As the North Atlantic Treaty Organization (NATO) turns 75 this year, it is hardly surprising that burden sharing—the degree to which allies are financially shouldering the costs of collective security—is once again an issue at the forefront of Washington, D.C. policy discussions. For decades, the issue has hinged on the amount that allies are spending on defense; since 2014, the focus has been on whether allies are spending 2 percent of their gross domestic product (GDP) on defense and military programs. Now, however, the tune has changed slightly. Congressional frustration is mounting regarding spending on Ukraine and whether U.S. taxpayers should continue to foot their part of the bill—especially when NATO allies have, in the view of many policymakers, failed to meet their own defense-spending commitments.

Given the stakes in Ukraine for the future of the alliance and global democracy, as well as the enormous utility that NATO has for advancing U.S. interests, Congress is getting perilously close to cutting off its nose to spite its face. Especially since the major metric against which burden sharing is measured—as a percentage of GDP spent on military and defense matters—does not adequately capture all the different ways that allies are spending their resources on capabilities and programs that improve transatlantic security writ large. This discrepancy exists because the 2 percent metric only captures what states are spending on their military capabilities and defense programs. But deterring and defending against a revanchist Russian adversary requires whole-of-government—if not whole-of-society—capabilities and plans that are not presently factored into the way NATO thinks about burden sharing. In other words, an expanded definition of what countries can count as contributions toward transatlantic security is needed since the international security environment is one in which the lines between domestic and foreign policy are being blurred, and defense spending is necessary but not sufficient to meet national security requirements. While the authors do not argue that allied nations should spend less than 2 percent of GDP on their own defense, there should be more flexibility associated with how they spend it and how it is “counted.”

From Burden Sharing to Responsibility Sharing

Many take the view that NATO is a military alliance; accounting for broader civilian and whole-of-government activities is, therefore, outside its purview. Yet this perception of the alliance and its responsibilities arguably misreads the 1949 North Atlantic Treaty (Washington Treaty) that founded it in the first place:

“It is often overlooked, but NATO’s collective strength isn’t just in its ability to defend its members from an attack (Article V of the 1949 Washington Treaty) but also in its requirements to ensure members are undertaking necessary domestic activities to prepare for crisis response and, potentially, military action (see Article III of the Washington Treaty). This allows for resilience, something which, in a dynamic world, needs to be more deeply invested in and more comprehensively approached.”

Article II of the Washington Treaty also establishes that the alliance has a role to play in protecting democracy and promoting economic cooperation. NATO’s adversaries are waging hybrid and political warfare against allies, which ultimately presents a direct Article II challenge. NATO allies should and are investing in whole-of-government responses to such threats; given their criticality of the survival to the alliance overall, such investments ought to be included in calculations of NATO responsibility sharing.

Two of the authors argued for a new “responsibility sharing” framework for NATO spending based upon an expanded view of what counts toward collective security spending:

“Allies should spend 4 percent of their GDP on defense and security annually: States should spend a minimum of 2 percent of GDP on defense, though NATO should continue to explore and allow more flexibility in the way those monies are spent nationally, particularly for states that do not have sufficient absorptive capacity to spend 2 percent on their defense capabilities and programs. The balance—between 2 and 4 percent—should be allocated toward activities that are strategically vital to the alliance but are not accounted for in NATO’s methodologies for defense spending, such as peacetime preparedness and resilience.”

Putting such a concept into practice, an analysis of responsibility sharing should include not only direct defense expenditures but also investments in resilience and select adjacent sectors. The question then becomes: what does responsibility sharing look like after applying this broader definition of collective security spending?

Redefining What “Counts”

To build a more comprehensive picture of responsibility sharing across the NATO alliance, the authors pulled publicly available data on:[1]

  • Defense spending—that is, those monies that were counted toward NATO defense pledges in 2022, including military costs and research and development (R&D).
  • Public order and safety expenditures in 2021, including on police, firefighting, judicial systems, and R&D related to public order and safety. These capabilities have traditionally been excluded from burden sharing calculations, as they are thought to “dilute” the focus on needed military spending. Yet they are essential for national resilience in the event of a crisis—and thus critical for the deterrence and defense of the transatlantic community. They, therefore, warrant inclusion in a new responsibility sharing framework.
  • Support for Ukraine in 2022, including military, financial, and humanitarian aid.
  • The costs of pivoting from Russian energy sources such as liquefied natural gas (LNG), oil, and coal—based on bilateral trade data from 2022. Following the invasion of Ukraine, the European Union and NATO members called for and adopted sanctions that significantly reduced trade with Russia. Subsequent sanctions and reduction of trade ensued as allies reoriented away from Russian energy sources. EU and NATO members had to find alternatives from countries such as Norway and have invested in LNG regasification terminals to enable imports from the United States and elsewhere. The European Union also imposed a ban on coal and embargoes on Russian seaborne crude oil and petroleum products, and the Group of Seven (G7) set an oil price cap.

There are likely other contributions to responsibility sharing that are not captured in this analysis, such as investments in cyber protection of critical national infrastructure, operational risk sharing, rail infrastructure and airport maintenance. These and other data will be incorporated into future CSIS analyses as appropriate and when made available.

Methods and Data

As a first step to quantify contributions and help policy practitioners move toward responsibility sharing, researchers within the Futures Lab at CSIS collected data to provide a more holistic measurement. The first dataset used was the Consumer Price Index (CPI), which is a measure of the average change over time in the prices paid for a market basket of consumer goods and services produced by the U.S. Bureau of Labor Statistics (BLS). The researchers queried 2021–2023 CPI monthly data, which was then aggregated for yearly averages.

Next, the team collected data on defense spending for 2022 from the Stockholm International Peace Research Institute (SIPRI) military expenditure database hosted by the World Bank; the CPI figures were then used to convert the current U.S. dollar (USD) listed in the dataset to 2022 USD.

To visualize support to Ukraine, the researchers leveraged the Kiel Institute for the World Economy’s Ukraine Support Tracker. Data pulled included military, financial, and humanitarian aid from 2022, and the 2021 USD figures were converted to 2022 USD.

To visualize nonmilitary or nondefense expenditures with national security implications, CSIS compiled data from various sources that use the United Nations’ classification of the functions of government (COFOG) standard. This included 2021 general government-spending data from Eurostat, the Organization for Economic Cooperation and Development (OECD), the UK Office of National StatisticsStatistics CanadaWorld Bank BOOST platform, and the International Monetary Fund on “public order and safety.” Multiple sources were necessary because no single repository held all NATO members’ expenditure for this COFOG category, which captures “police services; fire-protection services; law courts; prisons; [and] R&D related to public order and safety.” As 2021 was the final year of reporting by most NATO members in Eurostat, all queries from these other sources also looked at just 2021 for consistency. The next step was to convert reported local currencies (the euro, British pound, Canadian dollar, Albanian lek, Icelandic króna, and Swedish krona) to 2022 USD using exchange rates provided by the U.S. Department of the Treasury and GDP deflators provided by the U.S. Department of Defense to adjust for inflation.

The collection was then complemented by NATO allies’ energy divestment. To quantify Russian energy divestment by commodity, the team first collected NATO members’ reported imports from Russia between 2021 and 2022 using Harmonized System (HS) codes for traded products. This data, accessed through the UN Commodity Trade Statistics Database (UN Comtrade), reflects the net weight of commodities and trade value in current USD. The HS codes used captured the flow of petroleum and bituminous oils (excluding crude), bituminous coal (not agglomerated), and LNG. The team then converted the current USD reported to 2022 USD, calculated the average price in 2021 by dividing the total value by the net weight imported in 2021, and used the price per unit of weight against the total value in 2022 to control for post-war price shocks. Next, the research team calculated the difference in purchased value between both years to obtain a pre- and post-war comparison illustrating the extent of divestment from Russian energy, aggregating the figures for oil and coal while leaving LNG as a separate category. Finally, the team queried 2022 GDP data from the World Bank, converting current USD to 2022 USD to calculate the expense categories as a share of each country’s GDP.

Key Findings

Data pulled on allied spending from these sources presented a reasonably optimistic picture of responsibility sharing across the NATO alliance:

  • At least 13 allies are spending over 4 percent of GDP on “responsibility sharing,” (14 including Sweden).
  • At least further 11 countries are spending over 3 percent of GDP on responsibility sharing.
  • Some countries worked to shift away from Russian energy sources. Lithuania, for example, spent roughly 3.21 percent of its GDP as it diversified its energy sector away from Russian imports.

Of note, 18 NATO allies are reported to be on track to meet the 2 percent defense spending target in 2024, meaning that the defense portion of responsibility sharing is increasing—although not reflected in this analysis.

As shown in Figure 1, at least 13 allies (14 including Sweden) are spending over 4 percent of GDP when counting not only defense spending but also public order and safety expenditures, Ukraine aid, and Russian energy divestment. At least another 11 are spending over 3 percent of GDP on this responsibility sharing. (CSIS was unable to capture public order and safety spending for North Macedonia and Montenegro.) Some countries worked particularly hard to shift away from Russian energy sources, as shown in Figure 2. Lithuania, for example, spent roughly 3.21 percent of its GDP on diversifying its energy sector away from Russian imports.

Spotlight: Pivoting from Russian Energy Sources

After Russia’s invasion of Ukraine in 2022, many European countries chose to move away from Russian energy imports as part of several sanctions packages introduced by the European Union. This choice, while expensive, provided two major strategic advantages. First, such divestment was viewed as a way to impose costs on Moscow for its illegal invasion of Ukraine. Second, Europeans hoped that diversifying national energy sectors and reducing reliance upon Russian sources would remove any strategic leverage Moscow might seek to wield against its neighbors.

European countries were already looking to find substitutes for Russian pipeline gas following the curtailment of Russian export volumes beginning in 2021 and the closure of several major pipelines. Since the start of the war in Ukraine, these countries have increasingly relied on alternative gas supplies and greater energy efficiency to meet local demand, with LNG from the United States and additional pipeline gas imports from Norway playing a critical role. EU countries have constructed six new regasification terminals since early 2022 to accommodate rising LNG imports, including three in Germany, and more are under construction.

The European Union also imposed an embargo on Russian seaborne crude oil and petroleum products (exempting some supplies such as those imported via the Druzhba pipeline’s southern arm). These embargoes forced Russia to divert its crude and petroleum products to other markets, giving buyers greater leverage and reducing Russia’s oil revenue. This was shortly followed by an EU ban on coal and, a year later, a G7 oil price cap. As seen in Table 1, Lithuania exemplifies the shift, having slashed Russian oil imports by approximately 80 percent between 2021 and 2022, equating to a substantial 3.05 percent of its 2022 GDP. Finland follows with a 76 percent reduction in oil imports, corresponding to around 0.91 percent of its GDP for the same year. Greece also made significant cuts, with a 76 percent decrease in oil imports, equivalent to around 0.31 percent of its GDP.

That said, CSIS identified some instances in which European countries increased their energy imports from Russia during this timeframe, suggesting that a more coherent strategic approach to responsibility sharing on collective security is needed.


Continue increasing defense budgets. Understanding the broader national security spending of NATO allies provides a more comprehensive picture of responsibility sharing. But widening the aperture does not mean that states can lessen their spending on defense—quite the opposite. As agreed at the Vilnius NATO summit in 2023, states should spend a minimum of 2 percent of GDP on defense. That said, “NATO should continue to explore and allow more flexibility in the way those monies are spent nationally, particularly for states that do not have sufficient absorptive capacity to spend 2 percent on their defense capabilities and programs.”

Build a responsibility sharing data-capture process. This CSIS paper uses publicly available data to show what responsibility sharing looks like. It constitutes a rough approximation of what states are doing to share the security burden. Obviously, the responsibility sharing picture would become clearer if and when more granular data is made available. Considerably, more work can be done to flesh out responsibility sharing in such areas as critical infrastructure protection and energy trade flows.

NATO should lead the way in building this data picture by establishing a new responsibility sharing reporting process—perhaps during discussions at the upcoming 75th anniversary of the founding of the alliance. Yet there are also risks associated with building such a process. Namely, if done improperly, NATO and its member states might include miscellaneous spending that does not add value or capability to the alliance.

In order to mitigate this risk, the secretary general and international staff should centrally manage the data collection process and provide a centralized data repository open to the public. To that end, the secretary general should be tasked at the initiation of the tenure with producing a report that articulates nonmilitary risks that have strategic importance to, or impact upon, NATO’s defense capabilities. Annual updates should also be issued by the secretary general’s office. From there, NATO’s international staff should develop a list of the kinds of activities or spending that would count against the target. Much like the defense investment pledge, nations would then report what they are doing to meet the goal.

In designing this process, NATO leaders should:[2]

  • Factor in temporality. NATO must account both for actions taken in the short term to mitigate against immediate threats and for plans and concrete steps taken to prepare for, deter, and mitigate longer-term risks.
  • Assume strategic shocks. Many of the crises that have demanded allied attention rapidly move from more peripheral concerns to the top of the allied agenda. A responsibility sharing framework must, therefore, be routinely updated and have mechanisms for when an issue moves to the top of the strategic agenda.
  • Identify nonmilitary investments that have bearing on alliance strategic positioning. As noted earlier in regard to NATO’s Article III preparedness requirements, such investments could include those in dual-use logistics and infrastructure to enhance military mobility, societal resistance capabilities (in areas such as hybrid warfare and information operations), and energy independence, to name a few. Some NATO nations pride themselves on having resilient domestic infrastructure to manage shocks, disruptions, and contingencies. These efforts should be applauded but also be shared within NATO as part of the greater capability set needed to enable alliance responsiveness.
  • When appropriate, allow states to report spending through other organizations, including the European Union. A number of strategically important investments, such as improvements to infrastructure that enhance military mobility, occur through the European Union. But, at the end of the day, member nations and national budgets still form the building blocks for that spending. Such investments, when aligned with NATO strategic priorities, as outlined by the secretary general, should, therefore, be reportable by nations as they show how they are [bolstering responsibility sharing].
  • Make reports publicly available on security spending. Part of the reason the burden sharing debate has become so toxic is many of the measures of what allies are doing and how they are doing it remain shielded from public view. A more holistic assessment of how allies are risk sharing should therefore inform the public discourse on the many ways allies are contributing to the common defense and security of the transatlantic community.

Kathleen J. McInnis is a senior fellow with the International Security Program and director of the Smar Women, Smart Power Initiative at the Center for Strategic and International Studies (CSIS) in Washington, D.C. Daniel P. Fata served as the former U.S. deputy assistant secretary of defense for Europe and NATO policy from 2005 to 2008 and is a non-resident senior adviser at CSIS. Benjamin Jensen is a senior fellow for future war, gaming, and strategy in the International Security Program at CSIS. Jose M. Macias is a research associate in the Futures Lab within the International Security Program at the Center for Strategic and International Studies (CSIS).

This report is made possible by general support to CSIS. No direct sponsorship contributed to this report.

Please consult the PDF for reference.