## 1. Exports of Goods

## Source details

**Canonical URL:** [1. Exports of Goods](https://www.imf.org/-/media/files/publications/cr/2026/english/1mdaea2026002-source-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/cr/2026/english/1mdaea2026002-source-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/cr/2026/english/1mdaea2026002-source-pdf.pdf.json)

---

### Context
- Moldova is a small, open economy where expanding and diversifying exports is essential for sustained economic growth and income convergence.
- The EU Association Agreement (including the DCFTA) was signed in 2014 and entered into force in 2016; it lowered tariffs, simplified customs procedures, and harmonized standards, facilitating access to the EU single market.
- Moldova dismantled its tariffs on certain products (including plastics, furniture and textiles) in 2019.
- The 2025 update of the DCFTA envisages further reduction or elimination of customs duties for remaining products.
- Joining the EU would entail adoption of the Acquis Communautaire and participation in the single market’s four freedoms; evidence from other CEE countries suggests EU accession typically raises exports, FDI, and per capita incomes.
- Moldova’s Growth Plan is underpinned by a Reform and Growth Facility which could disburse up to €1.9 billion in concessional loans and grants for 2025-27, conditional on implementation of the Reform Agenda.

### Export performance indicators
- Total exports increased by 52 percent in the post-DCFTA period examined.
- The EU’s share of Moldova’s exports rose from 53 percent to 67 percent between 2014 and 2024.
- Main drivers of export growth: transport equipment, foods, and crude materials — together contributed over 70 percent of overall export growth in the period.
- Exports surged in 2021-22 but have declined in the last couple of years, likely reflecting difficulties related to the war in Ukraine.
- Moldova remains open but trade openness has been on a declining trend.

### Extensive and intensive margins
- Export growth decomposed into extensive (new products/destinations) and intensive (larger shipments of existing product–market pairs) margins following Hummels and Klenow (2005).
- Both margins improved over time; the intensive margin toward the EU stands out as the key driver in recent years.
- Largest gain came from exporting more of products already established in EU markets; extensive-margin gains continued but were smaller in comparison.

### Diversification and concentration
- Moldova’s share in global exports has gradually risen, driven mainly by deeper integration with the EU.
- Number of export products (HS6-level) and export destinations expanded over the past decade, though 2023 data show some decline in both—potentially linked to logistical difficulties from the war in Ukraine.
- Product concentration indicators point to increasing reliance on a relatively limited set of goods; product concentration is somewhat higher for exports to the EU except in 2022–23 when product HHI in trade with non-EU countries increased noticeably.
- Market diversification (share of Top 10 destinations) declined somewhat relative to the mid-2000s and has largely been flat in recent years.
- Relative to peers: EU members like Romania, Bulgaria, and Lithuania have broader product and market bases and lower concentration indices; non-EU peers (Armenia, Georgia, Belarus) show higher concentration. Moldova’s trajectory is toward greater diversification but convergence will take time.

### Export relationships durability (survival analysis)
- Median duration of country–product pair relationships for Moldova is 3 years.
- Kaplan–Meier estimates show steep declines in survival probabilities during the first 2–3 years across most markets, sectors, and cohorts; attrition slows thereafter.
- After three years, the EU survival rate is about 0.6, compared to 0.5 for non-EU markets—this difference is highly statistically significant.
- Initial export size matters: relationships beginning with larger shipment values (top quartile) survive substantially longer than small-value entries (bottom quartile).
- Variation by product: clothing and oil seeds display higher persistence; electrical equipment and apparatus, and animal or vegetable fats tend to be more short-lived.

### Export quality and complexity
- Export quality assessed using unit-value (UV) ladders at the HS6 level: higher unit values within identical product categories interpreted as higher-quality or more differentiated varieties.
- Moldova’s median UV position fluctuates around 0.3 vs. an EU average of about 0.55.
- 2014 UV distribution bi-modal (peaks ~20th and ~80th percentile); 2023 UV distribution shows the left (low-UV) peak much less pronounced and compression in the middle — some products upgraded or mid-range UV varieties entered.
- Top UV performers dominated by car seats and components and clothing items; high unit-value items are differentiated manufactured goods despite significant agricultural exports.
- Product space: Moldova’s RCA concentrated in peripheral areas, predominantly agricultural and textile sectors; number of HS4-level products with RCA declined from 128 to 122 between 2014 and 2023.
- ECI dynamics: ECI declined in the mid-2010s, recovered after DCFTA, with recent gains reflecting growth within existing low-to-mid-complexity categories rather than a shift into more sophisticated products.
- Feasibility of diversification using existing capabilities remains limited; most nearby products are low-to-moderate complexity.

### Potential effects of EU membership (gravity model evidence)
- Gravity model panel: bilateral exports for former transition economies currently in the EU and accession countries, 1996-2023, estimated in 3-year intervals and at HS2 level.
- Three specifications estimated:
  - (i) traditional gravity with exporter-time and importer-time fixed effects;
  - (ii) gravity with paired fixed effects;
  - (iii) model allowing heterogeneous impacts across industries.
- Key quantitative results:
  - Traditional gravity equation estimates a very large effect of joining the EU—nearly 45 percent (calculated as 100*(exp(β)-1)).
  - Paired fixed effects estimation associates EU membership with a nearly 16 percent increase in exports.
  - Hagemejer and Mućk (2025) estimate a positive effect from EU enlargement of about 14 percent.
- Selected gravity table coefficients (preserved):
  - RTA: 0.011 (model 1); 0.117*** (model 2); 0.112*** (model 3).
  - EU: 0.371*** (model 1); 0.148** (model 2).
  - Observations: 706306 (model 1); 817773 (models 2 and 3).
- Sector-specific interaction coefficients (model 3; interaction with EU indicator):
  - Textiles (HS50-63) #1: -0.202**.
  - Metals (HS72-83) #1: 0.205**.
  - Machinery/electrical/transport (HS84-89) #1: 0.378***.
  - Other (HS90-97) #1: 0.273***.
  - Agri-food (HS1-24) base: 1.EU -0.055 (not significant).
  - Minerals (HS25-27) #1: -0.203 (not significant).
  - Chemicals & plastics (HS28-40) #1: 0.142 (not significant).
  - Other light manuf (HS41-49, 64-71) #1: 0.035 (not significant).
- Interpretation:
  - EU membership could meaningfully increase exports, with particularly strong effects in machinery, electrical equipment, metals, and other sectors linked to European production networks.
  - Benefits are uneven across industries; weaker or negative effects in labor-intensive, low-technology sectors (e.g., textiles).
  - Impact on agricultural and food products is inconclusive (coefficient not statistically significant).

### Policy-relevant findings and implications (exports)
- DCFTA has been a major driver of export gains, especially through deeper penetration (intensive margin) into EU markets.
- EU accession could consolidate DCFTA gains and further boost exports, but effects will differ across sectors; structural and regulatory alignment will be critical.
- Key constraints to accelerating export diversification and sophistication include infrastructure bottlenecks, limited access to finance, and vulnerability to external shocks.
- Policy priorities to support more durable, higher-value exports:
  - Invest in quality upgrading and standards compliance to capture higher UV segments.
  - Improve logistics and reduce trade frictions to lower entry costs and improve survival rates of new relationships.
  - Enhance access to finance for exporters to support scaling and initial market commitment.
  - Pursue deepening structural reforms required for EU integration and to leverage the Reform and Growth Facility resources.
  - Target sector-specific measures recognizing that intensive benefits of EU market access vary across industries.

### EU Reform and Growth Facility and export-led growth
- Focused use of concessional EU resources can:
  - Address infrastructure bottlenecks.
  - Expand quality infrastructure.
  - Support firm-level upgrading to crowd in private investment and accelerate integration into European value chains.
- With sustained reform momentum, Moldova can transition from a low- to a mid-complexity export structure and achieve more resilient, export-led growth.

### Methodology and data (Appendix I)
- Data sources:
  - CEPII BACI (Gaulier, G., Zignago, 2010) and CEPII Gravity (Conte et al., 2022) databases.
  - Atlas of Economic Complexity (Hausmann et al., 2013).
- Data specifics:
  - CEPII BACI annual bilateral trade flows data at the HS6 level (2007 edition) for 2007-2023 used for diversification metrics, survival analysis, and export quality indicators.
  - Gravity model estimated with CEPII BACI bilateral data covering 1996-2023, aggregated to HS2 level, augmented with CEPII Gravity variables: DIST, CONTIG, COMLANG, COMCOL, RTA, EU.
- Export concentration and margins: normalized Herfindahl-Hirschman indices; extensive and intensive margins follow Hummels and Klenow (2005).
- Survival analysis: follows Cox and Oakes (1984) and Kleinbaum and Klein (2012); Kaplan–Meier estimator applied.
- Gravity model specification: Models (1)-(3) with exporter-time, importer-time, exporter-importer fixed effects, lnDIST, CONTIG, COMLANG, COMCOL, RTAij,t, EUij,t, and sector indicators Dkg.

### Export conclusions (summary)
- Deeper trade integration with the EU benefited Moldova’s exports: DCFTA supported export expansion and regulatory modernization; progress in penetrating EU markets and improving durability of export relationships.
- Export growth has been uneven; recent shocks exposed vulnerabilities; export basket remains concentrated in a narrow set of products.
- Quality and complexity improved gradually but Moldova lags peers with more diversified and sophisticated export structures.
- EU membership could amplify positive trends but will not automatically overcome structural constraints.

---

### Moldova’s inflation targeting regime and monetary policy (context and evolution)
- NBM adopted formal IT in late 2012; operationalized in 2013.
- Inflation target range: 5+/- 1.5 percent.
- NBM began publishing quarterly Inflation Reports and introduced forecast-based decision-making supported by a macroeconomic model and scenario analysis.
- Inflation has been volatile since end-2014 due to high share of exogenously determined goods (food and fuels) and regulated prices (gas and electricity) affected by Russian gas supply and pricing up to 2025.
- Median pass-through to CPI for MDL-US dollar exchange rate (exchange-rate-driven changes): 45 percent.
- Pass-through elasticity for MDL-euro exchange rate (exchange-rate-driven changes): 53 percent.
- For exchange rate changes driven by domestic monetary policy shocks: pass-through from an exchange rate change to the CPI is 62 percent for both MDL-US dollar and MDL-euro rates.
- Interpretation: around half of an exchange rate movement transmits to the CPI for exchange-rate-driven shocks.

### NBM instruments and evolving operations
- Complementary instruments: reserve requirements (RRs), open market operations, and standing facilities.
- Two typical IT challenges:
  - Fear of floating due to liability dollarization or high pass-through can lead to de facto exchange rate targeting and heavy use of FX interventions (FXI).
  - Shallow financial markets and excess liquidity (post-2014 bank fraud and emergency lending) led the NBM to increase RRs to exceptionally high levels and issue short-term certificates of deposit (CDs).
- CDs: base rate applied; maturities generally 14 days.
- Repo rate: set at +0.25 pp above the base rate.
- CDs peaked just below 5 percent of GDP in March 2018 and again in July 2023; by December 2025 CDs had declined to 1.4 percent of GDP.
- Repos generally used less, but saw a pickup in late 2025.

### Empirical findings on transmission and FXI
- Shift in base rate behavior:
  - Pre-2020: Base rate did not react significantly to the inflation gap; instead, base rate increased by 7 basis points to a quarterly depreciation in the leu-dollar exchange rate of 1 percent (significant at the 1 percent level).
  - Since 2020: Base rate displays a significant reaction to the inflation gap; exchange rate coefficient smaller and significant only at the 10 percent level.
- Interest-rate transmission:
  - Since 2021, deposit rate rises 0.19 percentage points on impact after a one-percentage-point increase in the base rate; pass-through complete after about 4 months with a total increase of 0.55 percentage points.
  - Lending rate: no immediate reaction on impact; after 3 months it settles at 0.3 percentage points higher.
  - T-bill rates react quickly and strongly: since 2021, 71 percent of base rate hikes were transmitted to T-bills within a few months.
  - Example: Energy price hike in January 2025 — base rate increase of 290 bps was followed by increase of bank loan rates of 110 bps until July 2025; lending pass-through somewhat higher at 0.38 in that episode.
- FX interventions (FXI):
  - Use of FXI declined substantially post-2020.
  - Pre-2020: one-percent month-over-month depreciation of leu against dollar associated with FX sales of 0.09 percent of annual GDP.
  - Post-2020: reaction coefficient not significant; interventions smaller relative to GDP and less biased toward dollar sales.
  - Degree of exchange rate management has decreased, indicating reduced focus on preventing depreciations.

### Taylor rule empirical coefficients (subsamples; notes preserved)
- 2013Q1-2025Q2
  - Inflation gap: 0.1-0.03
  - Output gap: 0-0.15-0.03
  - Change in exchange rate: 0.07***
- 2013Q1-2019Q4
  - Inflation gap: 0.1-0.03
  - Output gap: 0-0.15-0.03
  - Change in exchange rate: 0.07***
- 2020Q1-2025Q2
  - Inflation gap: 0.54***
  - Output gap: -0.03
  - Change in exchange rate: 0.04*
- Notes preserved: *0.05<p≤0.1, ** 0.01<p≤0.05, *** p≤0.01.

### Reserve requirements (RRs): history, distortions, and costs
- NBM raised the reserve requirement ratio on deposits in lei from 14 to 35 percent between February and November 2015 and then to 42.5 percent in October 2018.
- In December 2025, the NBM reduced the RR ratio to 20 percent.
- Required reserves remuneration: first five percent of the reserve base not remunerated; remaining amount remunerated at the overnight deposit rate (ONDR), which is currently two percentage points below the base rate.
- At RR ratio of 20 percent and a base rate of 5 percent in December 2025, the unit cost was 2.8 percent.
- Total cost in December 2025: about MDL 487 mln (roughly 1.6 percent of banks’ equity).
- Liability dollarization remains sizable: 20-percent share of FX loans in total loans.

### Model simulations: effects of lowering RRs (QIPF-based experiment)
- Experiment: 10 percent reduction in the RR from 22 to 12 percent as indicative experiment.
- Reduction assumed perfectly credible and perfectly anticipated; RR adjusts fully within four years, half-way point after 18 months.
- Simulated banking and macro responses:
  - Nominal loan rates and spreads fall on announcement; stocks of loans and deposits rise.
  - Higher loan volumes drive profitability; banks’ market-to-book ratios undergo a sustained rise.
  - Overall macroeconomic effects small in the model; new credit directed to households, boosting asset prices and consumption short term.
  - Model does not capture likely benefits from boosting business lending (e.g., more capital deepening and improvements in labor productivity).

### Policy discussion and recommendations (monetary and RRs)
- Monetary policy practice and credibility:
  - Continue focusing on the inflation gap in setting the base rate to bolster credibility and accountability.
  - Build a track record of following IT principles and improve transparency per Central Bank Transparency Code Review.
  - Strengthen NBM’s political, institutional, and financial independence.
- FXI practices:
  - Use FXI to address FX market disruptions, counter financial stability risks from FX mismatches, and support price stability when exchange rate changes risk de-anchoring inflation expectations.
  - Limit systematic exchange-rate targeting; avoid excessive reliance on FXI.
- RR policy guidance:
  - NBM should reduce the level of RRs further once inflationary pressures have subsided.
  - Pursue a carefully calibrated reduction of RRs while monitoring inflation developments.
  - A slow release over time would avoid a possibly undesirable credit boom; released liquidity could be absorbed by issuance of CDs.
  - Endpoint calibration: regional peers local currency RRs are in the range of 5.5 and 10 percent; optimal level for Moldova should consider banks’ liquidity needs, shocks, and interbank market development.
- Market and interbank benefits:
  - Lower RRs would support government debt market development and the interbank market.
  - Repos could be used increasingly for open market operations with government debt as collateral.
  - Better functioning sovereign bond market should help revive the interbank market.

*Prepared by Rossen Rozenov and Yiran Zha (EUR). Source: IMF staff analysis from "MOLDOVA – EXPORT DEVELOPMENTS AND PROSPECTS".*

### 1. Exports of Goods _________________________________________________________________  5

### 1. Exports of Goods

### Context
- Moldova is a small, open economy where expanding and diversifying exports is essential for sustained economic growth and income convergence.
- The EU Association Agreement (including the DCFTA) was signed in 2014 and entered into force in 2016; it lowered tariffs, simplified customs procedures, and harmonized standards, facilitating access to the EU single market.
- Moldova dismantled its tariffs on certain products (including plastics, furniture and textiles) in 2019.
- The 2025 update of the DCFTA envisages further reduction or elimination of customs duties for remaining products.
- Joining the EU would entail adoption of the Acquis Communautaire and participation in the single market’s four freedoms; evidence from other CEE countries suggests EU accession typically raises exports, FDI, and per capita incomes.
- Moldova’s Growth Plan is underpinned by a Reform and Growth Facility which could disburse up to €1.9 billion in concessional loans and grants for 2025-27, conditional on implementation of the Reform Agenda.

### Export Performance Indicators
- Exports expanded notably following the DCFTA: total exports increased by 52 percent in the post-DCFTA period examined.
- The EU’s share of Moldova’s exports rose from 53 percent to 67 percent between 2014 and 2024.
- Main drivers of export growth were transport equipment, foods, and crude materials, which together contributed over 70 percent of the overall export growth in the period.
- Exports surged in 2021-22 but have declined in the last couple of years, likely reflecting difficulties related to the war in Ukraine.
- Moldova remains open but trade openness has been on a declining trend.

### Extensive and Intensive Margins
- Export growth decomposed into extensive (new products/destinations) and intensive (larger shipments of existing product–market pairs) margins.
- Both margins improved over time, but the intensive margin toward the EU stands out as the key driver in recent years.
- The largest gain came from exporting more of products already established in EU markets, reflecting positive effects from trade liberalization; extensive-margin gains continued but were smaller in comparison.

### Diversification and Concentration
- Moldova’s share in global exports has gradually risen, driven mainly by deeper integration with the EU.
- The number of export products (HS6-level) and export destinations expanded over the past decade, though 2023 data show some decline in both—potentially linked to logistical difficulties from the war in Ukraine.
- Product concentration indicators point to increasing reliance on a relatively limited set of goods; product concentration is somewhat higher for exports to the EU except in 2022–23 when product HHI in trade with non-EU countries increased noticeably (likely reflecting increased shipments to Ukraine and decreased exports to Russia).
- Market diversification (share of Top 10 destinations) declined somewhat relative to the mid-2000s and has largely been flat in recent years.
- Relative to regional peers: EU members like Romania, Bulgaria, and Lithuania have broader product and market bases and lower concentration indices; non-EU peers (Armenia, Georgia, Belarus) show higher concentration. Moldova’s trajectory is toward greater diversification but convergence will take time.

### Export Relationships Durability
- Survival of product–destination export links is a key metric; most relationships face high termination risk in the first few years.
- Median duration of country–product pair relationships for Moldova is 3 years.
- Kaplan–Meier estimates show steep declines in survival probabilities during the first 2–3 years across most markets, sectors, and cohorts; attrition slows thereafter.
- EU vs non-EU: after three years, the EU survival rate is about 0.6, compared to 0.5 for non-EU markets—this difference is highly statistically significant.
- Initial export size matters: relationships beginning with larger shipment values (top quartile) survive substantially longer than small-value entries (bottom quartile).
- Variation by product: clothing and oil seeds display higher persistence; electrical equipment and apparatus, and animal or vegetable fats tend to be more short-lived.

### Export Quality and Complexity
- Export quality is assessed using unit-value (UV) ladders at the HS6 level: higher unit values within identical product categories are interpreted as higher-quality or more differentiated varieties.
- Quality assessment also considers the distribution of Moldova’s UV over time, the composition of high-UV products, and the income level of destination markets as a revealed-demand measure of sophistication.
- Sustaining export gains and moving up the value chain will require continued investment in quality upgrading, logistics, and firm productivity.

### Policy-relevant Findings and Implications
- The DCFTA has been a major driver of export gains, especially through deeper penetration (intensive margin) into EU markets.
- EU accession could consolidate DCFTA gains and further boost exports, but effects will differ across sectors; structural and regulatory alignment will be critical.
- Key constraints to accelerating export diversification and sophistication include infrastructure bottlenecks, limited access to finance, and vulnerability to external shocks.
- Policy priorities to support more durable, higher-value exports:
  - Invest in quality upgrading and standards compliance to capture higher UV segments.
  - Improve logistics and reduce trade frictions to lower entry costs and improve survival rates of new relationships.
  - Enhance access to finance for exporters to support scaling and initial market commitment.
  - Pursue deepening structural reforms required for EU integration and to leverage the Reform and Growth Facility resources.
  - Target sector-specific measures recognizing that intensive benefits of EU market access vary across industries.

*Prepared by Rossen Rozenov and Yiran Zha (EUR). Source: IMF staff analysis from "MOLDOVA – EXPORT DEVELOPMENTS AND PROSPECTS".*

### 15.      The UV ladder results show that Moldova’s median position is below that for the EU,

### Export Quality, Export Complexity, and Potential Effects of EU Membership

### Export quality (UV ladder)
- Moldova’s median position fluctuates around 0.3 vs. an EU average of about 0.55.
- Interquartile range and 10–90 percentile band show substantial dispersion, indicating some products perform strongly while others remain clustered in lower-value segments.
- 2014 UV distribution: bi-modal with peaks around the 20th percentile and the 80th percentile, implying two clusters of products (lower- and higher-value varieties).
- 2023 UV distribution: left (low-UV) peak much less pronounced; right peak remains strong. Compression occurs more in the middle, suggesting some products upgraded in relative quantity or new mid-range UV varieties entered.
- Moldova expanded its export basket in the last decade; new exports generally enter at moderate UV levels and do not reach top percentiles immediately.
- Top UV performers are dominated by car seats and components and clothing items, indicating the export-quality frontier is largely anchored in manufacturing linked to European value chains.
- Although Moldova exports significant agricultural products, the high unit-value items are differentiated manufactured goods, implying opportunities for quality upgrading through deeper participation in global production networks and investment in skills, logistics and standards compliance.

### Export destinations and income-level composition
- About 60 percent of total exports go to middle-income countries (World Bank classification).
- The largest share of exports goes to countries with per capita GDP in the range $10-20K (constant 2015 US dollars).
- The noticeable increase in the share of this $10-20K group after 2016 is due to Romania crossing the threshold.
- Expansion in the share of importers with income less than $3K in 2022 is largely due to higher exports to Ukraine.
- Pattern confirms UV-based findings: Moldova exports some high-quality goods, but aggregate export structure remains shaped by mid-range product lines.

### Export complexity and the product space
- Product space framework: products mapped by similarity of capabilities; complex products cluster near dense core; simpler goods on periphery.
- Moldova’s RCA concentrated in peripheral areas of the product space, predominantly in agricultural and textile sectors; these clusters are far from the dense core (chemicals, machinery, electronics).
- Number of HS4-level products with RCA declined from 128 to 122 between 2014 and 2023.
- Decline driven mainly by loss of RCA in agriculture, not offset by sufficient gains in other sectors that could shift Moldova toward more sophisticated capabilities.
- ECI dynamics:
  - ECI declined in the mid-2010s even though model-based projections suggested an increase.
  - Following the entry into force of the DCFTA, the ECI gradually recovered and converged to its projected path.
  - Recent gains reflect growth within existing low-to-mid-complexity categories rather than a shift into more sophisticated products.
- Feasibility of diversification using existing capabilities remains limited; most products near current export mix are low-to-moderate complexity and more sophisticated products lie farther away.
- Compared to Bulgaria and Romania five years prior to EU membership, Moldova has fewer nearby high-complexity diversification opportunities.

### Potential effects of EU membership (gravity model evidence)
- Gravity model estimations use panel data of bilateral exports for former transition economies currently in the EU and accession countries (Western Balkans, Georgia, Moldova), 1996-2023, estimated in 3-year intervals and at 2-digit HS level.
- Three specifications estimated:
  - (i) traditional gravity with exporter-time and importer-time fixed effects;
  - (ii) gravity with paired fixed effects;
  - (iii) model allowing heterogeneous impacts of EU membership across industries.
- Key quantitative results:
  - Traditional gravity equation estimates a very large effect of joining the EU—nearly 45 percent (calculated as 100*(exp(β)-1)).
  - Paired fixed effects estimation associates EU membership with a nearly 16 percent increase in exports.
  - Hagemejer and Mućk (2025) estimate a positive effect from EU enlargement of about 14 percent (study uses value added exports vs. gross exports in this paper).
- Gravity table coefficients (selected):
  - RTA: 0.011 (model 1); 0.117*** (model 2); 0.112*** (model 3).
  - EU: 0.371*** (model 1); 0.148** (model 2).
  - Observations: 706306 (model 1); 817773 (models 2 and 3).
  - Sector-specific interaction coefficients (model 3; interaction with EU indicator):
    - Agri-food (HS1-24) base: EU effect shown as 1.EU -0.055 (not significant).
    - Minerals (HS25-27) #1: -0.203 (not significant).
    - Chemicals & plastics (HS28-40) #1: 0.142 (not significant).
    - Textiles (HS50-63) #1: -0.202**.
    - Metals (HS72-83) #1: 0.205**.
    - Machinery/electrical/transport (HS84-89) #1: 0.378***.
    - Other light manuf (HS41-49, 64-71) #1: 0.035 (not significant).
    - Other (HS90-97) #1: 0.273***.
  - Note: Standard errors reported in parentheses; significance levels *, **, *** correspond to p < 0.10, p < 0.05, p < 0.01.
- Interpretation:
  - EU membership could meaningfully increase exports, with particularly strong effects in machinery, electrical equipment, metals, and other sectors linked to European production networks.
  - Benefits of EU integration are uneven across industries; stronger positive effects in higher-value manufacturing and weaker or negative effects in labor-intensive, low-technology sectors (e.g., textiles).
  - Impact on agricultural and food products is inconclusive (estimated coefficient not statistically significant).

### Conclusions and policy recommendations
- Findings:
  - Deeper trade integration with the EU has benefited Moldova’s exports: DCFTA supported export expansion and regulatory modernization; progress in penetrating EU markets and improving durability of export relationships.
  - Export growth has been uneven; recent shocks exposed vulnerabilities; export basket remains concentrated in a narrow set of products.
  - Quality and complexity improved gradually but Moldova lags peers with more diversified and sophisticated export structures.
  - Moldova’s export model is at an important turning point: longer-lived EU export relationships, improved quality distribution, and growing participation in certain manufacturing segments point to emerging strengths, while product space reveals persistent capability gaps.
  - EU membership could amplify positive trends but will not automatically overcome structural constraints.
- Policy priorities to realize full benefits of deeper integration:
  - Reducing trade costs through investments in transport corridors, border infrastructure, and logistics services to help exporters scale and diversify, particularly SMEs and firms producing time-sensitive goods.
  - Improving access to finance, including through export-oriented credit instruments, to support survival and growth of new export relationships that currently face high early-stage attrition.
  - Upgrading skills, standards, and firm capabilities—especially in manufacturing and agri-food—to bridge capability gaps highlighted by the product space analysis and facilitate movement toward more complex products.
    - Universities can provide relevant technical and analytical skills and support firms through research and development; realization requires close collaboration between corporate sector and academia to ensure effective transfer of knowledge and innovation.
  - Strengthening the investment climate and governance, including regulatory predictability and public administration reforms, to attract foreign direct investment, which anchors higher-value export upgrading and enhances domestic firm capabilities.

*REPUBLIC OF MOLDOVA  INTERNATIONAL MONETARY FUND*

### 29.      The EU Reform and Growth Facility provides a timely opportunity to align public

### 29.      The EU Reform and Growth Facility provides a timely opportunity to align public 

### EU Reform and Growth Facility and export-led growth
- Focused use of concessional EU resources can:
  - Address infrastructure bottlenecks.
  - Expand quality infrastructure.
  - Support firm-level upgrading to crowd in private investment and accelerate integration into European value chains.
- With sustained reform momentum, Moldova can transition from a low- to a mid-complexity export structure and achieve more resilient, export-led growth.

### Methodology and data (Appendix I)
- Data sources used:
  - CEPII BACI (Gaulier, G., Zignago, 2010) and CEPII Gravity (Conte et al., 2022) databases.
  - Atlas of Economic Complexity (Hausmann et al., 2013).
- Data specifics:
  - CEPII BACI annual bilateral trade flows data at the HS6 level (2007 edition) for 2007-2023 used for diversification metrics, survival analysis, and export quality indicators.
  - Gravity model estimated with CEPII BACI bilateral data covering 1996-2023, aggregated to HS2 level, augmented with CEPII Gravity variables: DIST, CONTIG, COMLANG, COMCOL, RTA, EU.
  - Gravity estimation subsample includes former transition economies from Central and Eastern Europe currently EU members and accession countries listed in the source.

### Export concentration and margins
- Export concentration measurement:
  - Normalized Herfindahl-Hirschman indices used for market and product concentration as specified (formulae preserved in source).
- Extensive and intensive margins:
  - Decomposition follows Hummels and Klenow (2005) with definitions of extensive margin (EXE) and intensive margin (IEX) preserved.
  - Country’s export share S is given by S = EXE × IEX.
  - Analysis considers exports to the rest of the world, the EU, and non-EU countries.

### Survival analysis
- Survival analysis framework:
  - Follows Cox and Oakes (1984) and Kleinbaum and Klein (2012).
  - Event defined as the end of an export relationship; right censoring acknowledged.
  - Survivor function S(t) = Pr(T > t).
  - Kaplan-Meier (KM) estimator applied (formula preserved from source).

### Gravity model specification
- Estimated models (as presented in the source):
  - Model (1), Model (2), Model (3) with exporter-time, importer-time, exporter-importer fixed effects, and controls including lnDIST, CONTIG, COMLANG, COMCOL, RTAij,t, EUij,t, and sector indicators Dkg (formulas preserved).
- Variable definitions preserved: lnXij,t is log of exports from country i to country j at time t; πi,t, χi,t, μiij are fixed effects; Dkg is indicator for aggregated sector k.

### Moldova’s inflation targeting (IT) regime — context and evolution
- Institutional setup:
  - NBM adopted formal IT in late 2012; operationalized in 2013.
  - Inflation target range: 5+/- 1.5 percent.
  - NBM began publishing quarterly Inflation Reports and introduced forecast-based decision-making supported by a macroeconomic model and scenario analysis.
- Inflation dynamics and pass-through:
  - Inflation has been volatile since end-2014 due to high share of exogenously determined goods (food and fuels) and regulated prices (gas and electricity) affected by Russian gas supply and pricing up to 2025.
  - Large share of imported goods in CPI implies sizable pass-through from exchange rate to CPI.
  - Median pass-through to CPI for MDL-US dollar exchange rate (exchange-rate-driven changes): 45 percent.
  - Pass-through elasticity for MDL-euro exchange rate (exchange-rate-driven changes): 53 percent.
  - For exchange rate changes driven by domestic monetary policy shocks: pass-through from an exchange rate change to the CPI is 62 percent for both MDL-US dollar and MDL-euro rates.
  - Interpretation: around half of an exchange rate movement transmits to the CPI for exchange-rate-driven shocks.

### NBM monetary instruments and challenges
- Complementary instruments to the base rate include reserve requirements (RRs), open market operations, and standing facilities.
- Two typical IT challenges highlighted:
  - Fear of floating due to liability dollarization or high pass-through can lead to de facto exchange rate targeting and heavy use of FX interventions (FXI).
  - Shallow financial markets and excess liquidity (post-2014 bank fraud and emergency lending) led the NBM to increase RRs to exceptionally high levels and issue short-term certificates of deposit (CDs).

### Evolving monetary policy operations and empirical findings
- Shift in base rate behavior:
  - Pre-2020: Base rate did not react significantly to the inflation gap; instead, base rate increased by 7 basis points to a quarterly depreciation in the leu-dollar exchange rate of 1 percent (significant at the 1 percent level).
  - Since 2020: Base rate displays a significant reaction to the inflation gap; exchange rate coefficient smaller and significant only at the 10 percent level.
- Interest-rate transmission:
  - Since 2021, deposit rate rises 0.19 percentage points on impact after a one-percentage-point increase in the base rate; pass-through complete after about 4 months with a total increase of 0.55 percentage points.
  - Lending rate: no immediate reaction on impact; after 3 months it settles at 0.3 percentage points higher.
  - T-bill rates react quickly and strongly: since 2021, 71 percent of base rate hikes were transmitted to T-bills within a few months.
  - Example: Energy price hike in January 2025 — base rate increase of 290 bps was followed by increase of bank loan rates of 110 bps until July 2025; lending pass-through somewhat higher at 0.38 in that episode.
- FX interventions (FXI):
  - Use of FXI declined substantially post-2020.
  - Pre-2020: one-percent month-over-month depreciation of leu against dollar associated with FX sales of 0.09 percent of annual GDP.
  - Post-2020: reaction coefficient not significant; interventions smaller relative to GDP and less biased toward dollar sales; interventions have become smaller relative to FX interbank transactions.
  - Degree of exchange rate management (ratio of FXI volatility over FXI + exchange rate volatility) has decreased, indicating reduced focus on preventing depreciations.

### Empirical coefficients (Taylor rule subsamples as reported)
- Taylor Rule Coefficients: Subsamples
  - 2013Q1-2025Q2
    - Inflation gap: 0.1-0.03
    - Output gap: 0-0.15-0.03
    - Change in exchange rate: 0.07***
  - 2013Q1-2019Q4
    - Inflation gap: 0.1-0.03
    - Output gap: 0-0.15-0.03
    - Change in exchange rate: 0.07***
  - 2020Q1-2025Q2
    - Inflation gap: 0.54***
    - Output gap: -0.03
    - Change in exchange rate: 0.04*
  - Notes preserved: *0.05<p≤0.1, ** 0.01<p≤0.05, *** p≤0.01.

### Policy-relevant implications (drawn from findings in the source)
- Continued focus on strengthening the IT framework by prioritizing inflation while limiting systematic exchange-rate targeting.
- Use of complementary instruments (RRs, open market operations, standing facilities) appropriate to manage liquidity and reinforce base rate control, while avoiding excessive reliance on FXI to counteract trend exchange rate movements.
- Structural reforms and use of concessional EU resources to relieve infrastructure bottlenecks and support firm upgrading to foster export diversification and resilience.

*Republic of Moldova — International Monetary Fund (content as provided in the source).*

### 7.      The intense use of high RRs, while appropriate during times of excess liquidity, is

### 7. The intense use of high RRs, while appropriate during times of excess liquidity, is

### Distortions from high reserve requirements (RRs)
- NBM actions and RR history:
  - NBM raised the reserve requirement ratio on deposits in lei from 14 to 35 percent between February and November 2015 and then to 42.5 percent in October 2018.
  - In December 2025, the NBM reduced the RR ratio to 20 percent.
- Economic frictions and costs:
  - Required reserves must be held as central bank reserves which are only partially remunerated; the amount of required reserves equal to the equivalent of five percent of the reserve base is not remunerated while the remaining amount is remunerated at the overnight deposit rate, which is currently two percentage points below the base rate.
  - The requirement to hold reserves creates an inefficient wedge between lending and deposit rates, raising the spread between the two.
  - The large increases in the RR ratio over short periods of time in 2015 and 2022 had visible effects with a delay of a few months.
  - Measured distortion: In December 2025, as the NBM reduced the RR ratio to 20 percent, the cost still stood at about MDL 487 mln, which was still sizable at roughly 1.6 percent of banks’ equity.

### Model simulations: effects of lowering RRs
- Experiment parameters and assumptions:
  - Calibrated an extended version of the quantitative IPF model (Adrian et al, 2021) to Moldova.
  - Considered a 10 percent reduction in the RR from 22 to 12 percent as an indicative experiment.
  - Reduction assumed perfectly credible and perfectly anticipated by all agents.
  - Adjustment is gradual: RR adjusts fully within four years, with the half-way point reached after 18 months.
- Simulated banking and macro responses:
  - Near-term banking response consistent with reduced leverage as reserves are shed from balance sheets easing financial frictions.
  - Nominal loan rates and spreads fall on announcement, while the stocks of loans and deposits rise.
  - Higher loan volumes drive profitability; banks’ market-to-book ratios undergo a sustained rise.
  - Overall macroeconomic effects are small in the model; new credit is directed to households, boosting asset prices and consumption in the short term.
  - Model does not capture likely benefits from boosting business lending (e.g., more capital deepening and improvements in labor productivity).

### NBM liquidity instruments: CDs and repos and interbank implications
- Instrument features and use:
  - CDs: base rate applied; maturities generally 14 days.
  - Repo rate: set at +0.25 pp above the base rate.
- Empirical patterns:
  - CDs used intensely over the past decade as banks shed excess liquidity beyond what was withdrawn through RRs.
  - As a share of GDP, CDs peaked at just below 5 percent in March 2018 and again in July 2023.
  - By December 2025 CDs had declined to 1.4 percent of GDP.
  - Repos generally used less, but saw a pickup in late 2025.
  - Observed substitution: repos pickup coincided with sales of CDs, implying NBM provided liquidity to some banks while absorbing liquidity from others; banks transacted with the NBM rather than with each other on the interbank market.

### Policy discussion and recommendations
- Monetary policy practice and credibility:
  - Greater focus on the inflation gap in setting the base rate signals a determination by the NBM to keep inflation within its target band and bolsters credibility and accountability.
  - Credibility will further grow with an increasing track record of following IT principles.
- Foreign exchange intervention (FXI) practices:
  - More limited response of the base rate and FXI to exchange rate movements has allowed exchange rate to reflect market forces more.
  - 2019 rules: FXI should address FX market disruptions but not fundamental exchange rate changes; leave discretion to react to turbulence.
  - IPF “use cases” for FXI relevant for Moldova:
    - i) address destabilizing premia from arbitrage frictions in shallow FX markets;
    - ii) counter financial stability risks from FX mismatches;
    - iii) support price stability when exchange rate changes risk de-anchoring inflation expectations.
  - Liability dollarization remains sizable: 20-percent share of FX loans in total loans.
  - Recommended actions to avoid unanchoring inflation expectations:
    - Continue building a track record of effectively fighting inflation.
    - Follow recommendations in the Central Bank Transparency Code Review for clearer communication of deliberations and decisions.
    - Strengthen NBM’s political, institutional, and financial independence.
- RR policy guidance:
  - NBM should reduce the level of RRs further once inflationary pressures have subsided.
  - Rationale: strong lending growth since early-2024 increased the credit to GDP ratio from 21 to 29 percent, reducing excess liquidity substantially.
  - Credit growth slowed towards the end of 2025 but is projected to continue in the coming year, and over time to further reduce excess liquidity.
  - Pursue a carefully calibrated reduction of RRs while monitoring inflation developments.
- Benefits of lower RRs for markets:
  - Lower RRs would support government debt market development and the interbank market.
  - Repos could be used increasingly for open market operations to manage aggregate rather than bank specific liquidity, with government debt as collateral.
  - Higher demand would raise valuations of government debt, boosting the sovereign bond market.
  - Scarcity of liquidity and a better functioning sovereign bond market should help revive the interbank market, supported by ending the fixed-rate-full-allotment and improving liquidity management and forecasting by banks and the NBM.
- Need for careful calibration:
  - A plan for a reduction path must consider how released reserves are used by banks.
  - A slow release over time would avoid a possibly undesirable credit boom; remaining excess liquidity could be absorbed by issuance of CDs.
  - Speed of reducing RRs might vary with economic conditions (booms vs. recessions).
  - Endpoint calibration: among regional peers local currency RRs are currently in the range of 5.5 and 10 percent; optimal level for Moldova should consider banks’ specific liquidity needs, shocks, and degree of interbank market development.

### Costs of reserve requirements: unit and total cost (methodology and numbers)
- Remuneration structure and unit cost expression:
  - NBM pays interest at the overnight deposit rate (ONDR) on the share of required reserves that exceeds 5 percent of the reserve base (RB); the first five percent of the reserve base are not remunerated.
  - Unit cost formula (as given in the source):
    - [5%*RB*Base rate+(RR ratio-5%)*RB*(Base rate - ONDR)]/(RR ratio*RB)
    - which simplifies to:
      - (5%/RR ratio)*base rate+[(RR ratio-5%)/RR ratio]*(base rate - ONDR)
  - As the gap between the base rate and the ONDR changes rarely, the unit cost varies mostly with the base rate and the RR ratio.
- Example calculation and values:
  - At the required reserve ratio of 20 percent of the reserve base and a base rate of 5 percent in December 2025, the unit cost was 2.8 percent.
  - Total cost in December 2025: about MDL 487 mln (roughly 1.6 percent of banks’ equity).

### Model framework and transmission mechanisms (methodology highlights)
- Model used:
  - Framework adapted from a quantitative model of the integrated policy framework (QIPF, Adrian et al. 2021): two-country open economy New Keynesian model with financial intermediaries added in the extended model.
- Key modeling features:
  - Firms produce for domestic and foreign consumption with sticky prices in local currency.
  - Households are split into ‘saver’ and ‘borrower’ types; borrowers face LTV-based borrowing limits tied to housing collateral; housing market affects credit demand and consumption.
  - Banks face leverage/incentive constraints due to moral hazard (depositors fear private benefits extraction by bankers); loan supply proportional to net worth.
  - Reserves occupy balance sheet “space”: holding reserves raises banks’ overall leverage (loans plus reserves to net worth), tightening the incentive constraint, raising spreads, and reducing lending capacity.
  - Creditor perceptions: reserves considered better collateral than loans; tightness of constraint depends on mix of loans and reserves and overall leverage.

*Italic: Source: Excerpt from IMF PDF chapter on Moldova monetary policy and reserve requirements (provided content).*

---


_Source: https://www.imf.org/-/media/files/publications/cr/2026/english/1mdaea2026002-source-pdf.pdf_