 85.8 percent in 2024 (among CIS countries). The average share of exports to the EU of total exports is 10 percent (Figure 3).
Figure 3. Wine exports by country groups (2014-2024)
Source: Geostat, 2025.
Although subsidies played a key role in revitalizing Georgia’s wine industry following the collapse of the Soviet Union, especially as grape production and processing have increased over the years, their long-term impact have been problematic (Ghvanidze, Bitsch, Hanf, & Svanidze, 2020). Since subsidies were introduced in 2008, Georgia’s grape market has become heavily distorted, with prices shaped by government support rather than supply and demand dynamics.
Even though a significant portion of government funding for the sector is allocated to subsidies, the way in which subsidies affect grape production levels is not obvious. Other sector insufficiencies, such as quality issues and exporting market diversification are inadequately addressed. Grape quality remains a key issue, as farmers lack incentives to improve production practices, knowing that the government will purchase their yield regardless. Additionally, Georgia’s heavy reliance on its main export partner, Russia, poses significant risks, and the share of exports to EU countries has not seen substantial growth. Overall, since the subsidies aim to stabilize producers’ income rather than to address structural issues in the sector, they may be considered social support.
The Apple Subsidy Program
The apple subsidy program in Georgia was introduced in 2014 to support the sale of non-standard apples after market prices dropped to a record low 0.02 GEL. Non-standard apples are damaged fruits that fall from trees due to wind, hail, or other natural factors. Typically unfit for direct consumption, these lower-quality apples are primarily used by factories to produce apple concentrate. The program aimed to stabilize prices and provide financial relief to farmers. Processing companies received financial support for each kilogram of non-standard apples purchased.
The program was discontinued between 2015 and 2019, before it resumed in 2020. The number of companies involved in purchasing non-standard apples for further processing ranges from 12 to15 over the years.
As for apple production levels, although there were significant production surges in 2016, 2018, and 2020, these increases have been volatile and unstable.
Figure 4. Apple production, subsidies and exports (2014-2023)
Source: Geostat, 2025.
In terms of exports, the volume increased sharply between 2018 and 2019, reaching its peak in 2021 before gradually declining. Most apple exports are directed to CIS countries, with Russia accounting for an average of 94 percent between 2018 and 2024. In contrast, the EU’s share remains minimal, averaging less than 1 percent, with no exports recorded to the EU in half of the considered years.
Figure 5. Apple exports by country group (2014-2024)
Source: Geostat, 2025.
While apple production is highly vulnerable to weather conditions, the adoption of insurance remains low. The provided subsidy program supports farmers in producing lower-quality non-standard apples, thus limiting the incentives to enhance product quality, productivity, or production practices, as farmers rely on the government to purchase their produce regardless. Similar to the grape industry, government support in the apple market functions more as a social assistance rather than a tool for industry advancement.
The Hazelnut Subsidy Program
Georgia introduced the Hazelnut Production Support Program in 2022 to enhance competitiveness, assist farmers, and improve disease management. The program registered hazelnut orchards in a national cadaster, enabling better monitoring and targeted support, to subsidize the purchase of pesticides and agrochemicals essential for hazelnut care and cultivation. The program has continued in 2023 and 2024, with subsidies amounting 22 and 22.6 million GEL, respectively.
Hazelnut production in Georgia has been highly volatile in the past decade. The sector experienced its most severe crisis in 2017-2018 when fungal diseases and an Asian stink bug (Pharosana) invasion devastated yields. Consequently, both the quantity and quality of hazelnut production declined. In 2019, the production began to recover, peaking in 2021. However, unfavorable weather conditions resulted in a decline in 2022, with only a partial rebound in 2023.
Figure 6. Hazelnut production and exports (2014-2023)
Source: Geostat, 2025.
Hazelnut is mainly exported to EU countries, with an average share of 65.3 percent, between 2014 and 2024. The share of CIS countries in this period is 20.2 percent. However, the share exported to EU countries has been declining 2023 and 2024, to 52.4 and 56.7 percent, respectively.
Figure 7. Hazelnut exports by country group (2014-2024)
Source: Geostat, 2025.
The subsidy scheme in the hazelnut sector seems to be more targeted at the issues the sector is facing, compared to the other discussed programs. The effects are however yet to be explored as the program began in 2022. However, several challenges remain, such as insufficient technical facilities for drying and storing goods essential for ensuring the quality of products (Gelashvili, Deisadze & Seturidze, 2023).
Conclusion and Recommendations
Although the government of Georgia provides substantial support for the agricultural sector, it still suffers from various challenges. Product quality, high vulnerability to weather events and export dependency on unstable partners are major issues for the grape and apple sectors. Further, the effectiveness of the direct financial support and the corresponding incentives within these sectors can be questioned.
For these crops, the subsidy programs seem to function more as social assistance rather than tools for industry development. In the grape sector, guaranteed government purchases reduce incentives for farmers to improve grape quality. Similarly, the apple subsidy program encourages the cultivation of non-standard apples, as farmers rely on state-backed purchases rather than market-driven quality improvements. Apple production has also shown significant volatility over the years, further highlighting the sector’s instability.
Additionally, heavy dependence on Russia as a primary export market for these crops presents economic risks. Diversification, particularly to the EU, has remained limited.
As for the hazelnut sector, the subsidy program aims to address some of the structural challenges, while this sector also relies less on the Russian market. However, some issues with infrastructural equipment remain unresolved.
Overall, the share of subsidies in agriculture is very high; further, the design of the programs mainly prioritizes short-term income stability for farmers rather than long-term market competitiveness and sectoral development. To address the discussed systemic challenges, it is essential to develop targeted policies tailored to the specific needs of each sector. While the priorities may differ across each crop, several key areas require focused attention:
- Quality of Products – Enhancing product quality through ensuring food safety standards, improved farming and manufacturing practices, and better regulatory frameworks can help increase competitiveness in both domestic and international markets.
- Market Diversification – Strengthening ties with new international partners and improving branding strategies can help industries access new markets and reduce risks associated with economic or political fluctuations in dominant trade partners.
- Infrastructure Development – Poor infrastructure remains a challenge for the sector. Investments in post-harvest drying and storage facilities, as well as modern machinery and equipment, will enhance efficiency, reduce losses, and improve product quality.
- Adoption of innovative farming practices– Adopting innovative farming practices boosts productivity, lowers costs, and enhances sustainability. It helps farmers adapt to changing weather conditions, making agriculture more efficient, environmentally friendly, and resilient.
By addressing these fundamental issues, policies can play a role in contributing to the long-term stability and growth of the agricultural sector, ultimately strengthening the economy and increasing global competitiveness.
References
- Gautam, M. (2015). Agricultural Subsidies: Resurging Interest in a Perennial Debate. Indian Journal of Agricultural Economics.
- Gelashvili, S., Deisadze, S., & Seturidze, E. (2022). An Overview of the Georgian Wine Sector.
- Gelashvili, S., Deisadze, S., & Seturidze, E. (2023). Overview of the hazelnut sector in Georgia: past trends and the way forward. Tbilisi: ISET Policy Institute.
- Ghvanidze, S., Bitsch, L., Hanf, J. H., & Svanidze, M. (2020). “The Cradle of Wine Civilization” – Current Developments in the Wine Industry of the Caucasus. Caucasus Analytical Digest, 117, 9-15.
- Jayne, T., & Rashid, S. (2013). Input Subsidy Programs in Sub-Saharan Africa: A Synthesis of Recent Evidence. Agricultural Economics, 44, 547-562.
- Schwartz, G., & Clements, B. (1999). Government subsidies. Journal of Economic Surveys, 13(2), 119-148. doi:10.1111/1467-6419.00079
- State Audit Office of Georgia. (2023). Audit Report on the Development and Management of the State Agricultural Insurance Program.
Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.
Exposure to Violence and Prosocial Attitudes
This policy brief examines the academic literature on the impact of conflict exposure on pro-social behavior, a crucial component of resilience and societal cohesion. It also explores potential implications for public opinion, particularly in relation to Ukraine’s prospective EU accession and foreign relations.
Introduction
Since the full-scale invasion of Ukraine began on February 24, 2022, Russian forces have launched daily attacks with varying intensity. Living in a conflict zone profoundly affects individuals in multiple dimensions, including physical and mental health, as well as economic and social conditions. While reports often focus on the destruction of physical and human capital, social capital is equally affected by violence, influencing community resilience, cohesion, and cooperation. In conflict settings, identity can become more pronounced, particularly in distinguishing allies from adversaries.
This policy brief overviews the academic literature on this topic; the impact of conflict exposure on pro-social behavior broadly defined. This literature primarily examines post-conflict settings within the broader discourse on sustaining peace. It focuses on individuals directly engaged in combat or civilians directly affected by violence, particularly regarding the reintegration of former combatants and the rehabilitation of affected populations. As discussed below, results vary, depending on indicators used and the specific context. There is more consistent support for an impact on cooperation than on trust for instance. Another key finding in the literature is the differential behavior towards in-group members – those with whom individuals identify – versus out-group members, raising important questions about national identity and attitudes towards foreign allies. Based on this literature, the brief proceeds to discuss potential implications for public attitudes in Ukraine, focusing on Ukraine’s prospective EU accession.
Literature Overview
This review focuses on the empirical literature, though the theoretical basis spans psychology and the social sciences. Post-traumatic growth theory posits that adversity can foster positive change, whereas post-traumatic withdrawal theory suggests that violence exposure leads to distrust and social withdrawal. Economic arguments emphasize the need for rebuilding, enhanced safety concerns, or reduced time constraints for civic participation due to economic disruptions. Other perspectives highlight the detrimental effects of fragmented communities, given that trust and cooperation take time to develop, or suggest that individuals directly involved in violence may face social ostracization (see Fiedler 2023 for a detailed discussion).
Empirical studies on pro-social behavior employ diverse methodologies and data, including survey responses, indicators of political engagement, and controlled experiments measuring cooperation and trust. Methodology and research design vary, but most studies compare those with direct exposure to violence (treatment group) to those indirectly exposed (control group) within a post-conflict context. It is thus important to note that even the control group experiences some degree of conflict-related impact, meaning that studies specifically capture the effects of direct exposure.
Fiedler (2023), in a recent overview, categorizes the impact of violence into three main domains: personalized and political trust, cooperation, and political engagement. Most studies suggest a negative effect on trust, as seen in Kosovo (Kijewski & Freitag, 2018) and across Europe, the South Caucasus, and Central Asia post-World War II (Grosjean, 2014). Bauer et al. (2016) conducted a meta-analysis of 16 early studies measuring the effects of war violence on social participation, cooperation, and trust. When it came to trust, no significant impact of exposure to violence was found. Cassar et al. (2013) found that Tajik civil war survivors exhibited lower trust in close neighbors but not distant villagers, suggesting that intra-community political divisions played a role. However, a small number of studies report positive effects, such as Hall & Werner (2022), who found that victimized Syrian and Iraqi refugees in Turkey exhibited higher generalized trust.
In terms of cooperation, early studies overwhelmingly support a positive effect, including the meta-analysis of Bauer et al. (2016). For example, Bauer et al. (2014) held experimental games in Sierra Leone and Georgia, demonstrating that those directly exposed to violence exhibited greater altruism and inequality aversion. More recent work has come to different conclusions, however. Hager et al. (2019) found that Uzbek victims of violence in Kyrgyzstan were less cooperative in experimental games with both in-group and out-group members. Similarly, Cecchi & Duchoslav (2018) found that violence-exposed caregivers in Uganda contributed less in public goods games.
When it comes to political engagement, most studies find a positive effect, including the meta-analysis by Bauer et al. (2016) looking at participation in social groups and political engagement. Early and influential studies by Bellows & Miguel (2006, 2009), found that individuals in Sierra Leone with direct war exposure were more likely to participate in community meetings, elections, and social or political groups. Interestingly, while Kijewski & Freitag (2018) found that violence reduced trust in Kosovo, Freitag et al. (2019) found increased political participation in the same setting. Grosjean (2014) also reported a negative effect on trust but found that conflict victims were more likely to engage in civic organizations and collective action. These findings suggest that broad measures of prosocial behavior may be overly simplistic.
A common, and important, finding in much of the literature is with regards to differential behavior towards in-groups and out-groups. Bauer et al. (2014) found that exposure to violence increased altruism and inequality aversion only when interactions occurred within the in-group. Similar findings emerge in studies on soccer players in Sierra Leone (Cecchi et al., 2016) and trust experiments in Colombia (Francesco et al., 2023). Calvo et al. (2019) found that in conflict-affected areas of Mali, participation increased in kinship-based groups while it decreased in more inclusive organizations. Similarly, Mironova & Whitt (2016) found that Kosovars exhibited greater altruism and cooperation when interacting with in-group members. These findings align with research on parochial altruism in general, where cooperation and altruistic behavior are evolutionarily linked to in-group solidarity in response to external threats (e.g. Bernhard et al., 2006, Tajfel et al., 1979). There is thus a risk that social identity becomes more based on a narrow in-group (defined by ethnicity, religion, or language) potentially exacerbating societal divisions.
Implications for Ukraine
What do these insights imply for Ukraine? Given the context-dependent nature of the literature, definitive conclusions are challenging. Two studies on conflict exposure in eastern Ukraine offer preliminary insights. Mironova & Whitt (2021) examined fairness preferences among young Ukrainian men in Donbas, finding that, while no bias against ethnic Russians existed at the onset of violence in 2014, such bias increased after a year of conflict – particularly among non-combatants, contradicting typical patterns in the literature. Coupe & Obrizan (2016) used survey data from November 2014, showing that direct exposure to violence affected political behavior: physical damage reduced voter turnout, while property damage increased support for Western-leaning parties and stronger opposition to Russian aggression.
The strong effect on non-combatants in Mironova & Whitt (2021) highlights a key limitation in the literature – findings on direct exposure may not generalize to entire populations under invasion. Comparing directly and indirectly exposed individuals does not capture the broader societal impact, potentially leading to an overly optimistic view of conflict-induced prosocial behavior. If everyone is negatively affected, those with direct exposure to violence may simply be impacted a little less.
Of particular interest is how the war shapes national identity, in-group perceptions, and political preferences. These dynamics matter for domestic cohesion, interethnic relations, and Ukraine’s foreign policy trajectory. Focusing on the latter, the EU and the U.S. have provided substantial support during the full-scale invasion but delays and insufficiencies in aid may influence perceptions of these allies. EU accession presents economic benefits but entails lengthy and costly reforms with uncertain outcomes. Additionally, shifting U.S. policies and emerging geopolitical alignments may alter Ukrainian attitudes toward Western institutions.
Terror management theory (Landau et al., 2004) suggests that fear strengthens support for charismatic leadership, which, in fragile democratic settings, may favor more authoritarian tendencies. If Western democratic institutions lose appeal, this could negatively impact Ukraine’s political engagement, trust in allies, and willingness to align with European values, which are crucial for successful EU integration.
Conclusions
This review examined the literature on exposure to violence and prosocial behavior, discussing implications for Ukraine’s societal resilience and international alignment. The findings suggest no universal relationship between conflict exposure and prosociality; instead, effects vary depending on the recipient of trust, cooperation, and engagement. Generally, prosocial behavior increases within in-groups, while attitudes toward out-groups may remain unchanged or worsen. In the Ukrainian context, this has ramifications for internal cohesion and external diplomatic relations, particularly regarding the country’s path toward EU membership.
References
- Bauer, M., Blattman, C., Chytilová, J., Henrich, J., Miguel, E., & Mitts, T. (2016). Can War Foster Cooperation? Journal of Economic Perspectives, 30(3), 249–274.
- Bauer, M., Cassar, A., Chytilová, J., & Henrich, J. (2014). War’s Enduring Effects on the Development of Egalitarian Motivations and In-Group Biases. Psychological Science, 25(1), 47–57.
- Bellows, J., & Miguel, E. (2006). War and Institutions: New Evidence from Sierra Leone. American Economic Review, 96(2), 394–99.
- Bellows, J., & Miguel, E. (2009). War and Local Collective Action in Sierra Leone. Journal of Public Economics, 93(11–12), 1144–57.
- Bernhard, H., Fehr, E., & Fischbacher, U. (2006). Group Affiliation and Altruistic Norm Enforcement. American Economic Review, 96(2), 217–221.
- Calvo, T., Lavallée, E., Razafindrakoto, M., & Roubaud, F. (2019). Fear Not for Man? Armed Conflict and Social Capital in Mali. Journal of Comparative Economics, 48(2), 251–76.
- Cassar, A., Grosjean, P. A., Khan, F. J., & Lambert, M. (2022). Mothers, Fathers and Others: Competition and Cooperation in the Aftermath of Conflict. UNSW Business School Research Paper.
- Cecchi, F., Duchoslav, J. (2018). The Effect of Prenatal Stress on Cooperation: Evidence from Violent Conflict in Uganda. European Economic Review, 101, 35–56.
- Cecchi, F., Leuveld, K., & Voors, M. (2016). Conflict Exposure and Competitiveness: Experimental Evidence from the Football Field in Sierra Leone. Economic Development and Cultural Change, 64(3), 405-435.
- Coupé, T., & Obrizan, M. (2016). Violence and political outcomes in Ukraine—Evidence from Sloviansk and Kramatorsk. Journal of Comparative Economics, 44(1), 201-212.
- Fiedler, C. (2023). What Do We Know about How Armed Conflict Affects Social Cohesion? A Review of the Empirical Literature. International Studies Review.
- Francesco, B., Gómez, C., & Grimalda, G. (2023). Crime-related exposure to violence and prosocial behavior: Experimental evidence from Colombia. Journal of Behavioral and Experimental Economics, 104.
- Freitag, M., Kijewski, S., & Oppold, M. (2019). War Experiences, Economic Grievances, and Political Participation in Postwar Societies: an Empirical Analysis of Kosovo. Conflict Management and Peace Science, 36(4), 405–24.
- Grosjean, P. (2014). Conflict and Social and Political Preferences: Evidence from World War II and Civil Conflict in 35 European Countries. Comparative Economic Studies, 56(3), 424–51.
- Hager, A., Krakowski, K., & Schaub, M. A. X. (2019). Ethnic Riots and Prosocial Behavior: Evidence from Kyrgyzstan. American Political Science Review, 113(4), 1029–44.
- Hall, J., & Werner, K. (2022). Trauma and Trust: How War Exposure Shapes Social and Institutional Trust among Refugees. Frontiers in Psychology, 13, 786838.
- Kijewski, S., & Freitag, M. (2018). Civil War and the Formation of Social Trust in Kosovo: Post-traumatic Growth or War-Related Distress? Journal of Conflict Resolution, 62(4), 717–42.
- Landau, M. J., Solomon, S., Greenberg, J., Cohen, F., Pyszczynski, T., Arndt, J., Miller, C. H., Ogilvie, D. M., & Cook, A. (2004). Deliver us from Evil: The Effects of Mortality Salience and Reminders of 9/11 on Support for President George W. Bush. Personality and Social Psychology Bulletin, 30(9), 1136–1150.
- Mironova, V., & Whitt, S. (2016). Social Norms after Conflict Exposure and Victimization by Violence: Experimental Evidence from Kosovo. British Journal of Political Science, 48(3), 749–65.
- Mironova, V., & Whitt, S. (2021). Conflict and parochialism among combatants and civilians: Evidence from Ukraine. Journal of Economic Psychology, 86.
- Tajfel, H., Turner, J. C., Austin, W. G., & Worchel, S. (1979). An integrative theory of intergroup conflict. Organizational Identity: A Reader, 56-65.
Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.
Energy Security at a Cost: The Ripple Effects of the Baltics’ Desynchronization from the BRELL Network
The Baltic States’ desynchronization from the BRELL network on February 7, 2025, cut ties with Russia and Belarus, ending electricity trade. Though the transition was smooth with no outages, recent underwater cable disruptions have highlighted vulnerabilities, raising energy security concerns. These events underscore the importance of both diversifying and decentralizing power systems, drawing lessons from Ukraine’s electricity market, which has remained operational despite sustained Russian attacks.
The Baltics’ power system was part of a large Russian-operated synchronous electricity system known as BRELL, which connected the electricity transmission systems of Belarus, Russia, Estonia, Latvia, and Lithuania (Figure 1). The desynchronization from BRELL and the integration into the European grid have been discussed since 2007, when the Prime Ministers of the Baltic States declared desynchronization as the region’s strategic priority. In 2018, a decision was made to join the Continental European Synchronous Area through a connection with Poland, leading to significant investments – financially supported by the European Commission – to ensure adequate infrastructure. Fully committing to their priority, the Baltic’s desynchronized completely from BRELL on February 7th, 2025.
Figure 1. The BRELL power ring
Source: Karčiauskas (2023)
A Successful Physical (De)synchronization
The desynchronization process proceeded smoothly, with no blackouts. This success was anticipated, given the project’s meticulous planning over several years. A comparable example is Ukraine, which disconnected from the Russian and Belarusian power systems less than a month after Russia’s full-scale invasion in 2022. Ukraine then synchronized with the Continental European power grid ENTSO-E, an event that had been in preparation since 2017.
After the desynchronization, the Baltic states temporarily operated in island mode, relying entirely on domestic generation for all grid operations. To maintain system stability, the commercial capacity of interconnectors with the Nordics (whose regional group is not part of the Continental European Synchronous Area) was reduced, ensuring they could serve as reserves in case of major generator outages. The NordBalt cable is one such connector linking Sweden’s SE4 region and Lithuania.
However, conditions are gradually returning to normal. As of February 17, 2025, 700 MW is now available for commercial trading, as shown in Figure 2. Despite this progress, the commercial trading capacity of the interconnector with Poland (the LitPol line) remains heavily restricted and is primarily used to maintain system stability.
Figure 2. Day-ahead commercial transfer capacities on the Nordic interconnectors around the desynchronization
Source: Nord Pool
The Baltic region’s synchronization with the European grid is currently achieved through a 400 kV overhead power line connecting Lithuania and Poland. A second link, the Harmony Link, an underground cable, is planned to become operational by 2030. This makes the existing interconnection an essential part of regional infrastructure and a potential security risk, particularly given the recent sabotage of cables in the Baltic Sea. In response to these threats, Lithuania has increased surveillance of the NordBalt cable. The country’s prime minister has estimated the cost of securing the Baltic cables at €32-34 million, seeking EU support for its funding. The government has also strengthened the protection measures. Initially, security was outsourced to a private security company, but plans are in place for the country’s Public Security Service (Viešojo saugumo tarnyba) to take over in spring 2025. Further, in preparation for the Baltics’ full desynchronization, the Polish Transmission System Operator deployed helicopters to patrol the interconnection, to enhance the security of the infrastructure.
From Trade Interruption to Infrastructure Sabotage
The most significant short-term impact of the desynchronization from the BRELL is the limitation of electricity trade for the Baltic states. The desynchronization has affected reserve balancing in the Baltic region, forcing the three states to rely more on their internal generation for system stability. This has resulted in reduced generation capacity for commercial trade, as the states must be prepared to again operate in island mode in case of an outage on the LitPol cable. Until February 19, 2025, the LitPol line remained unused for commercial trading. However, gradual increases are expected to eventually allow for 150 MW commercial trade between the Polish area and the Baltics, a significant reduction from the 500 MW previously available. This limited trading capacity could lead to higher prices in the Baltics, as the region is a net importer of electricity.
This is not the first time the Baltics have faced trade disruptions. In November 2020, after the construction of a Belarusian nuclear power plant near the Lithuanian border, Lithuania, followed by Latvia and Estonia, limited commercial electricity exchanges with Russia and Belarus. Furthermore, on May 15, 2022, electricity trade between Russia and Finland was halted, followed by the closure of the Kaliningrad-Lithuania connection the next day. While this event led to no blackouts, it clearly impacted the region’s price volatility (Lazarczyk & Le Coq, 2023).
Recently, the region has experienced sabotage to underwater interconnectors, significantly impacting electricity trade between the Nordics and the Baltics. On December 25, 2024, the Estlink 2 cable, one of two connections between Finland and Estonia, was cut, reducing transmission capacity between the two regions. Repair costs are expected to reach several million Euros. As disclosed via Nord Pool’s Urgent Market Message, repairs are expected to last until August 2025 – stressing the system. As Estlink 2 is offline, the Baltic system is not fully operating. If another major component fails, there may be insufficient capacity to maintain grid stability, increasing the risk of outages or the need for emergency interventions.
With the complete disconnection from the Russian and Belarusian power grids, Russia no longer has direct control over the Baltic electricity trade, effectively eliminating the risk of trade disruptions from Russia. However, a new energy threat has emerged: infrastructure sabotage. Although the perpetrators of recent sabotage incidents have not been clearly identified, both Lazarczyk & Le Coq (2023) and Fang et al. (2024) emphasize Russia’s strategic incentives to engage in such actions to maintain its geopolitical influence and discourage neighboring countries from reducing their energy dependence. Sabotaging critical infrastructure presents another efficient method of weaponizing electricity, particularly in the current context of limited Nord Pool imports and the Baltic States’ insufficient integration with the broader European grid.
From Diversification to Decentralization: Responses to Electricity Infrastructure Threats
The Baltic States have diversified their domestic energy supply sources to address the electricity infrastructure threat. In 2024, Estonia’s parliament approved the development of nuclear energy, with Fermi Energia planning to build two 300 MW light-water reactors. Other projects include a hydrogen-ready gas plant in Narva, which is expected to be completed by 2029, as well as an expansion of wind power capacity. While there was some support for extending the use of oil-fired plants in Estonia, their competitiveness has been undermined by high carbon prices and the closure of domestic oil fields. Elering, the Estonian Transmission system operator, has also begun long-term procurement to acquire 500 MW of new generation and storage for frequency management to ensure reserve capacity.
However, diversification alone will not be sufficient to address the challenges currently faced by the Baltic States. Incidents like the cutting of underwater cables underscore the growing need to decentralize the power system. Large, centralized power plants are more vulnerable to targeted attacks compared to decentralized energy systems. As a result, connected microgrids seem to be a viable solution for future energy resilience, as they can maintain functionality even when localized damage occurs. Again, Ukraine’s experience demonstrates the benefits of decentralization. Since the onset of the war, Ukraine has faced both physical and cyberattacks but has strengthened its energy resilience by decentralizing its system and expanding wind and solar power (Eurelectric, 2025). This approach has proven effective: while a single missile could destroy a nearly gigawatt-scale power plant, it would only damage an individual wind turbine or a small section of solar panels, significantly limiting the overall impact.
The desynchronization of the Baltic States from the BRELL network marked a complete break with Russia and Belarus, effectively ending any possibility of electricity trade between these countries and the Baltic region. This transition was successfully completed without any power outages. While the primary goal was to enhance energy security in the Baltics, several challenges remain, as highlighted in this policy brief. Recent disruptions to underwater cables, as well as Russia’s attacks on Ukraine’s electricity market, underscore the urgent need for both diversification and decentralization to strengthen the region’s energy security. While energy supply diversification reduces supply chain dependencies, decentralization enhances resilience against targeted attacks, creating a more robust and flexible energy system.
References
- Eurelectric, 2025, Redefining Energy Security In the age of electricity, Lexicon.
- Fang, S., Jaffe, A. M., Loch-Temzelides, T., and C.L. Prete. (2024). Electricity grids and geopolitics: A game-theoretic analysis of the synchronization of the Baltic States’ electricity networks with Continental Europe. Energy Policy, 188, 114068.
- Karčiauskas, J. (2023). Lithuania External Relations Briefing: Synchronization of the Baltic Electricity Network and Breaking Dependence on Russian Energy Market. China CEE Institude Weekly Briefing 2023 Eylül, 4, 3.
- Lazarczyk, E. and Le Coq, C. (2023). Power coming for Russia and Baltic Sea region’s energy security, Energiforsk report.
- Lazarczyk, E. and Le Coq, C. (2022). Can the Baltic States Do Without Russian Electricity?, FREE Policy Brief.
Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.
Ukraine’s Fight Is Our Fight: The Need for Sustained International Commitment
We are at a critical juncture in the defense of Ukraine and the liberal world order. The war against Ukraine is not only a test of Europe’s resilience but also a critical moment for democratic nations to reaffirm their values through concrete action. This brief examines Western support to Ukraine in the broader context of international efforts, putting the order of magnitudes in perspective, and emphasizing the west’s superior capacity if the political will is there. Supporting Ukraine to victory is not just the morally right thing to do, but economically rational from a European perspective.
As the U.S. support to the long-term survival of Ukraine is becoming increasingly uncertain, European countries need to step up. This is a moral obligation, to help save lives in a democratic neighbor under attack from an autocratic regime. But it is also in the self-interest of European countries as the Russian regime is threatening the whole European security order. A Russian victory will embolden the Russian regime to push further, forcing European countries to dramatically increase defense spending, cause disruptions to global trade flows, and generate another wave of mass-migration. This brief builds on a recent report (Becker et al., 2025) in which we analyze current spending to support Ukraine, put that support in perspective to other recent political initiatives, and discuss alternative scenarios for the war outcome and their fiscal consequences. We argue that making sure that Ukraine wins the war is not only the morally right thing to do, but also the economically rational alternative.
The International Support to Ukraine
The total support provided to Ukraine by its coalition of Western democratic allies since the start of the full-scale invasion exceeded by October 2024 €200 billion. This assistance, that includes both financial, humanitarian and military support, can be categorized in various ways, and its development over time can be analyzed using data compiled by the Kiel Institute for the World Economy. A summary table of their estimates of aggregate support is provided below.
A particularly relevant aspect in light of recent news is that approximately one-third of total disbursed aid has come from the United States. The U.S. has primarily contributed military assistance, accounting for roughly half of all military aid provided to Ukraine. In contrast, the European Union—comprising both EU institutions and bilateral contributions from member states—stands as the largest provider of financial support. This financial assistance is crucial for sustaining Ukraine’s societal functions and maintaining the state budget.
Table 1. International support to Ukraine, Feb 2022 – Oct 2024
Source: Trebesch et al. (2024).
Moreover, the EU has signaled a long-term commitment to provide, in the coming years, an amount comparable to what has already been given. This EU strategy ensures greater long-term stability and predictability, guaranteeing that Ukraine has reliable financial resources to sustain state operations in the years ahead. Consequently, while a potential shift in U.S. policy regarding future support could pose challenges, it would not necessarily be insurmountable.
What is crucial is that Ukraine’s allies remain adaptable, and that the broader coalition demonstrates the ability to adjust its commitments, as this will be essential for sustaining the necessary level of assistance moving forward.
Putting the Support in Perspective
To assess whether the support provided to Ukraine is truly substantial, it is essential to place it in context through meaningful comparisons. One approach is to examine it in historical terms, particularly in relation to past instances of large-scale military and financial assistance. A key historical benchmark is the Second World War, when military aid among the Allied powers played a decisive role in shaping the outcome of the conflict. Extensive resources were allocated to major military operations spanning multiple continents, with the United States and the United Kingdom, in particular, dedicating a significant share of their GDP to support their allies, including the Soviet Union, France, and other nations. As seen in Figure 1, by comparison, the current level of aid to Ukraine, while substantial and essential to its defense, remains considerably smaller in relation to GDP.
Figure 1. Historical comparisons
Source: Trebesch et al. (2024).
Another way to assess the scale of support to Ukraine is by comparing it to other major financial commitments made by governments in response to crises. While the aid allocated to Ukraine is significant in absolute terms, it remains relatively modest when measured against the scale of other programs, see Figure 2.
A recent example is the extensive subsidies provided to households and businesses to mitigate the impact of surging energy prices since 2022. Sgaravatti et al. (2021) concludes that most European countries implemented energy support measures amounting to between 3 and 6 percent of GDP. Specifically, Germany allocated €157 billion, France and Italy each committed €92 billion, the UK spent approximately €103 billion. These figures represent 5 to 10 times the amount of aid given to Ukraine so far, with some countries, such as Italy, allocating even greater relative sums. On average, EU countries have spent about five times more on energy subsidies than on Ukraine aid. Only the Nordic countries and Estonia have directed more resources toward Ukraine than toward energy-related support. Although not all allocated funds have been fully disbursed, the scale of these commitments underscores a clear political and financial willingness to address crises perceived as directly impacting domestic economies.
Figure 2. EU response to other shocks (billions of €)
Source: Trebesch et al. (2024).
Another relevant comparison is the Pandemic Recovery Fund, also known as Next Generation EU. With a commitment of over €800 billion, this fund represents the EU’s comprehensive response to the economic consequences of the Covid-19 pandemic. Again, the support to Ukraine appears comparatively small, about one seventh of the Pandemic Recovery Fund.
The support to Ukraine is also much smaller in comparison to the so-called “Eurozone bailout”, the financial assistance programs provided to several Eurozone member states (Greece, Ireland, Spain and Portugal) during the sovereign debt crisis between 2010 and 2012. The programs were designed to stabilize the economies hit hard by the crisis and to prevent the potential spread of instability throughout the Eurozone.
Overall, the scale of these commitments underscores a clear political and financial willingness and ability to address crises perceived as directly impacting domestic citizens. This raises the question of whether the relatively modest support for Ukraine reflects a lack of concern among European voters. However, this does not appear to be the case. In survey data from six countries – Belgium, Germany, Hungary, Italy, the Netherlands, and Poland – fielded in June 2024, most respondents express satisfaction with current aid levels, and a narrow majority in most countries even supports increasing aid (Eck and Michel, 2024).
A further illustration comes from the Eurobarometer survey conducted in the spring of 2024 which asked: “Which of the following [crises] has had the greatest influence on how you see the future?”. Respondents could choose between different crises, including those mentioned above, and the full-scale invasion of Ukraine.
Figure 3 illustrates the total commitments made by EU countries for Ukraine up until October 31, 2024, compared to other previously discussed support measures, represented by the blue bars. The yellow bars, on the other hand, show a counterfactual allocation of these funds, based on public priorities as indicated in the Eurobarometer survey. Longer yellow bars indicate that a higher proportion of respondents perceived this crisis as having a greater negative impact on their outlook for the future. By comparing the actual commitments (blue bars) with this hypothetical allocation (yellow bars)—which reflects how resources might have been distributed if they aligned with the population’s stated priorities—it becomes evident that there is substantial public backing for maintaining a high level of support for Ukraine. The results show that the population prioritizes the situation in Ukraine above several other economic issues, including those that directly affect their own personal finances.
Figure 3. Support to Ukraine compared to other EU initiatives – what do voters think?
Source: Trebesch et al. (2024); Niinistö (2024); authors’ calculations.
The Costs of Not Supporting Ukraine
When discussing the costs of support to Ukraine it is important to understand what the correct counterfactual is. The Russian aggression causes costs for Europe irrespective of what actions we take. Those costs are most immediately felt in Ukraine, with devastating human suffering, the loss of lives, and a dramatic deterioration in all areas of human wellbeing. Also in the rest of Europe, though, the aggression has immediate costs, in the economic sphere primarily in the form of dramatically increased needs for defense spending, migration flows, and disruptions to global trade relationships. These costs are difficult to determine exactly, but they are likely to be substantially higher in the case of a Russian victory. Binder and Schularik (2024) estimate increased costs for defense, increased refugee reception and lost investment opportunities for the German industry at between 1-2 percent of GDP in the coming years. As they put it, the costs of ending aid to Ukraine are 10-20 times greater than continuing aid at Germany’s current level.
Any scenario involving continued Russian aggression would demand substantial and sustained economic investments in defense and deterrence across Europe. Clear historical parallels can be drawn looking at the difference in countries’ military spending during different periods of threat intensity. Average military spending in a number of Western countries during the Cold War (1949-1990) was about 4.1 percent of GDP, much higher in the U.S. but also in Germany, France and the UK. In the period after 1989-1991 (the fall of the Berlin Wall, the dissolution of the Soviet Union), the amounts fell significantly. The average for the same group of countries in this period is about 2 percent of GDP and only 1.75 percent if the U.S. is excluded.
Also after 1991 there is evidence of how perceived threats affect military spending. Figure 4 plots the change in military spending over GDP between 2014-2024 against the distance between capital cities and Moscow. The change varies between 0 (Cyprus) and around 2.25 (Poland) and shows a very clear positive correlation between increases in spending and proximity to Moscow. There has also in general been a substantial increase in military spending after 2022 in several European countries, but in a scenario where Russia wins the war, these will certainly have to be increased further and maintained at a high level for longer. An increase in annual military expenditure in relation to GDP in the order of one to two percentage points would mean EUR 200-400 billion per year for the EU, while the total EU support to Ukraine from 2022 to today is just over €100 billion.
Figure 4. Increase in military expenditures in relation to distance to Moscow
Source: SIPRI data, authors’ calculations.
A Russian victory would also have profound consequences for migration flows, with the most severe effects likely in the event of Ukraine’s surrender. The Kiel Institute estimates the cost of hosting Ukrainian refugees at €26.5 billion (4.2 percent of GDP) for Poland, one of the countries that received the largest flows. Beyond migration, a Russian victory would also reshape the global geopolitical order. Putin has framed the war as a broader conflict with the U.S. and its democratic allies, while an emerging alliance of Russia, Iran, North Korea, and China is positioning itself as an alternative to the Western-led system. A Ukrainian defeat would weaken the authority of the U.S., NATO, and the rules-based international order, potentially driving more nations in the Global South toward authoritarian powers for military and economic support. This shift could disrupt global trade, affect access to food, metals, and energy. Estimating the full economic impact of such a shift is difficult, but comparisons can be drawn with other global shocks. The European Union’s GDP experienced a significant contraction due to the Covid-19 pandemic, 5.9 percent contraction in real GDP according to Eurostat, 6.6 percent according to the European Central Bank. While the economy rebounded relatively quickly from the pandemic, a permanent geopolitical realignment caused by a Russian victory would likely have far more severe and lasting economic consequences.
Given that Ukraine is at the forefront of Russia’s aggression, its resilience serves as a critical test of Europe’s ability to withstand potential future threats. Thus, strengthening our own security and economic stability in the long term is inseparable from strengthening Ukraine’s resilience now. The fundamental difference lies in the long-term trajectory of these investments. In a scenario where Ukraine is victorious, military and financial aid during the war would eventually transition into reconstruction efforts and preparations for the country’s integration into the EU. This outcome is undeniably more favorable—both economically and in humanitarian terms—not only for Ukraine but for Europe as a whole. Therefore, an even more relevant question is whether the level of support is enough for Ukraine to win the war.
Is Sufficient Support Feasible?
Is it even reasonable to think that we in the West could be able to support Ukraine in such a way that they can militarily defeat Russia? Russia is spending more on its war industry than it has since the Cold War. In 2023, it spent about $110 billion (about 6 percent of GDP). By 2024, this figure is expected to have increased to about $140 billion (about 7 percent of GDP). These amounts are huge and represent a significant part of Russia’s state budget, but they are not sustainable as long as sanctions against Russia remain in place (SITE, 2024). For the EU, on the other hand, the sacrifices needed to match this expenditure would not be as great. The EU’s GDP is about ten times larger than Russia’s, which means that in absolute terms the equivalent amount is only 0.6-0.7 percent of the EU’s GDP. If the U.S. continues to contribute, the share falls to below 0.3 percent of GDP.
Despite the economic advantage of Ukraine’s allies over Russia, several factors could still shift the balance of power in Russia’s favor. One key issue is military production capacity—Russia has consistently outproduced Ukraine’s allies in ammunition and equipment. While Western economies have the resources to manufacture superior weaponry, actual production remains insufficient, requiring both increased capacity and political will. Another challenge is cost efficiency. Military purchasing power parity estimates suggest that Russia can produce approximately 2.5 times more military equipment per dollar than the EU, giving it a cost advantage in volume production. However, this does not fully compensate for its overall economic disadvantage, particularly when factoring in quality differences.
Manpower is also a critical factor. Russia’s larger population allows for sustained mobilization, but at a steep financial cost. Soldiers are recruited at a minimum monthly salary of $2,500, with additional bonuses bringing the first-year cost per recruit to three times the average Russian annual salary. Compensation for injured and fallen soldiers further strains state finances, with estimated payouts reaching 1.5 percent of Russia’s GDP between mid-2023 and mid-2024. Over time, these costs limit Russia’s ability to fund its war effort, making mass mobilization financially unsustainable.
Overall, advanced Western weaponry and superior economic capacity can match Russia’s advantage in manpower if the political will is there. Additionally, Russia’s already fragile demographic situation is deteriorating due to battlefield losses and wartime emigration. Any measure that weakens Russia’s economic capacity—particularly through sanctions and embargoes—diminishes the strategic advantage of its larger population and serves as a crucial complement to military and financial support for Ukraine.
Conclusion
Ukraine’s western allies have provided the country with substantial military and financial support since the onset of the full-scale invasion. Yet, relative to the gravity of the risks involved, previous responses to economic shocks, and citizens’ concerns about the situation, the support is insufficient. The costs of a Russian victory will be higher for Europe, even disregarding the human suffering involved. With U.S. support potentially waning, EU needs to pick up leadership.
References
- Becker, Torbjörn; and Anders Olofsgård; and Maria Perrotta Berlin; and Jesper Roine. (2025). “Svenskt Ukrainastöd i en internationell kontext: Offentligfinansiella effekter och framtidsscenarier”, Commissioned by the Swedish Fiscal Policy Council.
- Binder, J. & Schularick, M. (2024). “Was kostet es, die Ukraine nicht zu unterstützen?” Kiel Policy Brief No. 179.
- Eck, B & Michel, E. (2024). “Breaking the Stalemate: Europeans’ Preferences to Expand, Cut, or Sustain Support to Ukraine”, OSF Preprints, Center for Open Science.
- Niinistö, S. (2024) .“Safer Together – Strengthening Europe’s Civilian and Military Preparedness and Readiness” European Commission Report.
- Sgaravatti, G., S. Tagliapietra, C. Trasi and Zachmann, G. (2021). “National policies to shield consumers from rising energy prices”, Bruegel Datasets, first published 4 November 2021.
- SITE. (2024). “The Russian Economy in the Fog of War”. Commissioned by the Swedish Government.
- Trebesch, C., Antezza, A., Bushnell, K., Bomprezzi, P., Dyussimbinov, Y., Chambino, C., Ferrari, C., Frank, A., Frank, P., Franz, L., Gerland, C., Irto, G., Kharitonov, I., Kumar, B., Nishikawa, T., Rebinskaya, E., Schade, C., Schramm, S., & Weiser, L. (2024). “The Ukraine Support Tracker: Which countries help Ukraine and how?” Kiel Working Paper No. 2218. Kiel Institute for the World Economy.
Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.
Revisiting the Impact of Rising Gasoline Prices on Swedish Households
Sweden has a long-standing tradition of fuel taxation, but recent shifts in transport policy have significantly altered the cost of driving. This policy brief examines the impact of gasoline tax cuts and reductions in biofuel mandates introduced between May 2022 and January 2024. These measures, alongside a drop in global crude oil prices, have led to a 34 percent decline in pump prices, bringing the cost of driving to one of its lowest levels in the past 25 years. Using a comparative analysis with Denmark, the brief quantifies the impact of the tax cuts and biofuel policy changes, showing how they kept fuel prices lower. However, these short-term financial benefits have broader implications. Lower gasoline taxes have increased household exposure to crude oil price volatility and slowed electric vehicle adoption, reversing progress toward Sweden’s long-term climate targets. Given these trade-offs, the brief argues for a reassessment of transport policies to balance affordability with long-term environmental sustainability.
The Cost of Driving in Sweden
Sweden has a long history of fuel taxation, having introduced an excise tax on gasoline in 1924. For over seventy years (1951–2021), the nominal tax rate steadily increased without significant reductions. This trend stopped in 2022, when the first of a series of tax cuts was implemented on May 1. This shift in transport policy came in response to a 60 percent surge in gasoline prices between early 2021 and mid-2022. This price spike was driven by pandemic-related supply-chain issues and Russia’s invasion of Ukraine in February 2022. Moreover, the 2022 elections in Sweden, which brought a conservative coalition to power, further transformed transport policy, as the new government had campaigned on reducing pump prices.
In 2022, Celina Tippmann and I published two policy briefs on the impact of surging gasoline prices in Sweden. The first, titled The Impact of Rising Gasoline Prices on Swedish Households – Is This Time Different?, found that despite record-high real gasoline prices, driving was historically affordable due to improved fuel efficiency and rising real wages over the past three decades. The second brief, Who Benefitted from the Gasoline Tax Cut in Sweden?, examined Sweden’s first major gasoline tax cut in decades, implemented on May 1, 2022 in response to the surging price. We found that the tax cut was fully passed through to consumers but likely caused spill-over effects that raised gasoline prices in neighboring countries, shifting part of the burden onto their households.
In this brief, I analyze the developments since the May 2022 gasoline tax cut. This tax cut marked the beginning of significant changes to Sweden’s transport policies. While a part of the May tax cut was reversed by design (the majority of the 1.81 SEK (€0.17) per liter tax cut expired by October 2022), it was followed by the removal of subsidies for electric vehicles in November 2022 and additional tax cuts; one tax cut on January 1st, 2023, and a further reduction in gasoline tax rates on January 1st, 2024, alongside a lower biofuel mandate. Meanwhile, global crude oil prices dropped by more than a third since their June 2022 peak. Together, these changes have likely reduced the cost of driving using gasoline and diesel and created a relative cost advantage for vehicles with internal combustion engines.
Figure 1. Gasoline pump price: 2000-2024
Source: Monthly data on gasoline prices are provided by Drivkraft Sverige (2025).
Figure 1 illustrates the dramatic price movements over the last couple of years. After the sharp increase in gasoline prices from early 2021 to mid-2022, the subsequent drop has been equally dramatic. Since June 2022, pump prices have fallen by 34 percent, bringing real gasoline prices just below the 25-year average of 15 SEK per liter.
Figure 2. Gasoline expenditure per 100 km
Source: Trafikverket (2022) and Drivkraft Sverige (2025).
Furthermore, the recent drop in driving costs is even more dramatic if we factor in improvements in average fuel efficiency over time. New vehicles sold in Sweden today can drive 50 percent further on a liter of gasoline compared to the year 2000. Accounting for this, Figure 2 shows that the cost of driving is now 20 percent below the average cost over the last 25 years.
Lastly, real wage growth has further enhanced the affordability of driving. Since 1991, average real wages in Sweden have risen by nearly 60 percent. As a result, the cost of driving, measured as a share of income, has steadily declined. Figure 3 shows a temporary increase in driving costs in 2022, but today, households spend less than 40 percent of their hourly wage to drive 100 kilometers – a near-historic low.
Figure 3. Cost of driving as share of income
Source: Data on average hourly real wages are provided by Statistics Sweden (2025).
The Cost in the Counterfactual Scenario
While Figures 1–3 show the evolution of driving costs, they do not isolate the impact of recent transport policies. The causal effect of the tax cuts and changes to the biofuel mandate hinges on the pass-through rate to consumers and how much of the benefit of the policy changes has been captured by producers. In addition, we need to separate the price change that is due to policy changes from the part that is due to the falling crude oil price.
Two strategies are available to estimate the pass-through rate to households. The first involves using price elasticities of demand and supply for gasoline, where the relatively inelastic side captures most of the benefit from a tax reduction (Andersson and Tippmann, 2022). However, the unusual conditions in the gasoline market over the past few years – characterized by supply restrictions from underinvestment during the pandemic, sanctions on Russia following its invasion of Ukraine, and shifts in consumer travel behavior – have made elasticity estimates from historical data less reliable for assessing tax incidence today.
The second approach involves a comparative analysis, examining the evolution of gasoline pump prices in Sweden against those in a ”twin“ country – one similar to Sweden but unaffected by recent transport fuel policy changes. This is the method I adopt in this brief. A benefit of using a comparative analysis is that the crude oil price is not a confounder as it affects the gasoline price in the comparison country equally. I selected Denmark as the comparison unit due to its geographical proximity, socio-economic similarity, and minimal changes to gasoline tax rates over the past two and a half years (Drivkraft Danmark, 2025).
Figure 4. Gasoline pump price 2022-2024
Note: Gasoline prices in Sweden and Denmark are provided by CirkleK (2025). Daily exchange rates are provided by Riksbanken (2025). The horizontal lines indicate the four tax changes over the sample period.
Figure 4 shows that nominal gasoline prices in Sweden and Denmark closely tracked each other until the first tax cut on May 1, 2022. Following the tax cut, Sweden’s prices fell by an amount roughly equivalent to the tax cut. When part of this initial tax reduction was reversed on October 1, 2022, the price gap narrowed before widening again due to a new tax cut on January 1, 2023. The gap widened further at the start of 2024 with another tax cut and a reduction in the biofuel mandate (biofuel is typically much more expensive than crude oil). In total, the pump price in Sweden fell by more than 3 SEK relative to the counterfactual scenario. With a full pass-through of the tax cuts to consumers, approximately half of this reduction is attributed to the tax cuts, with the other half resulting from the reduced biofuel mandate (Andersson and Tippman, 2022).
It may seem surprising that a reduction of the biofuel mandate from 7.8 percent to 6 percent has such a significant impact on the pump price in Sweden. However, one needs to account for the indirect effect on the price of biofuel itself from a reduction in its demand. Sweden also reduced its biofuel mandate for diesel, from 30.5 percent to 6 percent, a far more drastic cut. Together, these reductions significantly lowered biofuel demand, likely driving down biofuel prices in the market and amplifying their impact on pump prices.
Conclusion
The cost of driving in Sweden is at a historic low. Over the past two and a half years, tax cuts and reductions in the biofuel mandate have significantly lowered pump prices, with the benefits passed directly to consumers. Compared to a scenario with no policy changes, Swedish households now enjoy drastically reduced costs at the pump. However, these short-term benefits come with a long-term risk that warrant careful consideration.
In our first policy brief in 2022, Celina Tippmann and I cautioned that reducing gasoline tax rates could encourage households to purchase less fuel-efficient vehicles, leaving them more vulnerable to future crude oil price spikes. Previously, excise taxes – comprising more than half of Sweden’s pump price – acted as a buffer against global oil price volatility. Lower fuel taxes now mean crude oil prices make up a larger share of the pump price, increasing price volatility and household exposure to market fluctuations.
Emerging evidence suggests that households are responding to the latest policy changes as anticipated. In 2024, the share of electric vehicles in new car sales dropped for the first time in years, from 38.7 percent to 35 percent, while average carbon emissions from new vehicles increased by 5 percent (Mobility Sweden, 2025), breaking a long-run downwards trend. This reversal of progress in emissions reductions makes achieving Sweden’s 2030 climate target – a 70 percent reduction in transport sector carbon emissions relative to 2010 – significantly more challenging.
While the election campaign promise from the conservative coalition of reducing gasoline prices may have been politically and electorally effective, its consequences on the transport market are becoming clearer. Swedish households have become more vulnerable to crude oil price volatility as they are buying less fuel-efficient vehicles, and progress toward emission reduction goals has stalled. As such, it is time for a more ambitious climate policy in the transport sector. Sweden should consider reintroducing higher gasoline tax rates and strengthening financial support for electric vehicle adoption. These measures would help balance the affordability of driving with the urgent need to meet climate objectives.
References
- Andersson, J. and Tippmann, C. (2022). “Who Benefitted from the Gasoline Tax Cut in Sweden?” FREE Policy Brief.
- CircleK. (2025). ”Drivmedelspriser – bensinpriser & dieselpriser.” Retrieved from: https://www.circlek.se/drivmedel/drivmedelspriser.
- CircleK. (2025). ”Brændstof-, El-, og fyringsoliepriser.” Retrieved from: https://www.circlek.dk/priser.
- Drivkraft Danmark (2025). “Priser Benzin.” Retrieved from: https://www.drivkraftdanmark.dk/priser/benzin-95-oktan/
- Drivkraft Sverige (2025). “Fakta & Statistik.” Retrieved from: https://drivkraftsverige.se/fakta-statistik/priser/
- Mobility Sweden. (2025). “Svagt fordonsår 2024 med ökade utsläpp, men viss ljusning 2025.” Retrieved from: https://mobilitysweden.se/statistik
- Riksbanken. (2025). ”Daily exchange rates SEK-DKK.” Retrieved from: https://www.riksbank.se/sv/statistik/.
- Statistics Sweden. (2025). “Average hourly wage statistics.” http://www.statistikdatabasen.scb.se.
- Trafikverket. (2022). “Vägtrafikens utsläpp 2021.” Tech. rep., Swedish Transport Administration.
Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.
Road Congestion Pricing with A Public Transport Cashback Mechanism
Traffic jams are a major problem in cities leading to wasted time, air pollution, reduced accessibility, and, in turn, lower economic activity. Transport economists widely agree that charging drivers fees for using busy roads during rush hours (congestion pricing) is the best answer to road congestion problems. However, such a policy is rarely used, mostly because people see it as unfair in how it affects different income groups. We propose an innovative personalized public transport cashback mechanism to make congestion pricing more acceptable. Recent surveys in Riga and Vienna show that people are more willing to support the introduction of congestion pricing when it includes a cashback component.
Road Congestion Pricing and Its Discontent
Road Congestion Pricing
Traffic jams happen when too many cars at the same place and at the same time use a road of a limited capacity. Building new roads or lanes is expensive, especially in cities, and it only provides short- and medium-term traffic improvements, with little impact on congestion in the long term (Ossokina et al., 2023; Hymel, 2019). Duranton and Turner (2011) show that when major roads are expanded, more people start using them and, over time, congestion returns to the same level as before. Meanwhile, a travel mode shift from cars to public transport and bicycles also requires investments and is difficult to implement in practice.
Dynamic congestion pricing, when road tolls vary based on the time of day, is designed to spread out traffic flow over time without the need to expand road infrastructure (Small and Verhoef, 2007). Notably, this approach does not aim to reduce the total number of cars on the road. Instead, it encourages them to spread their travel times more evenly, ensuring that the road capacity can handle the traffic without congestion.
Dynamic congestion pricing typically works as follows: there is no charge at night, the toll is small in the early morning, then it gradually increases during the morning until it reaches its peak. The toll then decreases in the afternoon before rising again during the evening. This system works in a congestion zone, which is usually the busiest areas of a city. When a car enters the zone, video cameras automatically identify it without stopping the car. There are no toll booths on the streets – an electronic system calculates the toll based on the time of day and charges the driver automatically through a linked account. Cities can tailor the system to fit their specific geography and infrastructure, offering exemptions for certain vehicles and pass-through traffic (for practical examples, visit the Swedish Transport Agency’s website to learn more about congestion pricing in Stockholm and Gothenburg).
By reducing the number of cars during congested hours, such dynamic pricing benefits both the city and its residents:
- (i) Drivers enjoy faster travel times as road toll allows them to gain time in exchange for money. For example, in the morning, drivers can leave for work later as they no longer need to account for time spent in traffic jams.
- (ii) Non-drivers enjoy congestion-free neighborhoods with improved air quality and overall higher quality of life.
- (iii) The city can tackle congestion without making large investments in new roads. The funds collected from drivers not only cover the toll system maintenance, but also contribute to the cost of the infrastructure they use. The funds may also be used to improve public transportation.
Low Public Acceptability
In light of the benefits of congestion pricing, it seems surprising that very few cities actually use it. Notable examples include London, Singapore, Stockholm and Gothenburg. New York City introduced its congestion charge on the 5th of January 2025, the first in the US. This stands in stark contrast to paid on-street parking, another transport policy measure that has been successfully implemented in almost every large city across Europe. The disparity arises because the general public often sees congestion pricing as an additional tax, believing it unfairly affects lower-income individuals. Presumably, low-income individuals have less flexible work schedules and fewer travel choices, making it harder for them to avoid traveling during high-toll periods (Selmoune et al., 2020). Moreover, they would spend a larger share of their income on road tolls compared to wealthier drivers, which makes congestion pricing a regressive policy.
Even though congestion pricing is not a tax and is not meant to redistribute funds, it may still appear as such to the public. This perception leads to vocal public resistance to road pricing which, in turn, discourages politicians from implementing the policy. Another reason for public skepticism is a lack of trust in politicians and municipal officials to manage the collected funds effectively, with concerns that the money may not be spent in ways that benefit the city.
Public Transport Cashback
Cashback Mechanism
To address the perceived unfairness of congestion pricing and fears about the misuse of collected funds, we propose a personalized public transport cashback mechanism – a novel approach that has not yet been implemented anywhere. Instead of collecting the tolls, we suggest immediately transferring the money back to drivers in the form of public transport vouchers or cashback. That is, when a driver pays road toll, almost the entire amount is credited directly to their personal public transport account/card as cashback, while a small portion of the toll is retained to cover maintenance costs of the road pricing system. The cashback can only be used to pay for public transport. Since the road toll is returned to drivers in the form of public transport cashback, there is no need for money redistribution by public authorities.
Our pricing mechanism retains the core feature of conventional dynamic road pricing: the road toll motivates drivers to adjust their travel times, helping to prevent traffic jams. The toll values are likely to be different though, as the toll now has additional value to drivers who might use the cashback for public transport. While this feature reduces the efficiency of the toll compared to conventional congestion pricing, the cashback mechanism also introduces a new beneficial property. By motivating some drivers to occasionally switch to public transport, it further reduces car use and helps ease congestion. The interplay between these two factors ultimately determines the required congestion toll values.
The cashback can be accumulated over several years and is non-transferable to prevent drivers from using their cars more often. The cashback mechanism would likely work for private cars only, though exceptions and specific features can be adjusted to local circumstances. Public transport companies are likely to benefit from additional revenue through increased ticket sales and unused, expired cashback. However, since public transport ticket prices do not always cover the full cost of providing the service, it is important to balance the additional costs of implementing the cashback mechanism with the expected revenue gains. This could potentially be done by reducing the cashback portion relative to the toll share retained for system maintenance.
However, congestion pricing with a cashback mechanism is not a standalone solution or a silver bullet. It works best when combined with improvements of the public transport network, as this encourages drivers to make regular use of their cashback.
Transport Survey Data
The key idea behind the cashback mechanism is that it gives drivers direct and transparent control of their money, which is expected to make road pricing policy more acceptable. Whether this holds true or not is an empirical matter. This was tested by considering the means of a representative survey conducted in Riga (Latvia) and Vienna (Austria) in the summer of 2024. The survey includes 1,000 residents in both capitals and their respective surrounding municipalities. It features questions about respondents’ socio-demographic characteristics, current travel options, commute patterns (including accompanying trips with children), and their political and social attitudes. It also includes two stated-choice experiments exploring the acceptability of congestion pricing and potential changes in travel behaviour if such pricing is introduced. While detailed data analysis is still ongoing, this policy brief highlights some intriguing preliminary insights.
In the survey, we ask the respondents whether they would vote in a referendum in favor of congestion pricing under four different scenarios for using the collected toll funds: (i) transferring them as a public transport cashback, (ii) sharing them equally among all city inhabitants, (iii) leaving the allocation decisions to local politicians, or (iv) using them to support eco-friendly transport. Respondents were familiarized with the topic before answering the question by participating in a stated-choice experiment about congestion pricing acceptability. The experiment included a detailed explanation of how congestion pricing works, along with a potential congestion zone map. Figure 1 shows responses from Riga, and Figure 2 from Vienna.
Figure 1. Responses from Riga. “Would you support congestion pricing in a referendum if the collected toll funds were used this way?”
Source: Representative survey in Riga in summer 2024.
Figure 2. Responses from Vienna. “Would you support congestion pricing in a referendum if the collected toll funds were used this way?”
Source: Representative survey in Vienna in summer 2024.
In Riga, the cashback option is the most popular, with more participants supporting than opposing it. The overall positive attitude towards congestion pricing with the cashback option suggests that Riga might already be ready to implement it. In Vienna, the cashback ranks a close second after the green transport option. This result shows that cashback might be a viable option also in Vienna.
Conclusion
To overcome public skepticism towards road congestion pricing, we propose a cashback mechanism. It involves returning toll money back to drivers as public transport cashback. The cashback mechanism has several benefits: drivers retain some control of their money, there is no need to redistribute collected toll funds, and it helps reduce congestion without major investments in road infrastructure. Surveys in Riga and Vienna in 2024 show support for the cashback option. While the specifics of such a solution should be tailored to each city’s needs, many cities struggling with congestion could benefit from implementing road congestion pricing with a public transport cashback mechanism.
Acknowledgment
This policy brief is based on a collaborative research effort by economists Sergejs Gubins from Riga (BICEPS) and Stefanie Peer and Martina Reggerova from Vienna (WU) as part of the “Tolls That Work” project, supported by the ERA-NET research grant. Agreement No ES RTD/2023/11. See project updates on the webpage:
https://www.wu.ac.at/en/spatialeconomics/projects/city-tolls-that-work
References
- Duranton, G., & Turner, M. A. (2011). The fundamental law of road congestion: Evidence from US cities. American Economic Review, 101(6), 2616–2652. https://doi.org/10.1257/aer.101.6.2616
- Hymel, K. (2019). If you build it, they will drive: Measuring induced demand for vehicle travel in urban areas. Transport Policy, 76, 57–66. https://doi.org/10.1016/j.tranpol.2018.12.006
- Ossokina, I. V., van Ommeren, J., & van Mourik, H. (2023). Do highway widenings reduce congestion? Journal of Economic Geography, 23(4), 871–900. https://doi.org/10.1093/jeg/lbad025
- Selmoune, A., Cheng, Q., Wang, L., & Liu, Z. (2020). Influencing factors in congestion pricing acceptability: A literature review. Journal of Advanced Transportation, 2020, 4242964, 11 pages. https://doi.org/10.1155/2020/4242964
- Small, K. A., & Verhoef, E. T. (2007). The economics of urban transportation. London: Routledge.
- The Swedish Transport Agency. https://www.transportstyrelsen.se/en/road/vehicles/taxes-and-fees/road-tolls/congestion-taxes-in-stockholm-and-gothenburg
Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.
How Social Assistance Shapes Election Outcomes: The Case of Georgia
This policy brief investigates the relationship between social assistance programs and election outcomes in Georgia, focusing on the 2024 parliamentary elections. Our regression analysis establishes a statistically significant link between an increase in social assistance beneficiaries and the vote share obtained by the incumbent Georgian Dream party. The results raise critical questions about the potential use of social assistance programs as a strategic political tool. Specifically, a 1 percentage point increase in targeted social assistance beneficiaries as a share of the population lead to an, on average, 0.5 percentage point increase in the Georgian Dream’s vote share, even after controlling for poverty-related factors. The findings recognize the dual impact of social assistance programs – alleviating poverty while shaping political behavior. They also underscore the need for ensuring that social assistance remains focused on addressing the needs of vulnerable populations without exerting undue political influence.
Introduction
The relationship between social assistance programs and electoral outcomes has gathered significant attention in both academic and policy circles, especially in the last decade. Social assistance programs, designed to support vulnerable populations, often carry political implications, particularly in developing democracies where incumbent governments may leverage these programs to secure voter loyalty. In Georgia, one of the largest components of social assistance is the targeted living allowance program, which, unlike other types of social assistance – such as those for individuals with special needs, internally displaced people, or elderly population –relies on assessing the beneficiaries’ poverty levels through proxy means testing (PMT). This makes subsistence allowance benefits vulnerable to biased, favorable selection by those in power. Allegations exist that the government may have strategically used this program, including increasing the number of beneficiaries in the lead-up to elections to secure votes or threatening existing beneficiaries with the withdrawal of their assistance based on their disclosed political preferences (Shubladze (2024); Japaridze (2023); Social Justice Center (2024)).
This policy brief explores the impact of social assistance on electoral outcomes in Georgia, specifically assessing whether increases in the number of targeted subsistence allowance beneficiaries during the 2020-2024 period influenced the votes received by the incumbent party in the 2024 parliamentary elections.
This analysis is especially important given the recent developments in Georgia’s political landscape. The 2024 parliamentary elections marked a critical juncture, with the Georgian Dream claiming to have secured 53.93 percent of the votes. Concerns over the fairness and transparency of the elections have been widespread. The Organization for Security and Co-operation in Europe’s (OSCE) Office for Democratic Institutions and Human Rights (ODIHR) reported systematic election irregularities, including pressure on voters, media bias, unequal campaign conditions, and election-day practices that compromised the ability of some voters – including public sector employees and recipients of social assistance – to cast their ballots without fear of retribution.
Our regression analysis documents a positive relationship between the number of living allowance beneficiaries and the votes garnered by the incumbent party across Georgian municipalities, raising further concerns about the integrity of the electoral process, and the allocation of state funds.
Political Implications of Social Assistance Programs: A Global Perspective
The link between social assistance and electoral outcomes has been widely studied. Social assistance programs often serve a dual purpose: they alleviate poverty and provide tangible support to vulnerable citizens while also shaping political behavior, particularly voting patterns. These programs can enhance incumbents’ electoral support by fostering gratitude among recipients, signaling government competence in addressing social needs, or creating concerns among beneficiaries that their political preferences if exposed, may influence the government’s decisions when choosing the beneficiaries of social assistance.
Research by De La O (2013) provides a compelling case in the context of Mexico. Examining the Progresa/Oportunidades conditional cash transfer program, De La O find that the program led to an increase in both voter turnout and incumbent vote share.
Zucco (2013) contributes further evidence from Brazil, where the Bolsa Família conditional cash transfer program emerged as a cornerstone of electoral strategy. Zucco demonstrates that municipalities with higher proportions of cash transfer beneficiaries tended to favor incumbent candidates in three different presidential elections, establishing a clear link between social assistance and voting behavior.
Adding to this body of work, Layton & Smith (2015) provide further insight into the nuanced ways in which targeted social assistance programs influence voter behavior in Latin America. The authors theorize that such programs simultaneously mobilize non-voters and convert opposition supporters, with variations based on country-level political and programmatic differences.
In addition, recent research from Indonesia further illustrates the impact of social assistance budgets on electoral outcomes. A study by Dharma, Syakhroza, and Martani (2022) examines 212 regencies and cities in Indonesia where incumbents participated in local elections. The findings reveal a direct positive effect of social assistance spending on incumbent votes. The authors further claim that high political competition counteracts incumbents’ advantages and mitigates the effectiveness of such spending.
The international literature thus provides valuable motivation for exploring the Georgian case, where social assistance may play a similar role in shaping voting behavior.
The Georgian Context
The deeply controversial October 26, 2024, parliamentary elections in Georgia mark a pivotal moment in the country’s political history. According to the Central Election Committee of Georgia, the ruling party, the Georgian Dream, secured 53.93 percent of the votes, maintaining its dominant position in Georgian politics. However, the elections were accompanied by widespread allegations of electoral malpractice, casting a shadow over their legitimacy and raising concerns about the future of democratic governance in the country.
OSCE’s ODIHR provided a comprehensive observation of the electoral process, noting both positive aspects and critical shortcomings. While the elections were generally well-administered, ODIHR’s final report emphasized significant concerns related to the broader political environment. Key issues included the adoption of legislation that undermined fundamental freedoms, restrictions on civil society, and a pervasive atmosphere of voter intimidation. Specific election-day practices, such as pressuring voters and leveraging administrative resources, were highlighted as undermining the integrity of the process. The report also mentions instances in 16 municipalities where public sector employees and economically vulnerable groups, particularly those reliant on social assistance, faced pressure to support the ruling party. Such fear of losing social benefits or facing retribution at work creates an atmosphere where voters struggle to form independent opinions and vote independently.
Further scrutiny from independent analysts has shed light on systematic irregularities that suggest the elections may not have reflected the genuine will of the Georgian electorate. Gutbrod (2024) suggests that tactics such as vote buying, mass intimidation, and direct manipulation of electoral outcomes were employed, leading to statistical anomalies. Specifically, the Georgian Dream’s support increased disproportionately in precincts linked to reported violence and irregularities. Additionally, social assistance beneficiaries were identified as a target group for snowball mobilization, organized by individuals affiliated with the Georgian Dream – a method where participants are encouraged to mobilize or identify a certain number of additional people to expand voter outreach and engagement.
Social Assistance in Georgia
The Law of Georgia on Social Assistance outlines several types of social welfare programs aimed at addressing the needs of vulnerable populations. These include living allowance, reintegration assistance, foster care allowance, adult family member care allowance, non-monetary social assistance, and social package. Among these, the targeted social assistance program, commonly referred to as the “living allowance“, holds particular significance. This program is designed to provide financial support to families living in extreme poverty. Eligibility for the living allowance is determined through a proxy means test that evaluates the socioeconomic conditions of applicants, ensuring that the assistance reaches those most in need. For this policy brief, the focus will be on beneficiaries of the living allowance (hereafter social assistance beneficiaries), as their numbers and electoral behavior present a unique opportunity to analyze the intersection of social assistance and voting patterns in Georgia.
As of October 2020 (previous parliamentary elections’ date) 142,870 families in Georgia received social assistance, benefiting a total of 510,343 individuals. The total amount of social assistance transfers during this period amounted to 28,825,259 GEL. Over the next four years, leading up to the 2024 Parliamentary Elections, social assistance grew significantly. By October 2024, the number of families receiving assistance had increased by 25 percent to 178,107 families, while the number of individual beneficiaries rose by 34 percent, reaching 684,432. The most notable expansion occurred in the total amount of social assistance transfers, which surged by 143 percent to 69,936,512 GEL. This corresponds to a cumulative annual growth rate of 25 percent (Social Service Agency of Georgia, 2024).
In 2022, an important modification was introduced for social assistance beneficiaries aged 18 years to retirement age and without disabilities or serious health conditions, offering employment opportunities mainly, in the public sector with a salary of up to 300 GEL per month. These wages did not affect recipients’ existing social assistance benefits. Participants had the option to take suitable public sector jobs, formalize any informal employment, or, if formally employed in the private sector, provide necessary documentation. The modification covered also new beneficiaries who were not already formally employed. Notably, families or individuals enrolled in the program were guaranteed eligibility for a living allowance for four years, as their social assistance status would not be reassessed during this period.
As of October 2024, 50,962 families were enrolled in the program with guaranteed social assistance, accounting for 28.6 percent of all families receiving social assistance. The monthly spending of social assistance transferred to these families amounted to 22,766,706 GEL, representing 33 percent of the total social assistance transfers.
The significant increase in social assistance beneficiaries and the introduction of the 2022 program for employing social assistance recipients, guaranteeing them four years of social assistance transfers, highlight the growing scope and influence of targeted social welfare initiatives in Georgia. While these developments may have addressed pressing socioeconomic challenges, they also raise important questions about the potential political motivations. Specifically, the substantial increase in the number of beneficiaries and the guaranteed eligibility linked to employment programs could be interpreted as mechanisms to foster voter loyalty and mobilization in favor of the ruling party.
Methodology and Results
To examine the relationship between the increase in social assistance beneficiaries and electoral outcomes, particularly the votes garnered by the incumbent Georgian Dream party, we employ a regression analysis framework. This statistical method allows us to explore whether and to what extent the growth in social assistance recipients is associated with the changes in the vote share of the incumbent party. Since social assistance depends on the varying levels of poverty across municipalities, we incorporate control variables that isolate the effect of economic well-being, minimizing potential confounders.
The study utilizes data from two primary sources: information on social assistance recipients, including families, and individuals, and the total amount of transfers across municipalities, was retrieved from the Social Service Agency of Georgia. This dataset covers 64 municipalities (and self-governing cities) in Georgia. From 2022, data includes families and individuals guaranteed to retain their socially vulnerable status for four years under the State Program for Promoting Public Employment. Second, election data was sourced from the Central Election Commission of Georgia, covering both the 2024 and 2020 parliamentary elections. The 2024 data covers the results from both electronic and non-electronic voting. Key variables include the number of registered voters, total votes cast, and votes obtained by the Georgian Dream and opposition parties. This election data is also aggregated at the level of the 64 municipalities (and self-governing cities).
Information on poverty levels in Georgian municipalities is not publicly available; therefore, we utilize control variables for employment and economic activity with the latter proxied by either the municipalities’ tax revenues or the value added generated in the private sector. Information on employment and value added are gathered from the National Statistics Office of Georgia, while data on tax revenues is retrieved from the Ministry of Finance.
The following table describes the results of the regression analysis.
Table 1. Regression analysis results
Source: Author’s calculations.
Note: The values in parentheses indicate the p-value. *Significant at the 10 percent level; **Significant at the 5 percent level; ***Significant at the 1 percent level.
The first regression (column 1) investigates the relationship between the change in social assistance beneficiaries as a share of the population and the change in the Georgian Dream party’s vote shares, displaying a significant relationship between the two. Specifically, the coefficient (0.49) is significant at the 5 percent level, suggesting that a 1 percentage point increase in social assistance beneficiaries as a share of the population, increases the vote share for the Georgian Dream by approximately 0.49 percentage points.
To control for the effect of poverty, we first use employment rates in 2023 (the latest available data) as a proxy for poverty. Column 2 presents these results. In this specification, the coefficient for the change in social assistance beneficiaries remains significant at 5 percent level and its value (0.47) remains consistent with the previous specification. The model further suggests that poverty is also positively and significantly (at the 1 percent level) associated with incumbent votes – the higher the poverty (lower employment) in municipalities, the higher the Georgian Dream vote share.
In the next step (column 3), we model the relationship between the change in the Georgian Dream’s vote share, change in employment as a share of the population, mobilized local tax revenues per capita, and the change in number of social assistance beneficiaries as a share of the population. Change in employment, calculated as the difference between 2019 and 2023 employment levels (as a share of the population), is used as a proxy for change in poverty. Tax revenues per capita for 2023 reflect economic activity across municipalities and self-governing cities, serving as a proxy for well-being. As seen in the table, change in employment is not statistically significant, however, the amount of tax revenues mobilized across municipalities is modestly significant. The coefficient for change in social assistance beneficiaries is once again statistically significant and consistent with the other specifications in terms of magnitude (at 0.51).
As a robustness test (column 4) we replace the previously used proxy for economic well-being (tax revenues), with the private sector value added per capita for 2023, which significantly (at the 1 percent level) correlates with an increase in the vote share for the incumbent party. Changes in employment remain insignificant. Importantly, the coefficient for change in social assistance beneficiaries remains positive (0.53) and statistically significant at the 1 percent level.
The discussed regression models were tested for a different dependent variable as well. In addition to observing the impact of change in vote shares, we also analyzed the impact on the number of votes cast for the Georgian Dream party between the 2020 and 2024 Parliamentary Elections. Changes in social assistance beneficiaries remain a significant explanatory variable in this specification as well.
The estimated impact of social assistance is consistent across all models, both in magnitude and significance, reinforcing the finding that increases in living allowance beneficiaries are strongly associated with higher vote shares for the Georgian Dream party, underscoring the critical role of social assistance in shaping electoral outcomes.
Conclusion
The analysis demonstrates a strong and statistically significant relationship between the increase in social assistance beneficiaries and the vote share obtained by the incumbent Georgian Dream party in the 2024 parliamentary elections. Even after controlling for poverty and economic well-being, the results highlight the impact of social assistance in shaping electoral outcomes. The findings suggest that a 1 percentage point rise in social assistance beneficiaries as a share of the population translates into a 0.47–0.53 percentage point increase in the Georgian Dream’s vote share. When contextualized within the overall election results, these estimates suggest that the expansion of the targeted social assistance program may have garnered the Georgian Dream an additional 45 ,500 to 50,000 votes, representing 2.2–2.5 percent of the total votes.
The results raise critical questions about the potential use of social assistance programs as a strategic political tool. The robustness of the relationship across multiple models suggests that the observed trends are not merely byproducts of economic conditions but reflect a deliberate link between social assistance expansion and electoral outcomes. The implications are significant for democratic governance in Georgia. The strategic use of social welfare programs risks undermining public trust in the electoral process and highlights the need for greater transparency and accountability in the implementation of social assistance policies. Recognizing the dual impact of these programs – alleviating poverty while potentially shaping political behavior – will be critical in fostering fairer electoral conditions and ensuring that social assistance remains focused on addressing the needs of vulnerable populations without undue political influence.
References
- De La O, A. L. (2013). “Do conditional cash transfers affect electoral behavior? Evidence from a randomized experiment in Mexico”, American Journal of Political Science, 57(1), 1-14.
- Dharma, F., Syakhroza, A., & Martani, D. (2022). “Does the social assistance budget realization affect incumbents’ votes? (Study in Indonesia Local Election)”, International Journal of Professional Business Review, 7(6), e0636-e0636.
- Gutbrod, H. (2024). “A dozen daggers: How Georgia’s 2024 elections were rigged”.
- Japaridze, T. (2023). “Social Policy in contemporary Georgia: liberal narratives, intervention and welfare state”, King’s College London.
- Law of Georgia on Social Assistance (2024). Parliament of Georgia.
- Layton, M. L., & Smith, A. E. (2015). “Incorporating marginal citizens and voters: the conditional electoral effects of targeted social assistance in Latin America”, Comparative Political Studies, 48(7), 854-881
- OSCE Office for Democratic Institutions and Human Rights (ODIHR), (2024). “Georgia Parliamentary Elections: Final Report”, Organization for Security and Co-operation in Europe.
- Shubladze, R. (2024). “Targeted Social Assistance Program in Georgia and Its Link to Electoral Outcomes”, Social Justice Center.
- Social Justice Center (2024). “What are the challenges in the livelihood support system, and what changes are necessary for its improvement?”.
- Social Service Agency of Georgia. (2024). https://ssa.moh.gov.ge/index.php?lang=1&v=0
- Zucco Jr, C. (2013). “When payouts pay off: Conditional cash transfers and voting behavior in Brazil 2002–10”, American journal of political science, 57(4), 810-822.
Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.