We aim to discover what factors affect global arms transfers, with particular scrutiny of whether states are more or less likely to export materiel to countries actively engaged in conflict.

Many countries claim to have ethical foreign policies, but does war act as a restraint or as encouragement when they choose where to permit arms exports?

This website accompanies the report, “Business as Usual”, in which full details of our methodology, analysis and conclusions can be found.

The research comprised two branches, examining the question from opposite perspectives:

For all wars between 2000 and 2018, we examined which states supplied arms to the conflict participants as the violence evolved over time.

List of Conflicts

We compiled data on all conflicts considered wars so far in the 21st century, and the countries that participated in them. For each participant, we counted the value of deliveries of Major Conventional Weapons (MCW) from the top 11 arms-supplying states in years when the recipient was actively involved while the conflict was at the intensity of war. The results are illustrated both in total and year-by-year.

For the top arms-selling countries, we analyzed where they did and did not sell arms, and whether this correlated with the presence or absence of war in the destination country, or with a range of other economic, political and conflict metrics

List of Arms Suppliers

We focused on the top 11 suppliers of Major Conventional Weapons (MCW) for the period of 1990-2018. We examined each possible combination of supplier country, recipient country and year, and in each case noted whether or not there were transfers of MCW from supplier to recipient in that year. We likewise collected data on the value of each metric for each recipient-year pair, or for some variables, each combination of supplier, recipient, and year. We looked at how the likelihood of transfers of MCW varied with different values of these other metrics.

We used the Stockholm International Peace Research Institute (SIPRI) Arms Transfer Database, which details transfers of Major Conventional Weapons between states.

For conflict analysis, we counted, for each conflict, the total volume of [Major Conventional Weapons] that were delivered by each suppier to any country in a year when it was actively engaged in that conflictm and when the conflict was classified as war at that time.

For arms supplier analysis, we treated any transfer record in the SIPRI database as constituting a significant transfer of weaponry from supplier to recipient for the year that the materiel was delivered. Each combination of supplier, recipient, and year thus has a "yes/no" (or 1/0) variable for arms supply.

Data was gathered from the Uppsala Conflict Data Project (UCDP) database of conflicts. The UCDP lists levels of conflict for each conflict in each calendar year, and the states and non-state groups actively engaged in the conflict in that year. For more information see the UCDP methodology page.

For conflict analysis, we selected all confrontations that had reached the intensity of war at any point in their timeline during the period 2000-2018. We also included episodes of one-sided violence against civilians perpetrated by state forces that reached a similar level of lethality.

For arms supplier analysis, we considered for each country and year, the intensity level of each conflict - whether it was war, minor armed conflict, or neither - that the country was involved in that year. The country was then rated as being at "war" if it was involved in at least one conflict with the intensity of war that year; as being in "minor armed conflict" if it was NOT at war, but was involved in at least one conflict with the intensity of minor armed conflict that year; or neither if it was not involved in any conflict reaching at least the intensity of minor armed conflict that year.

In addition, all other metrics were evaluated for each country and year.

We limited analysis to the top eleven arms exporting states, by value of transfers in the SIPRI Arms Transfer Database in the ten-year period 2009-2018. While a top ten might be more usual, we decided to include the eleventh country, Ukraine, as it offers an additional case of an emerging, non-Western supplier.

All examined metrics concerned the recipient state, or related to both the supplier and recipient state (for example, whether each supplier had recently supplied arms to the recipient). See the full list below. Note that some metrics were ignored where they were not considered relevant for a particular supplier state. This is detailed in footnotes on the specific pages involved.

The coefficients and significance of each metric were calculated using panel data logistic regression analysis. Full results can be found at the bottom of each supplier page.

For certain countries (Iraq, Libya, and Democratic Republic of Congo) a period of uninterrupted warfare was split into separate conflicts for the purpose of better analysis. Though this division was based on many nuanced elements, the essential common factor was a change of government in the country where the conflict took place. The choice to analyze these phases separately reflects a belief that the motives of arms suppliers and participants may be significantly different before and after.

Major Conventional Weapons covered by SIPRI's data include large weapons such as missiles and artillery; armed and armored military vehicles; and other large military items like engines, sensors and satellites. Notably it does not include small arms and other man-portable weapons; components of military equipment, or most sub-systems; nor military services or technology. For full details see SIPRI's full coverage description.

The UCDP records two levels of intensity of conflict for any conflict in any given calendar year. These levels are “Minor Armed Conflict” (25-999 battle-related deaths) and “War” (1000 or more battle-related deaths). For more information see the UCDP intensity definition.

The Trend Indicator Value (TIV) is a constructed measure of the notional value of military equipment, defined by the Stockholm International Peace Research Institute. It aims to provide a way of assigning comparable values to any arms transfer regardless of the countries involved, the price paid (which is not always known), or the date of the transfer. Weapons systems from different countries that are judged to have similar capabilities are thus given similar TIV values. As such, the value is normalized for inflation and currency, and reduced based on previous ownership and use. More information is available on SIPRI's methodology page.

Aside from Major Conventional Weapons, many other kinds of weaponry can have a significant effect on a country's military capacity, such as key components for weapons system. For analysis in the associated report “Business as Usual”, countries were considered to be “other significant suppliers” to a conflict under any of the following circumstances:

• €100m or more of export licenses (for the UK and Germany) or actual exports (for other EU countries) recorded by the European Union to participants at war in total, or €1m or more in a single year.
• $100m or more of arms deliveries recorded by the the Security Assistance Monitor (United States only) to participants at war, or $1m in a single year
• Having accepted orders of Major Conventional Weapons from any participant engaged in war at time of ordering, even though the deliveries were made in a year when the recipient was no longer at war.

Note that equivalents of the EU and US data described above were not available for Russia, China, Israel or Ukraine, so similar criteria could not be considered for these countries.