Partial reorientation as a working model - and the one condition it still lacks.

Armenia is still in the EAEU. Its gas still comes from Russia. In 2024 trade between Moscow and Yerevan hit a record $12.4 billion, according to data from Armstat. And in that same 2024 Armenia froze its CSTO membership, sent the Russian border guards home, signed a charter with the United States, and launched an EU accession process. This is not a pivot in the classic sense; it is an attempt to live inside two geopolitical circuits at once, and so far the only such model that actually works for a post-Soviet state.

The scale of the drift is clearest in the timeline of the past two years. By the end of 2024 the Russian border guards stationed at Zvartnots for 32 years were gone; since January 2025 Armenian guards have manned the stretch of border with Iran themselves. In January 2025 Yerevan signed a Strategic Partnership Charter with Washington; by May 2026 it was already a Charter on a *Comprehensive* Partnership. In March 2025 parliament passed a law launching the EU accession process (signed by the president in April); in November 2025 Brussels handed Yerevan a visa-liberalization roadmap. According to an IRI poll (May 2026), 75% of Armenians support EU membership.

To measure the distance, you need a second reference point. In March 2014 Armenia voted in the UN General Assembly against the resolution on Ukraine's territorial integrity: eleven states voted "no," and Yerevan found itself yoked to Russia, Belarus, Cuba, North Korea, and Zimbabwe. Even Kazakhstan, Moscow's closest EAEU ally, abstained. Twelve years ago Armenia was more pro-Russian than Astana.

And running alongside today's pivot is a baseline reality that has not gone anywhere. Almost all the gas Armenian consumers use comes from Russia: Gazprom is the sole commercial supplier, and the gas-transport network belongs to its subsidiary. The Metsamor nuclear plant runs on Russian fuel, and its operation, reportedly with an eye to extending its life to 2036, is handled by the same Rosatom. The record trade turnover of 2024 is roughly a third of Armenia's entire foreign trade. And Yerevan formally remains in the EAEU, stressing that it has no plans to leave.

So this is not a story of "Armenia walking out of Russia's orbit." It is a story of partial, asymmetric reorientation, in which the political-symbolic layer moves markedly faster than the economy and the infrastructure. The comparative question is more interesting than the normative one: why can some post-Soviet states pull off this maneuver while others no longer can?

The most instructive contrast is Georgia. In late 2023 Tbilisi received EU candidate status; a year later the picture had flipped: a foreign-agents law modeled on Russia's, elections with obvious violations, accession talks effectively halted by Brussels. In 2026 Armenia and Georgia are driving in opposite directions, though ten years ago their paths looked like mirror images. How did they swap places?

What Makes the Pivot Possible at All

The Catalyzing Shock: When the "Security Guarantor" Stops Guaranteeing

Without September 2023, none of the above would have happened - at least not at this speed. Azerbaijan's 24-hour operation in Karabakh, the exodus of more than 100,000 Armenians, the liquidation of Artsakh - this was a public demonstration that the Russian contingent, deployed precisely to keep the sides apart, did not intervene; that the CSTO does not, in principle, provide for help in such a situation; and that Moscow, busy with Ukraine, was unable or unwilling to lean on Baku.

The force of the shock was that it was not an argument but empirical confirmation of something long known. Between 2011 and 2020, by SIPRI's count, Russia supplied Armenia with about 94% of its heavy weaponry, while simultaneously providing about 60% of Azerbaijan's arms imports - 8.2 times larger than Armenia's by volume; in the "Four-Day War" of 2016, Baku went in on hardware supplied by Russia - Armenia's formal CSTO ally. For a decade and a half Russia armed the Armenian state's adversary better than it armed the Armenian state, all while remaining its nominal guarantor. Post-Soviet identity rested on three pillars: independence, Karabakh, victory; 2020 took the third, 2023 the second. After a catalyst like that, freezing the CSTO, withholding dues, and pulling out the border guards look less like radicalism than like logical consequences.

Georgia lived through an equivalent shock in August 2008 and responded by turning West. Catastrophe alone predetermines nothing; it opens a window through which elites may or may not steer the country. 2023 "worked" for Armenia because it coincided with an elite coalition already in place and ready to capitalize on it.

An Elite Coalition With No Old Obligations

The "velvet revolution" of spring 2018 was not anti-Russian: the protests against Serzh Sargsyan and the "Karabakh clan" that had dominated politics since the mid-nineties were aimed at a corrupt establishment, not at Moscow, and in his first months Pashinyan repeated to the Kremlin that the course would not change. Civil Contract spent five years in office before the pivot - the pivot was produced by the shock of 2023, not by the government's arrival. That lag matters: reorientation and a change of domestic elites do not necessarily go hand in hand.

But it was the coalition's sociology that made the pivot possible once the shock came. Civil Contract grew out of none of the milieus through which Moscow traditionally worked with the Armenian elite (the Soviet nomenklatura, the security services, the Karabakh military networks) - it is a party of Yerevan's urban middle class. Two effects follow. First, the absence of personal and material obligations to Moscow. Compare Georgian Dream, built around Bidzina Ivanishvili, whose fortune was made in Russia: for an actor like that, sanctions and asset freezes are an existential risk, and loyalty is baked into the balance sheet. Armenia's ruling party carries no such ballast. Second, the absence of an ideological stake in the Karabakh narrative: Pashinyan was already calling to "forget the heroic past" back in 2005, and it was precisely this that let him sign, in Prague in October 2022, a recognition of Karabakh as part of Azerbaijan - a step no "Karabakh" government could have taken without splitting apart.

The paradox: an elite politically vulnerable on "national" ground (the loss of Karabakh, the handover of villages, the clash with the Church) turns out to be the one most capable of sharp foreign-policy maneuvers - the cost of exiting the orbit is lower for it than for any predecessor. And its base supports it. The ground had been prepared for years: by CRRC's data, the share of Armenians who named Russia the country's main ally fell from 83% in 2013 to 14% in 2024. The electoral test came as early as June 2021: half a year after the lost war, Civil Contract took 53.9% against 21% for Kocharyan's revanchist bloc. The geography is the most telling part: in Shirak province, home to the soldiers taken prisoner near Hadrut, the party won 65%; in border-region Syunik, 53.4%. The regions that paid the most for the war voted for peace.

An External Anchor - Someone to Catch You

Reorientation requires a far shore. In 2018 Armenia had no "escalator": after Sargsyan walked away from the Association Agreement in 2013, Yerevan signed the more modest CEPA in 2017 - no security component, no free-trade zone, no membership prospect. By 2024 the situation had changed radically. The list of what Brussels and Washington put on the table in two years is striking even as a bare inventory: the EUMA monitoring mission on the border since 2023; a €270 million Resilience and Growth Plan; the European Peace Facility deployed, for the first time ever, for a CSTO member; a European Parliament resolution (March 2024) inviting Armenia to apply for candidate status; a visa-liberalization roadmap (November 2025); Strategic Partnership Charters with the US (January 2025) and the TRIPP mechanism (2026); the EUPM Armenia mission (April 2026); a high-level EU–Armenia meeting (July 2025); the first summit (May 2026).

These are not declarations but institutional infrastructure with funding, experts, and deadlines: each procedure is reversible on its own, but together they lock Yerevan ever tighter into Western institutions, and unlocking it grows costlier with every step - the same logic by which Central Europe was drawn in, only compressed in time and without the promise of membership. The key nuance: Armenia receives most of the package without membership, without candidate status, and without any obligation to leave the EAEU. Brussels is operating in a mode of "functional integration without accession" - which, for Yerevan, lowers the risk of an immediate Russian reaction.

What Holds It Back

Gas Is a Pipe

The turn West runs into a wall, and that wall is, quite literally, a pipe. Gazprom supplies about 82% of Armenia's gas; the remaining 18% comes from Iran through a "gas-for-electricity" barter. Gas covers roughly 63% of primary energy consumption; the single nuclear plant, providing 30-40% of electricity, runs on Rosatom fuel - and replacing that chain is impossible either quickly or without Moscow's consent.

The dependence is no accident: it was built by two deals of the Kocharyan era - "assets-for-debt" (2002-2003), which handed Russia industrial and energy assets in exchange for writing off a $93 million debt, and "assets-for-gas" (2006) - the handover to Gazprom of the Iranian pipeline, the fifth unit of the Hrazdan thermal plant, and an enlarged stake in the gas-distribution network. Gazprom took the final 20% of ArmRosGazprom in January 2014, in exchange for forgiving a $300 million debt - a few months after Sargsyan, in September 2013, was explaining to Putin that, for a CSTO member, joining the Customs Union was the only "rational decision." Most telling is the shift in the public grammar: in 2006 Gazprom deputy CEO Alexander Medvedev assured Yerevan that gas prices were "not tied to political alliances"; in 2025 Peskov put the opposite bluntly - the preferential price for Armenia is a function of EAEU membership. Not a threat, a formula.

A different gas source requires pipelines that do not exist; different nuclear fuel, years of licensing; replacing the plant, capital that the budget does not have. So even the most pro-Western government is forced to keep a working channel to Moscow open - literally, like a pipe. Hence both the caution in Pashinyan's rhetoric and the careful distancing from any talk of leaving the EAEU.

Trade That Grew Against the Politics

Standard logic says: the greater the political distance, the smaller the trade. With Armenia it is the reverse: trade with Russia grew fivefold from 2021 to 2024 - to a record $12.4 billion, and by the odd statistics of 2024 Armenia became the largest exporter of telephones to Russia ($620 million). The reason is obvious: much of the growth is parallel imports. Staying in the EAEU, Armenia is perfectly positioned as a transit point between Western producers and the Russian market. This delivers a GDP effect but makes the country hostage to two opposite processes: a tightening of sanctions threatens secondary sanctions, while an end to the war threatens a collapse of the transit margin. And the process has already begun: over the first eight months of 2025, trade with Russia collapsed by roughly half - as soon as the re-export of Russian gold, on which the record largely rested, was wound down. Remittances from Russia - whose net inflow fell 52% in 2024, per Armenia's finance ministry - add a second layer: tens of thousands of families still physically depend on the Russian labor market.

Security: A Hole With Nothing to Fill It

The most uncomfortable point. Armenia has left the Russian security umbrella without formally leaving it. What came instead? Weapons from India and France (the share of Russian hardware in arms imports fell from over 90% to under 10% by 2025). An EU mission - yes. But an Article 5 - no: in the event of a fresh Azerbaijani attack or a Turkish-Azerbaijani operation to force open a transport corridor through Syunik (the one Baku calls the Zangezur corridor), none of the new partners has given a binding guarantee.

Then again, the old system was no real umbrella either: Russia was contracting artillery and armor to Baku while remaining Yerevan's ally, and the 2015 "Lavrov plan," per leaked accounts, envisaged a Russian peacekeeping mission in place of an international one. Today's "hole" is the accumulated cost of two decades of single-track reliance on Moscow. And it is the main risk variable: if, in a new border crisis, the West - whose missions (EUMA) are observational and whose financial support (EPF) carries no combat guarantees - cannot or will not intervene, domestic politics will quickly raise the question of what the choice has cost. 2008 shows how this works: in Georgia the Western reaction to the war proved weaker than the leadership expected, and that later became one of the factors in the elite's turn.

Why Armenia Is Moving Faster Than Georgia - Right Now

Ten years ago the question ran the other way. Georgia had the Association Agreement, visa-free travel, then candidate status, and a steady 75-80% for the EU. Armenia, meanwhile, was still in the CSTO and the EAEU, with a Russian base at Gyumri. What changed? Not the public mood - in Georgia over 80% are still for the EU, and the protests in Tbilisi showed real mobilizing potential. What changed is the elite structure - and here the logic is awkward for optimistic theories of "transition":

The ownership structure of the elite matters more than its formal ideology. Georgian Dream is built around capital extracted from Russia; Armenia's ruling party is not. It also matters how that elite came to power: Ivanishvili won the 2012 election entirely lawfully, within an established system - and therefore inherited its networks, balances, and vulnerabilities instead of severing them.

The Western "escalator" is vulnerable to being stalled. With candidate status, Georgia became the object of heightened Brussels attention; in parallel, mechanisms of Russian counter-influence kicked in that had begun even before December 2023. Armenia is at an earlier, technical phase: the discrediting mechanisms that worked in Tbilisi do not yet "reach" it.

The war in Ukraine changed the EU's readiness. Georgia secured its status only after the doors were opened to Ukraine and Moldova; Armenia rides the same climate - but the terms are tightening: unanimity in the EU, enlargement fatigue, the positions of Hungary (which did not block the status in 2023 but then actively slowed the process) and Slovakia.

Karabakh closed the "open wound." While the conflict was active, exiting Russia's orbit was impossible - Moscow remained the only power broker. After 2023 that argument vanished, and the Armenian-Azerbaijani peace, shaky as it is, freed Yerevan's hands.

The comparison works not as "Armenia is better than Georgia" but as an illustration of how fragile post-Soviet reorientation is in general. Georgia shows you can travel 80% of the way and turn back. Armenia shows you can start from the opposite pole and cover a substantial stretch of functional integration in two or three years - if shock, elite coalition, and external anchor line up.

Where It Could Break

Elections - already held. Reorientation passed its first electoral stress test, but not cleanly. On June 7, 2026, Civil Contract took 49.8% of the vote and 64 of roughly 105 seats - a three-fifths constitutional majority (enough to pass constitutional laws and key appointments), but short of the two-thirds needed to amend the constitution. Telling is who came second: not Kocharyan's old bloc (it collapsed to about 10%, against 21% in 2021), but a new pro-Russian bloc, "Strong Armenia," led by billionaire Samvel Karapetyan - the very man the authorities detained in 2025 amid the clash with the Church and the one Moscow had bet on. In other words, the pro-Russian pole did not vanish; it reassembled around the "Church plus Russian money" pairing and won a notable 23% - but lost decisively, while the governing vector held its constitutional majority. The victory is nonetheless contested: Strong Armenia and the Kocharyan bloc are heading to the Constitutional Court, the CEC allowed a recount, and cases and travel bans have been opened against opposition leaders (Karapetyan, Kocharyan, Tsarukyan). Opponents call it pressure, the government calls it a fight against vote-buying.

Moscow's reaction underscored how much the frame has shifted. The Kremlin dragged out its congratulations ("we'll wait for the results"), Zakharova spoke of "unprecedented pressure on the opposition and Western interference" - a complaint about unfair elections, from Moscow, comments on itself - and on June 10, at a CSTO foreign-ministers' meeting in Kazan, Lavrov announced that the organization would consider applying a charter article to Armenia: stripping it of voting rights and quota posts over more than two years of unpaid dues. Pashinyan's answer: "we don't intend to change anything."

The constitution and peace. It is precisely the missing two-thirds that leads to the central unresolved knot - the peace treaty with Azerbaijan. Since 2023 Baku has demanded that Armenia strike from its constitution's preamble the reference to the 1990 Declaration of Independence (and through it, to the 1989 act on "reunification" with Nagorno-Karabakh), treating it as a territorial claim. Yerevan replies that its Constitutional Court has already found the reference no obstacle to recognizing Azerbaijan's territorial integrity. But legally, amending the preamble requires adopting a new constitution by national referendum, and the draft must first clear parliament by a two-thirds vote - which Civil Contract does not have. Pashinyan has announced a new constitution and has not ruled out a second vote if the first fails; the timeline is already slipping toward 2027. Here is the flip side of the result: the absence of a constitutional supermajority is also a reason not to rush an unpopular referendum, and to share responsibility for it with society at large rather than one party.

Domestic legitimacy. The government is in conflict with almost every load-bearing structure of Armenian identity - the Church, the Karabakh narrative, part of the diaspora: the standoff with the Catholicos became open after 2020, and in 2024 Archbishop Bagrat Galstanyan led protests against the handover of villages to Azerbaijan. The paradox: the American and French diasporas would prefer a pro-Western course - but they distrust the government carrying it out.

Gas and the nuclear plant. Any serious friction (a tariff, a supply restriction) instantly becomes a political instrument. Yerevan's strategy is to play for time and build alternatives (Iranian gas, a new nuclear plant, renewables), but their horizon is a decade, not a five-year plan.

Georgia. In the European Parliament in March 2026, Pashinyan called Georgia's frozen status the "main obstacle" to Armenian integration: Armenian gas and most of its trade with the West run through Georgian territory. What happens in Georgia after 2028 is a variable Yerevan cannot influence but on which it depends.

What This Says About Post-Soviet Reorientation in General

The Armenian example neither confirms theories of "democratic transition" nor refutes them - it shows how narrow the corridor of real choice is for a post-Soviet state, and on what conditions that corridor widens.

A shock is necessary but not sufficient: it only opens the window, and the elite that can pass through it is one with no old obligations to the departing patron. Hence the uncomfortable question: is a change of elites through rupture a precondition for genuine reorientation - as opposed to electoral rotation within an established system? The Georgian experience suggests elections alone do not do that work.

Full economic decoupling is unattainable - and perhaps not required. "Functional integration without accession" is not the 2004 enlargement with its full turn, but a hybrid mode in which political loyalty is detached from economic dependence, and in which instruments below candidate status (EUMA, EUPM, EPF, the visa plan) create more real integration than formal status: Georgia, with status, has less EU presence today than Armenia without it. This is a possible new model for the Eastern Partnership.

And the central (unsolved) problem of every post-Soviet pivot: security. Every case where it was closed with hard guarantees belongs to the window of 1999-2004, when NATO enlargement was still a realistic trajectory. The Baltics made it through that window and got Article 5 along with EU membership, the only post-Soviet case of a reorientation carried all the way to a hard anchor. Ukraine was left with the paper of the Budapest Memorandum, Georgia with the Bucharest "someday" of 2008. After 2008 the window closed for everyone: since then no post-Soviet state has obtained binding guarantees - and none left without them has completed its reorientation. Armenia is not making it through that window, and what to put in place of the anchor is an open question. TRIPP is a bet on economic distraction: that an American stake in lithium, copper, and transit will make the region "too valuable to attack." In peacetime it works; in a crisis it is untested.

The model passed the electoral test in June 2026: the bearer of the course held on to power and its constitutional majority, while the pro-Russian alternative lost without disappearing. That is already a precedent - an example of a small post-Soviet state getting through an electoral cycle without turning back. But the electoral test is not the decisive one. The decisive one has not yet come: the test of security. Until it does, partial reorientation holds firm right up to the first armed crisis - and it is that variable, not the next round of elections, that will determine whether the Armenian pivot has truly happened.

Kamshat Zhumagulova | Lead Analyst, Steppe Intelligence

June 17, 2026