"Hayk's Arrow" Armenian Strategic Center
Azerbaijan’s provisional 2025 data show slower real growth, a much narrower merchandise surplus and continued expansion of non-oil output and domestic finance.
Azerbaijan’s provisional 2025 data show slower real growth, a much narrower merchandise surplus and continued expansion of non-oil output and domestic finance. The strategic picture is therefore one of substantial resources alongside sectoral and external concentration, rather than uniform strength across all activities.
A common 2025 WDI frame separates the composition of output, employment and expenditure. Reported national-account shares and ILO modeled employment shares answer different questions; they are shown separately.
| Indicator | Value and reference period |
|---|---|
| Agriculture, forestry and fishing: value addedSector value added as a share of GDP; a composition measure, not a sector growth rate. | 5.9 % of GDP2025Reported result[122] |
| Industry, including construction: value addedIncludes mining, manufacturing, construction and utilities; manufacturing is a subset, not an additional sector to add. | 39.5 % of GDP2025Reported result[123] |
| Manufacturing: value addedManufacturing value added divided by GDP; excludes mining and is contained within the industry aggregate. | 6.0 % of GDP2025Reported result[124] |
| Services: value addedServices value added divided by GDP. Major sector shares need not sum to 100 because GDP includes taxes and other accounting adjustments. | 45.3 % of GDP2025Reported result[125] |
| Employment in agricultureILO modeled share of total employment in agriculture; it is not a national household-survey count or a measure of hours worked. | 34.2 % of total employment2025Estimate[126][131] |
| Employment in industryILO modeled sector share of employment, including construction. Different output and employment denominators cannot establish a wage or productivity ranking. | 15.5 % of total employment2025Estimate[127][131] |
| Employment in servicesILO modeled share of total employment in services; no conclusion about formality, job quality or earnings follows from the share alone. | 50.2 % of total employment2025Estimate[128][131] |
| Gross capital formationIncludes fixed capital, inventories and valuables. It is broader than fixed investment and is not the FDI inflow or proof of investment efficiency. | 18.9 % of GDP2025Reported result[129] |
| Exports of goods and servicesNational-accounts exports of goods and services relative to GDP; not customs-only goods exports or Armenian domestic value added embodied in exports. | 43.0 % of GDP2025Reported result[130] |
Azerbaijan’s industry share is much larger than its manufacturing share, making the distinction between resource-linked production and processing economically important. For Armenian firms, the relevant competitive question is a specific product, input cost and customer requirement; the broad industry total is not evidence that Azerbaijan has the same advantage in every manufacturing activity. Its employment composition also differs from its output structure. This can guide a search for complementarities, but cannot establish prices, quality or a legally usable bilateral supply channel. [123][124][127]
The agriculture employment share is materially larger than agriculture’s value-added share, so diversification cannot be assessed only by national output categories. Armenia should compare particular food and processing chains, including the ultimate origin of inputs, rather than assume a new regional seller removes an old upstream dependence. Azerbaijan’s capital-formation and export ratios add context but not a forecast of Armenian market access or a measure of available public support. Actual counterpart orders, payment and delivered margin remain the relevant commercial tests. [126][122][129][130]
The WDI vintage is 8 October 2026. Manufacturing is already included in industry; sector value-added shares need not sum to 100. The employment split is modeled and must not be substituted for a national labour-force survey. Dividing an output share by an employment share would not, without compatible prices, hours and coverage, establish a cross-country productivity or wage ranking. ILO models combine reported and imputed observations and may retain an older employment definition where changing standards would create a break. The exact model-to-survey reconciliation has not been established here; treat labour comparisons as qualified, not a ranking of national labour-market performance. [123][124][126][129][131]
Armenia's immediate choice is whether to spend scarce commercial-policy capacity anticipating an Azerbaijani market or reducing a dependency it already has.
Armenia's immediate choice is whether to spend scarce commercial-policy capacity anticipating an Azerbaijani market or reducing a dependency it already has. Azerbaijan's fiscal assets can cushion a resource-income shock, while its imports reveal dependencies shared with other regional buyers. Neither fact proves that Armenian firms currently have usable access to that market. The more robust starting point is an origin-and-route test for essential Armenian inputs. It can change purchasing and contingency decisions under today's lawful trading arrangements, and remains useful if bilateral opening is delayed. [119][118][72]
Give origin-diversification priority over building a broad new demand panel. A small review of actual invoices, origins and substitution times can identify a procurement decision; another national dashboard may have little value for firms with no demonstrated Azerbaijani exposure. The opportunity cost is delaying speculative market promotion, not abandoning it. Retain only the public indicators needed to explain a real commercial exposure, and protect firms' confidential terms. [68][72][59]
Proposed horizons run from authorisation. Later stages depend on the conditions below; these are analytical choices, not adopted deadlines or funding commitments.
Within the proposed first 90 days, use willing importers' existing records to distinguish producer country, intermediary, route and replacement lead time for their most consequential inputs. Choose a disruption case and a purchasing decision before collecting additional data. Record unknown origin rather than treating the invoice seller as the producer. [68][72]
Over the proposed next three to twelve months, test a commercially credible substitute or delivery chain through available legal channels. Compare identical quality, payment, inventory and full delivery costs. A buyer-validated supplier pilot can proceed in parallel where access, counterparties and settlement are already confirmed; it should not await a complete national database. [72][14][36]
Beyond twelve months, support testing, skills or logistics improvements only where repeated transactions reveal a common bottleneck and firms continue to pay. Revisit diversification when supplier concentration actually changes. End information collection that no longer changes a decision, and avoid converting a temporary service into a permanent subsidy. [36][37][116]
The binding constraint is reliable firm-level evidence and usable contracting options, not access to every Azerbaijani statistic. Customs origin, a seller's location and upstream production can diverge; sensitive invoices also require an agreed handling basis. Armenian agencies can convene and standardise the exercise, but firms must validate substitutions. Public financing should cover a defined common service, with operating responsibility identified before any scale-up. [72][59][116]
Put supplier-pilots first instead when Armenian firms already have a lawful channel, verified buyer demand and workable payment terms, while shared-origin exposure is small. Then paid orders provide better information than another diagnostic exercise. Conversely, a supply interruption calls for immediate procurement continuity before market development. These conditional choices are analytical priorities; national resource wealth alone cannot decide between them. [14][15][72]
Azerbaijan's fiscal buffer means lower oil receipts need not produce an immediate contraction in public spending or demand. SOFAZ's 2024 budget transfer was about one-third of state revenue, while the fund also held substantial financial assets. For Armenia, the transmission would depend on which expenditure is protected, which investment is deferred and whether import demand changes. Armenian businesses should therefore avoid interpreting an oil-price decline as either imminent Azerbaijani collapse or an automatic export opportunity. A larger neighbour can maintain purchases by drawing on savings, but prolonged adjustment can later affect contractors, distribution and household demand. This is an exposure channel, not a forecast of government behaviour. [118][51][32]
The manufacturing data point to an expanding productive base, but its output and domestic value added differ substantially. This creates two possible Armenian exposures if practical trade access develops: competition from Azerbaijani products and demand for complementary inputs, software, logistics or specialist services. Neither follows simply from GDP size. A supplier needs a specific buyer, usable payment arrangements, suitable standards and a delivered price that survives border costs. Energy-linked manufacturing may retain advantages unavailable to an Armenian competitor, while a specialised Armenian service can compete through quality and responsiveness. The appropriate unit of analysis is therefore a product and customer relationship, not a national verdict that one economy will inevitably displace the other. [14][15][36][42]
The State Statistical Committee’s macroeconomic table reports provisional 2025 GDP of 129.094 billion manats. The accompanying constant-price index implies 1.4% real growth, down from 4.2% in 2024 in the national series retrieved on 1 November 2026. Nominal output therefore grew while the real expansion slowed. The distinction matters because commodity prices and exchange-rate valuation can change the apparent size of an economy without an equivalent increase in produced goods and services. [1][2][9]
For Armenia, this describes a neighbour with significant resources but a changing growth cycle, rather than an unlimited economic advantage. National income, expenditure and constant-price series provide a better basis for assessing durable capacity than one dollar GDP ranking. A useful comparison is the persistence of growth across several years, together with household income and productive investment, because those channels influence demand, government room for manoeuvre and business confidence. [8][47][10]
The IMF Executive Board concluded the consultation in April 2026, following the February staff mission. The later release records nonhydrocarbon growth slowing from 6.5% in 2024 to 2.7% in 2025 and projects 2.1% overall growth for 2026. The forecast is identified by its publication vintage; it is not a measured 2026 outcome. [119]
The sector table separates oil and gas, non-oil activity and net product/import taxes. For 2025 it records 34.8886 billion manats in oil-gas output and 81.8073 billion in the non-oil sector. These categories do not sum to GDP until net taxes are included. This breakdown makes it possible to examine diversification more carefully than equating every non-oil manat with a new tradable manufacturing industry. [32][13][25]
Industrial output, industrial value added and intermediate consumption answer different questions about domestic depth. A processing activity can generate high sales while retaining less value locally if imported inputs are large. For Armenia, competition and cooperation are sector-specific: suppliers, logistics and services may benefit from expansion, while headline diversification can still coexist with hydrocarbon-led export earnings. The input-output tables offer a route to testing those linkages. [15][17][42]
The customs-based trade series records 2025 merchandise exports of $25.0428 billion and imports of $24.3797 billion, giving a surplus of $663.1 million. The comparable 2024 surplus was $5.4961 billion. The source also publishes a separate statistically adjusted export series; the figures above consistently use the ordinary customs series. The narrowing reflects rising nominal imports and lower exports, not necessarily an equivalent change in physical freight. [57][59][61]
Geographic tables put Europe at 77.5% of exports and Asia at 45.8% of imports in 2025. Europe is a geographic category, not the European Union alone. These asymmetries show why market access and supplier diversification require separate analysis. For Armenia, the practical question is which commodities and counterparties might overlap with Armenian firms or regional routes; aggregate bilateral turnover does not itself prove dependence or political alignment. [62][63][58]
Trading partners have different economic functions. In 2025 Italy bought USD11.342 billion of Azerbaijani exports, about 45.3% of the total; the commodity-country table shows its prominent role in oil and gas. China supplied USD4.790 billion of imports but bought only USD84.4 million of exports. Russia supplied USD3.736 billion and bought USD1.184 billion, including substantial agricultural goods. Thus European energy demand, Chinese equipment supply and Russian food-market access create separate exposures. For Armenia, a diplomatic disagreement with one partner need not interrupt every commercial channel, but disruption in a concentrated product market can still reach farms, logistics and household income. [59][61][70]
The World Bank’s April 2026 update places the full-year 2025 current-account surplus at 4.6% of GDP, down from 6.3% in 2024. This balance includes services and income flows, unlike customs merchandise trade. It therefore complements rather than contradicts the customs surplus above; the IMF’s 5.4% figure refers only to 2025’s first three quarters. [121][119]
The commodity tables separately identify animals and food, machinery, chemicals, metals and energy products, with values and selected physical quantities. They are more informative for supply-chain research than the overall trade balance. For example, a country can be a net exporter in total while depending on imports of specific industrial equipment, medicines or food categories. Product-level import and export series expose those different positions. [66][67][68]
Retail and wholesale series then show how goods reach domestic buyers, distinguishing turnover, real-volume indices, stocks and domestic versus imported products. For Armenia, this helps identify potential commercial niches under any future lawful market opening, while avoiding the assumption that political normalisation automatically creates demand. Competitiveness would still depend on product standards, payment channels, distribution costs and whether domestic production already serves that market. [103][94]
The macroeconomic baseline reports 2025 state-budget revenue of 39.1846 billion manats and expenditure of 38.6046 billion. These are state-budget figures, not a comprehensive consolidated public-sector balance sheet. The detailed historical budget workbook identifies tax and expenditure categories, while population-income and compensation tables indicate how resources are distributed between households and sectors. Combining these sources is more informative than reading spending capacity from GDP alone. [51][39][26]
Armenian analysis should distinguish resources available for discretionary policy from obligations already financing public services, infrastructure and household support. Regional income differences also matter for the sustainability of domestic demand and the geographic reach of economic growth. The material supports these questions but does not by itself identify all sovereign-fund transfers, contingent liabilities or financing commitments; those require separate fiscal and institutional records. [49][27][20]
SOFAZ turns resource income into both financial savings and budget spending. In its 2024 execution statement the fund reports 12.244 billion manats of oil-and-gas agreement revenue, 5.795 billion of investment income and a 12.781 billion transfer to the state budget. That transfer equals about 34.4% of the statistical committee’s 37.162 billion total budget revenue for 2024, calculated across the two sources. The fund’s USD60.031 billion year-end assets are a stock, not annual spending capacity: investment returns, valuation changes and withdrawals all affect its evolution. This mechanism cushions short-term oil shocks but transmits resource dependence into expenditure planning. [118][51]
SOFAZ reports USD 73.541 billion of assets at end-2025, up USD 13.510 billion over the year. Gold accounted for 38.2% of its investment portfolio, or 200 tonnes. These are fund-reported asset stocks: the increase must not be equated with new oil receipts, cash available to spend, or the fund’s investment-return rate. [120]
The investment series records 34.218 billion manats from all sources in 2025, including 16.6818 billion classified as foreign investment. This broad category must not be relabelled net foreign direct investment. Year-end loans to the economy reached 31.9467 billion manats, compared with 29.2882 billion in 2024. These figures show financing scale; they do not alone establish the quality, additionality or productivity of investment. [54][55][37]
Household bank savings rose to 16.7724 billion manats in 2025, with 11.6028 billion held in national currency. Alongside labour productivity and fixed-asset tables, this provides a basis for examining domestic financing and capital deepening. For Armenia, the strategically relevant question is whether financing builds competitive firms and transferable capabilities, or mainly supports consumption and existing asset values. That distinction requires sector-level follow-through rather than a league table of financial totals. [56][36][34]
External context: the IMF’s February 2026 staff statement also reports 1.4% growth in 2025 after 4.2% in 2024, and emphasises productivity, competition and capital-market development. This is a staff assessment, not an Executive Board decision. The World Bank’s January 2025 country framework prioritises better jobs and resilience. Together they support examining the quality of diversification rather than treating larger nominal financial totals as sufficient evidence. [116][117]
Manufacturing is a material productive base, although smaller than extraction. Provisional 2025 manufacturing output is 21.365 billion manats and manufacturing value added 7.720 billion; the difference largely represents purchased inputs rather than income retained by manufacturers. The physical output index indicates 7.9% growth in 2025. Manufacturing value added per employed person is 25,112 manats, compared with 4,311 in agriculture, forestry and fishing, in current prices. This gap helps explain why moving workers into more productive activities can matter more than expanding employment headcounts. It does not establish that every worker can switch sectors without training, housing or investment. [14][15][34][36]
The supply-and-use accounts distinguish food processing, refined petroleum, chemicals, plastics, metals and construction materials. Their economic roles differ: refining upgrades hydrocarbons, food processing links farm supply to urban and export demand, and construction materials depend heavily on domestic building. Manufacturing fixed assets reached a reported 17.106 billion manats in 2025, compared with 7.210 billion in 2015 at current prices. That is evidence of a larger recorded capital base, but inflation and valuation prevent calling it a doubling of physical capacity. Armenia’s possible commercial openings would lie in specific complementary inputs and services, conditional on practical trade access. [40][42][52][37]
Azerbaijan remains a substantial wheat importer despite its domestic agricultural sector. In 2025 it imported 1.2673 million tonnes worth USD275.6 million; the partner table assigns 604,900 tonnes to Russia. Imported grain therefore connects bread costs and milling to harvests, export conditions and transport outside Azerbaijan. This is a shared regional exposure rather than evidence that Azerbaijani farms produce nothing. For Armenia, reopening trade could add sourcing options, but exchanging goods through a neighbour that also imports them would not remove upstream dependence. Price transmission would depend on competition, border costs and the ability of firms to switch suppliers. [68][72][96]
Lower hydrocarbon receipts can narrow fiscal room even as non-oil output expands, because budget transfers and export earnings remain linked to resource income. [118][32]
Concentrated energy buyers and food-supply relationships create product-specific vulnerabilities; a diversified list of diplomatic partners does not diversify every commercial market. [61][70][72]
Manufacturing growth and a larger capital base can create complementary supplier and service demand for Armenian firms if market access and commercial payment arrangements become dependable. [14][15][36]
Accumulated financial assets can finance productive diversification and human capital, provided investment generates sustained output rather than only construction expenditure. [118][54]
These scenarios explore how documented drivers could affect Armenia. They are conditional analytical paths, not predictions or assigned probabilities.
Conditional scenario, not a forecast: if reciprocal trade becomes reliable and manufacturing continues to expand, Armenian firms could supply selected inputs and services. Gains would depend on actual orders, transport costs and payment reliability; shared imported-input dependence would persist. [14][15][72]
Conditional scenario, not a forecast: if hydrocarbon receipts weaken while public commitments remain high, accumulated assets could cushion spending before adjustment reaches domestic demand. Armenia should expect uneven effects across Azerbaijani buyers and sectors rather than infer either immediate collapse or unlimited fiscal endurance. [118][32][57]
Follow revisions to provisional 2025 GDP and distinguish real growth from current-price changes. [2]
Monitor the merchandise surplus with one consistent customs definition and commodity breakdown. [57]
Test diversification through value added, productivity and investment outcomes, not sector labels alone. [15][36]
These are analytical recommendations based on the evidence, not adopted government decisions or commitments by the named institutions. Timelines and measures below are proposed implementation tests.
Create a small quarterly panel linking Azerbaijan's resource receipts, budget execution, merchandise imports and sector demand to Armenian firms' actual exposures.
Create a small quarterly panel linking Azerbaijan's resource receipts, budget execution, merchandise imports and sector demand to Armenian firms' actual exposures. Begin with existing public series and a voluntary, anonymised survey of exporters and distributors; record indirect exposure through third countries separately.
Ask major Armenian food and input importers to map origin, intermediary, border route, settlement currency and substitution time for essential products.
Ask major Armenian food and input importers to map origin, intermediary, border route, settlement currency and substitution time for essential products. Begin with a confidential tabletop interruption exercise and compare an Azerbaijan-linked option with existing suppliers on identical product specifications.
Invite Armenian manufacturers and service firms to identify a short list of complementary products, then prepare buyer-validated pilot offers.
Invite Armenian manufacturers and service firms to identify a short list of complementary products, then prepare buyer-validated pilot offers. Start with cost sheets covering standards, origin documentation, delivery, insurance and payment; implement shipments only through legally available channels and voluntary commercial counterparties.
Use product-level evidence from pilots to select Armenian bottlenecks that public policy can plausibly address, such as testing access, workforce skills or logistics coordination.
Use product-level evidence from pilots to select Armenian bottlenecks that public policy can plausibly address, such as testing access, workforce skills or logistics coordination. Begin with firm-level diagnostics and publish the reason for each intervention before approving support for equipment or training.
The count includes distinct documents whose content was retrieved and checked for relevance to this profile. Translations, duplicate links and generic homepages are excluded. A document may support more than one topic when its content is relevant to each. Historical documents retain their original dates; they do not establish that a project or policy remains unchanged today.
Sources labelled “Excerpt checked” were reviewed only in the retained relevant passage. They can support the cited claim but are excluded from the full-document total.
Filters use the source date recorded in this library. Indicator reference periods are shown separately.
Annual real growth in the national statistical series.
The 2025 figure is provisional. These are changes in real output, not nominal GDP values.
| Observation | Value · % year on year | Reference period | Status | Sources |
|---|---|---|---|---|
| 2024 | 4.2 | 2024 | Reported result | |
| 2025 | 1.4 | 2025 | Provisional |
A common 2025 WDI frame separates the composition of output, employment and expenditure. Reported national-account shares and ILO modeled employment shares answer different questions; they are shown separately.
On smaller screens, scroll the table sideways to compare all three countries.
| Indicator | 🇦🇲 Armenia | 🇦🇿 Azerbaijan | 🇬🇪 Georgia |
|---|---|---|---|
| Agriculture, forestry and fishing: value added% of GDPMatched period and definition | 7.9 % of GDP2025Reported resultSource | 5.9 % of GDP2025Reported resultSource | 5.2 % of GDP2025Reported resultSource |
| Industry, including construction: value added% of GDPMatched period and definition | 23.1 % of GDP2025Reported resultSource | 39.5 % of GDP2025Reported resultSource | 18.6 % of GDP2025Reported resultSource |
| Manufacturing: value added% of GDPMatched period and definition | 11.0 % of GDP2025Reported resultSource | 6.0 % of GDP2025Reported resultSource | 8.0 % of GDP2025Reported resultSource |
| Services: value added% of GDPMatched period and definition | 61.3 % of GDP2025Reported resultSource | 45.3 % of GDP2025Reported resultSource | 63.9 % of GDP2025Reported resultSource |
| Employment in agriculture% of total employmentRead with stated qualifications | 27.0 % of total employment2025EstimateSource 1Source 2 | 34.2 % of total employment2025EstimateSource 1Source 2 | 34.0 % of total employment2025EstimateSource 1Source 2Source 3 |
| Gross capital formation% of GDPMatched period and definition | 24.2 % of GDP2025Reported resultSource | 18.9 % of GDP2025Reported resultSource | 21.3 % of GDP2025Reported resultSource |
Sector value added as a share of GDP; a composition measure, not a sector growth rate. Same WDI indicator, common observation year and denominator across the three countries. National revisions remain possible.
Compare sector structure, not agricultural growth or the return from moving a worker. Armenia still needs product, land, water and buyer evidence for a specific intervention.
Includes mining, manufacturing, construction and utilities; manufacturing is a subset, not an additional sector to add. Same WDI indicator, common observation year and denominator across the three countries. National revisions remain possible.
Azerbaijan’s larger industry share cannot be read as an across-the-board manufacturing advantage; the aggregate includes mining, construction and utilities.
Manufacturing value added divided by GDP; excludes mining and is contained within the industry aggregate. Same WDI indicator, common observation year and denominator across the three countries. National revisions remain possible.
The narrower manufacturing measure helps Armenia distinguish processing capacity from a resource- or construction-heavy industry total; it does not measure competitiveness at a delivered price.
Services value added divided by GDP. Major sector shares need not sum to 100 because GDP includes taxes and other accounting adjustments. Same WDI indicator, common observation year and denominator across the three countries. National revisions remain possible.
Service-heavy output can support Armenian regional partnerships, but the aggregate cannot identify a buyer or separate tourism, finance, transport and digital work.
ILO modeled share of total employment in agriculture; it is not a national household-survey count or a measure of hours worked. Same WDI model series and year, but not interchangeable with national LFS. ILO may retain older definitions to avoid breaks; Georgia changed national standards in 2020. No complete country-specific reconciliation of the model-to-survey gaps is established here.
The common modeled series provides a structural comparator; Armenian regional training and service choices still require local demand and must not use a value-added-share/employment-share quotient as a wage ranking.
Includes fixed capital, inventories and valuables. It is broader than fixed investment and is not the FDI inflow or proof of investment efficiency. Same WDI indicator, common observation year and denominator across the three countries. National revisions remain possible.
A higher ratio is not proof of a more productive public project. Armenia should assess completion readiness, inventories, recurring costs and the service actually delivered.
The WDI vintage is 8 October 2026. Manufacturing is already included in industry; sector value-added shares need not sum to 100. The employment split is modeled and must not be substituted for a national labour-force survey. Dividing an output share by an employment share would not, without compatible prices, hours and coverage, establish a cross-country productivity or wage ranking.
All values are rounded to one decimal; original precision and indicator metadata are retained in the research notes. Source counts do not imply independent corroboration.
Several countries or indicators can share one underlying data release. Multiple source links do not mean independent confirmation.
These dated records preserve the earlier research edition. Later source releases and statistical revisions are reflected in the detailed analysis above.
Reference period: IMF assessment published April 2025
The IMF's April 2025 assessment identified lower oil and gas prices as a risk to growth, the external position and fiscal revenue.
Reference period: 2024; IMF April 2025 assessment
The IMF reported 2024 real GDP growth of 4.1%, with non-hydrocarbon output up 6.2% and hydrocarbon output up 0.3%.
This is a curated starting collection, not an exhaustive country database or a live feed. Each record identifies its reference period and links to original evidence. “Announced” does not mean funded, ratified or operational; those distinctions are stated in the text. Armenia implications are editorial analysis, not source quotations. Census counts, population estimates and forecasts must not be treated as interchangeable. Source-check dates do not change the underlying data period.