## 1. Remittances and Monetary Transmission ## Source details **Canonical URL:** [1. Remittances and Monetary Transmission](https://www.imf.org/-/media/files/publications/cr/2022/english/1mdaea2022003.pdf) ## Other formats - [Markdown version](/-/media/files/publications/cr/2022/english/1mdaea2022003.pdf.md) - [Structured JSON version](/-/media/files/publications/cr/2022/english/1mdaea2022003.pdf.json) --- ### Monetary policy framework and main instruments - National Bank of Moldova (NBM) adopted inflation targeting (IT) in 2013 after a three-year transition period. - NBM’s first objective: keep inflation at 5 percent with a variability range of ±1.5 percentage points, at a horizon of 18–24 months. - NBM’s secondary objective: promote growth and employment. - Policy rate: official main instrument; transmitted through open market operations in the market for NBM certificates. - Interest rates on standing facilities establish a symmetric corridor at ±3 percentage points around the policy rate. - Other tools influencing monetary conditions: reserve requirements (RRs) and foreign exchange interventions (FXIs). - Since 2020, FXIs are guided by a formal FXI strategy that specifies intervention criteria indicating disorderly market conditions and/or excess exchange rate volatility. ### Key quantitative and structural constraints in Moldova - Supply shock vulnerability: - Share of food in CPI in 2020: almost 43 percent. - Large external inflows reduce scope for stabilization: - Remittances inflows: around 16 percent of GDP. - Foreign aid: 2 percent of GDP. - Dollarization and trade openness: - Loan dollarization: 33 percent. - Deposit dollarization: 41 percent. - Import share: 55 percent of GDP. - Exchange rate management and FXIs: - Frequent NBM interventions in the FX market; benchmarking in 2020 shows a moderate footprint, but comparison of volumes vs interbank transactions indicates NBM’s footprint was large in most months. - Financial account restrictiveness: - Financial Account Restrictiveness Index (FARI) based on the 2018 AREAER: inflows-side restrictiveness in line with peers but substantially higher than in Advanced Economies; outflows-side deemed very restrictive even compared to peer countries. - Shallow interbank and securities markets: - Interbank market in local currency largely inactive and completely dried up after the 2014 banking crisis. - Primary and secondary T-Bill markets increased over the last two years but remain small compared to peers. - Weak financial intermediation and lending: - Middle-range bank capitalization among peers, but lending indicators are particularly low. - High remittances inflows tend to increase deposits that are often not matched by equivalent growth in credit. - Bank concentration is at the higher end among peers. - Excess liquidity legacy of the 2014 banking fraud: - NBM injected sizeable liquidity after the 2014 banking fraud; excess liquidity has only partially been mopped up by increases in RRs. - Excess liquidity impairs interbank lending and pass-through of policy rate changes to bank rates and loan supply. - Growth of externally funded non-bank credit organizations: - Regulatory arbitrage after the banking crisis led to growth of non-bank credit organizations funded predominantly from abroad; domestic monetary policy can only marginally influence non-banks’ lending conditions. ### Channels through which remittances affect monetary transmission (Box 1) - Dutch-Disease-like effects: - Persistent remittance inflows can appreciate the real exchange rate (REER), weakening tradable sector competitiveness and possibly driving frequent, non-coordinated FX purchases. - Remittances allow higher spending on tradable and non-tradable goods; only non-tradable prices increase, expanding non-tradable supply and shifting resources away from tradables. - Institutional and financial market effects: - High remittance inflows associated with weaker institutional environment, undermining financial market development. - Remittances provide a stable inflow not sensitive to interest rates, narrowing the scope of monetary policy. - Remittances increase bank deposits but are often not matched one-to-one by credit increases, generating excessive liquidity and reducing interbank market activity. - Moldovan evidence: - Deposit-to-GDP ratio increased in the early 2000s and remained relatively stable since then; loan indicators have been going down since the early 2010s. - The 2014 banking fraud was responsible for a large part of the decrease in credit to the private sector; high remittances inflows might have contributed as well. ### Empirical analysis of monetary transmission (overview) - Empirical approach: VAR to identify monetary policy shocks and estimate inflation response; sample period 2010–2020 split into two subperiods to account for structural break after the banking fraud became public. - Overall finding: - Effectiveness of transmission of policy rate changes to inflation weakened after the banking fraud. - First subperiod (2010–14): in response to an increase in the policy rate by one percentage point, prices decreased by around 0.4 percent after three quarters (statistically insignificant). - After the banking fraud: peak response weakened to a decrease of 0.2 percent after five quarters. ### Interest rate channel — causation and pass-through estimates - Granger causality: policy rate affects money market, T-Bill, deposit, and lending rates, but not vice versa. Money market rates affect lending and deposit rates, but not vice versa. - ADL regression example (2010–2014, change in policy rate = 100 basis points): - Deposit rate: short-term change of 6 basis points in the following month, long-term change of 4 basis points. - Lending rate: short-term pass-through of 24 basis points. - Money market rate: short-term pass-through of 48 basis points. - T-Bill rate: short-term pass-through of 60 basis points. - Data limitation: Money market rate data available only for 2010M1 to 2015M4. - Evolution: - After 2014, short-term pass-through from policy rate changes to market interest rates remained low. - Long-term transmission improved (possibly linked to slowly deepening financial markets). - Results for lending and deposit rates are statistically insignificant and very sensitive to sample period. - Possible cause of low short-term pass-through: excess liquidity injected after 2014 reducing interbank activity and banks’ incentives to immediately pass on interest rate changes. ### Credit (bank lending) channel — findings - Identification uses bank-level lending and liquidity data and interaction between policy rate changes and a liquidity indicator. - Findings: - First subperiod: evidence of an operative bank lending channel (positive sum of coefficients on the interaction term), driven mainly by small banks (likely liquidity constrained). - Second subperiod: no evidence for an active bank lending channel (negative sum of the interaction term coefficients). - Likely drivers of breakdown: injection of liquidity and efforts to clean up the banking sector after the banking fraud. ### Exchange rate channel, FX interventions (FXIs), and their role in transmission - Aggregate VAR finding: Transmission via the exchange rate channel is weak, reflecting restricted capital mobility. - Subperiod differences: - First subperiod: exchange rate channel was active but with only a small impact. - After 2015: exchange rate channel effectively undermined transmission of the policy rate; CPI response would have been stronger without the exchange rate channel. - Degree of exchange rate management: - Quarterly frequency: exchange rate management moderate and aligned with IT emerging economies. - Daily frequency: exchange rate management much stronger, suggesting active high-frequency FXIs that may over-manage lower-frequency movements. - Drivers of FXIs: exchange rate volatility is the dominant driver; deviations from medium-term exchange rate level play only a minor role. - Empirical likelihood of interventions (average over whole sample, weekly basis): - likelihood of purchases: about 40 percent, - likelihood of no intervention: 45 percent, - likelihood of sales: 15 percent. - In crisis times, likelihood of FX sales increases. - Asymmetry and effectiveness: - NBM purchases much more often than it sells; average purchase volume is smaller than average sale volume. - Event-window analysis: - FX purchases happen in weeks after the exchange rate appreciates relative to average by about 0.3 percent; after purchase the difference comes down to not different from zero. - FX sales occur in weeks after the exchange rate depreciates relative to average by about 0.8 percent; after sale the difference comes down to not different from zero. - Interpretation: FXIs are effective in influencing the exchange rate in the very short term; one-sided, repeated interventions can have more permanent effects. - Coordination with interest rate policy: - FXIs are not always aligned with interest rate policy and can counter it, weakening monetary transmission. - Example episodes: - Crisis years 2015/16: policy rate increase accompanied by large FX sales, which stemmed depreciation and fostered tightening. - 2017/18: NBM increased policy rates while purchasing FX, preventing appreciation and impairing tightening effects on inflation. - Mechanisms of harm: - One-sided FX purchases can prevent exchange rate appreciation when policy rates are hiked. - FX purchases increase liquidity (if only partially sterilized), potentially adding inflationary pressure. - Conflicting signals from FX purchases during tightening can blur messaging and undermine IT credibility. - Strategy and practice since 2020: - 2020 FXI strategy adopted to foster two-way exchange-rate flexibility; specifies intervention criteria for disorderly conditions and excess volatility. - Daily intervention data suggests the strategy was broadly followed: most interventions occurred on days when at least one criterion was satisfied; some interventions happened on days without criteria but with very small volumes (reported as preemptive). - The more intervention criteria fulfilled, the larger the average intervention size. ### International experience on FXIs in IT frameworks (selected lessons) - FXIs can coexist with IT if narrowly defined and targeted, but can entail costs (hindering market development, undermining credibility). - Peru example: - Dollarization: 26 and 35 percent of loan and deposit dollarization, respectively, at the beginning of 2021; over 60 and 70 percent, respectively, at the beginning of the 1990s. - Sterilized FXIs helped guard against exchange-rate-related financial risk and reduced dollarization over time, but interventions constrained financial development (e.g., low credit-to-GDP ratio, limited FX hedging instruments). - Ukraine example: - Adopted IT in 2016; FXI strategy implemented from the start with objectives including accumulating reserves and smoothing excess volatility. - Research caveat: FXIs warranted primarily when currency mismatches and shallow FX markets exist; FXIs can encourage moral hazard, disincentivize hedging, and risk undermining credibility of the policy rate in IT frameworks. ### Policy recommendations (monetary policy and FXI management) - Strengthen IT framework credibility and policy consistency: - Policy rate should always be NBM’s primary instrument to achieve its inflation target. - Additional instruments (RRs for liquidity management, FXIs for disorderly FX conditions) must be coordinated with the policy rate to avoid undermining IT credibility and effectiveness. - Improve public communication on outlook, objectives, and instruments to help anchor inflation expectations and improve MTM. - Gradually reduce NBM’s presence in the FX market to facilitate external adjustment and improve exchange rate channel transmission: - Rationale: persistent and one-sided FXIs can prevent exchange rate appreciation when policy rates are hiked; FX purchases increase liquidity and can add upward pressure on inflation if only partially sterilized; conflicting signals from FX purchases during tightening can blur messaging and undermine IT credibility. - Continue evaluating and possibly improving parameterization of the 2020 FXI strategy to gradually improve exchange rate flexibility. - Communicate the main objectives of the strategy to banks so they can make informed risk-pricing decisions. - Foster FX market development: - Improve pricing of exchange rate risks and develop hedging instruments. - Review FX market liberalization policies to identify and eliminate obstacles given a still less-open capital account relative to peers. - National de-dollarization strategy: - Implement a national de-dollarization strategy to limit balance sheet exposure to exchange rate shocks. ### Facilitating financial market development and reducing excess liquidity — objectives and priorities - Key objectives: - Foster transmission via the interest rate and credit channels. - Foster lending to households and small and medium enterprises without compromising financial stability. - Safeguard progress in reforming the banking sector and continue progress in financial sector reforms. - Policy recommendations and priorities: - Strengthen regulation and supervision of non-bank financial institutions (NBFIs) to eliminate regulatory arbitrage and ensure financial stability while addressing risks and increasing consumer protection. - Leverage forthcoming transfer of supervision from NCFM to NBM to use NBM’s expertise in supervising NBFIs. - Adopt a strategy to foster financial inclusion and capital market development to broaden the financial sector’s customer base and widen NBM’s scope of action. - As credit to the economy grows and excess liquidity is reduced, normalize reserve requirements (RRs) to reduce banks’ funding costs. - Move away from passive full-allotment system towards proactive liquidity forecasting and management when excess liquidity is reduced. - Strengthen budgetary liquidity and debt management procedures and promote development of domestic securities. ### Public enterprises, governance, and macroeconomic vulnerabilities — key findings - Public enterprise sector size and concentration: - Over 900 state-owned companies. - Analysis covers 326 public enterprises for which data are available (roughly one-third). - Combined assets valued at about 21 percent of GDP (US$2.5 billion), concentrated in public works, energy, telecom, and transportation. - Central-level enterprises: in 90 percent the controlling share of the state was 100 percent; only 44 companies had state participation below 50 percent. - Central-level companies hired almost 40 thousand employees (as of 2018)—about 4 percent of total national employment. - Largest five enterprises among the 326 had assets and outstanding liabilities constituting roughly 6–7 percent of total assets and total liabilities of the sample. - Quantified corporate vulnerabilities (as of end-2018 and 2019 data): - Total debt outstanding on companies with available data amounted to 7.3 percent of GDP; arrears amounted to 3.3 percent of GDP (end-2018). - Total net worth improved in 2019 driven by growth in equity in most central government sectors; net worth as a share of GDP dropped slightly from 2018 to 2019. - Net worth grew 5 percent annually (aggregate measure cited). - One-fifth of all companies suffered negative equity; majority at local level. - Total subsidies extended to only four central-level public enterprises reached 1.5 percent of GDP; the overwhelming amount granted to one state roads administration. That subsidy exceeded the enterprise’s operating revenue by 20 percent. - Financial performance metrics (2019, 326 companies): - Profitability: 195 companies incurred net losses; ROA averaged 5 percent overall (5 percent central level, -2 percent sub-national); ROE central level 1.3 percent, sub-national -10 percent. - Cost recovery: mildly above “1” in 10 sectors; six sectors significantly below 1 (agriculture and water). - Labor costs: one-third of enterprises pay salaries >75 percent of total expenses; 47 enterprises pay salaries >90 percent of total expenses; labor costs absorb almost 15 percent on average of total enterprise revenue; 14 companies have salary costs that exceed revenue. - Liquidity: current ratios above “1” on average except four sectors; debtor turnover averaged 2 months; creditor turnover over 4 months on average; 46 enterprises recorded debtor or creditor turnover exceeding 1,000 days. - Solvency: 9 companies had debt exceeding assets; 2 companies had debts three-to-four times the size of assets; debt-to-EBITDA extremely high in five sectors (20–50 times). - Overall risk classification for 326 enterprises: - Overall financial risk: “High” at central and sub-national levels. - Profitability risk: “Moderate” overall, “High” at sub-national level; 200 enterprises faced “high” or “very high” profitability risks. - Liquidity risk: “High” overall, “Very high” at sub-national level; 102 enterprises with “high” or “very high” liquidity risks. - Solvency risk: “High” overall; 128 enterprises with “high” or “very high” solvency risks. ### Scale and concentration of risks; stress testing of largest enterprises - 250 companies face combined or multiple financial risks (over three-quarters of the sample). - 44 companies form a “core” group with “high” or “very high” risks across profitability, liquidity, and solvency: - 14 central-level companies (including 3 in energy). - 30 sub-national companies (including a large water and sanitation company). - Core group’s liabilities: over 13 percent of the 326 companies’ liabilities, or 1 percent of GDP. - Stress testing of 15 largest-liability companies (2021 scenarios, IMF FAD tool): - Growth shock: assumes 1.5 percentage points weaker growth recovery in 2021 relative to baseline and a weaker leu (10 percent). - Liquidity shock (in addition): 30 percent share of receivables potentially materializing as a liquidity constraint. - Stress test macro-fiscal impact (worst-case where government assumes liability burden of 15 largest enterprises): - Adds about 8 percentage points of GDP to public debt in the year of the shock (2021). - Adds 16 percentage points of GDP to public debt over the next 5 years. - Additional burden of absorbing economic losses estimated at 0.7 percent of GDP in 2021 and 1.4 percent of GDP cumulatively over the medium term. - Estimates exclude other forgone financial flows (taxes, dividends, tariff collections) and are described as conservative. ### Case studies (selected) - State Roads Administration: - Manages state assets worth almost 6 percent of GDP, employs about 170 staff, receives government subsidies amounting to 1.4 percent of GDP, and reported net income of MDL 0.8 million in 2019 (inclusive of subsidies). - Removing state support reveals severe operational weaknesses and imminent liquidity and solvency risks; tapping reserves would deplete about 40 percent of equity. - National Railway Company: - Manages state assets worth 1.6 percent of GDP, employs around 7,600 staff, received on-lent external financing. - 2019: sales revenues increased but cumulative net loss for a second consecutive year and reduction in equity; ROA positive, ROE negative; liquidity moderate; solvency very low insolvency risk due to subdued debt relative to assets and equity. - March 2021: salary arrears and voluntary leave arrangements reported. ### Reform roadmap and governance recommendations (state and PPA level) - State-wide (MoF, line ministries): - Adopt ownership rationale and state-ownership policy specifying objectives, criteria for establishment/termination, classification, legal forms, commercial vs non-commercial mandates. - Strengthen MoF’s fiscal risk oversight; expand Fiscal Risk Statement (FRS) to cover at least the 44 “core” companies and other systemically important enterprises. - Improve coverage, monitoring, coordination, and reporting of contingent liabilities. - Clarify state aid rules and intervention criteria; reduce fragmentation of roles and responsibilities across MoF, line ministries, legislature, and other bodies. - Strengthen oversight, accountability, and capacity of the PPA. - Create level playing field between public enterprises and private firms; review laws on privileged access to resources, public procurement transparency, and anti-monopoly legislation. - At the PPA level: - Comprehensive review of institutional arrangements and legal/regulatory frameworks to root out corruption, clarify remuneration and dividend policies, safeguard board independence, and strengthen operational efficiency and incentives. - Require forward-looking KPIs and limits on liabilities; enforce accounting standards and robust data/record keeping. - Hire and retain BoD and BoA members via transparent, merit-based processes. - Strengthen annual consolidated reporting to include employment, ownership changes, reform synopses, and disclosure of mismanagement and remedies. - Adopt strategy to strengthen accountability over public funds and improve asset management. ### Sequencing, political prerequisites, and data needs - Proper sequencing, political will, and ownership of the reform process are key to success. - A comprehensive assessment of financial positions of all SOEs at all government levels is a first step to identify enterprises that could generate fiscal costs. - Results should inform a time-bound state-ownership strategy identifying enterprises for restructuring, privatization, or liquidation. - Additional recommendations: - Update assessment with 2020 data to gauge COVID-19 impact. - Strengthen disclosure of missing data (employment, minority stakes, cross-ownership, intra-company debts). - Undertake a comprehensive Public Sector Balance Sheet (PSBS) in future to assess public wealth and fiscal solvency. ### Methodological and risk-threshold definitions used in benchmarking - Key ratios and thresholds used (examples preserved): - Return on assets = EBITDA / assets - Return on equity = net profit (loss) after tax / equity - Cost recovery = total revenue / (cost of goods sold + other expenses) - Current ratio = current assets / current liabilities - Quick ratio = (current assets - inventory) / current liabilities - Debtor turnover (days) = trade receivables*365 / total revenue - Creditor turnover (days) = trade payables*365 / cost of goods sold - Debt to assets = liabilities / assets - Debt to equity = liabilities / equity - Debt to EBITDA = liabilities / EBITDA - Selected risk thresholds (as assumed): - Profitability — ROA: Very Low Risk > 8.00; Low Risk 4.00; Moderate Risk 0.00; High Risk -5.00 - Liquidity — Current Ratio: Very Low Risk > 2.00; Low Risk 1.50; Moderate Risk 1.25; High Risk 1.00 - Solvency — Debt to Assets: Very Low Risk < 0.25; Low Risk 0.50; Moderate Risk 0.75; High Risk 1.00 - (Full list of thresholds preserved in the source.) ### Non-Bank Credit Organizations (NBCOs), financial inclusion, and systemic risks - NBCO overview: - NBCOs became an important source of financing after the 2014 banking fraud; planned transfer of NBFI supervision from NCFM to NBM expected by 2023. - Access to credit challenges: - Commercial bank credit as share of GDP around 22 percent of GDP — lowest among regional peers. - Commercial bank lending share fell from above 30 percent to below 15 percent following the 2014 fraud. - NBCO market trends: - NBCO lending grew by average around 20 percent annually over the last decade; accelerated in 2019 then slowed due to COVID-19. - NBCO share of credit market rose from around 7 percent in 2010 to almost one-fifth at end-2020. - End-2019: 570,000 individuals and firms had a credit relationship with an NBCO (about 20 percent of citizens). - NBCO consumer lending share rose from under 2 percent in 2010 to around one-third in 2020. - NBCO structure and concentration: - Almost 200 NBCOs on record; 144 NBCOs have equity < MDL 10 million (USD 569,000); 63 have < MDL 1 million (USD 56,900); 12 have negative equity. - Fewer than 40 NBCOs have assets > MDL 30 million; many undercapitalized. - Top 30 NBCOs account for ~90 percent of market; 40 NBCOs had zero loan assets at end-2020Q1. - Profitability and business model: - Banks’ ROA and ROE (2017–19): around 2.5 percent and 14.6 percent. - NBCOs’ ROA and ROE (2017–19): 9.3 percent and 27.8 percent. - NBCO loan interest rates averaged almost four times the average bank loan rate as of early 2021. - NBCOs charge high fees and penalties; much funding came from non-residents. - Funding and vulnerabilities: - Borrowing ~50 percent of NBCO assets over last ten years; borrowing rose from MDL 832 million in 2010 to almost MDL 6.1 billion at end-2020. - Non-resident funding averaged ~60 percent of total; banks ~24 percent. - NBCOs cannot accept deposits from the general population (law effective early 2020). - Capital requirement increased from MDL 100,000 to MDL 1,000,000 from 2021; active NBCOs reduced to 144 by end-2020Q1. - Consumer protection and overindebtedness risks: - Reported abusive practices and low financial literacy among targeted consumers. - NBCOs not required to consider credit bureau information; no overall debt-to-income limits—risk of overindebtedness. - Supervisory fragmentation and contagion risks: - Fragmented supervision limits systemic visibility; Financial Stability Council coordination required but has implementation delays. - Concentrated bank exposure to NBCOs: in three banks, lending to NBCOs constitutes 17.6, 14.9, and 10.4 percent of their regulatory capital (end-2020Q1). - 93 percent of bank lending to NBCOs goes to the top 14 NBCOs by assets. - Asset quality and buffers: - NBCO asset quality deteriorated in 2020: loans 31–90 days past due share increased by >10 percentage points since March 2020 through end-2021Q3. - Total buffers (equity + provisions) around half of outstanding loans at aggregate level. - Stress-test summary (Box 2 — NBCOs: A Stress Test): - Definition: “bad loans” = loans delinquent >90 days (require 100 percent provisioning). - Focus on 15 and 30 largest NBCOs (80 and 90 percent of market share respectively). - Downside scenarios: - Scenario 1: 25 percent of loans 60–90 days delinquent turn into “bad loans”. - Scenario 2: 50 percent convert. - Scenario 3: 75 percent convert. - Scenario 4: “Bad loans” double. - Scenario 5: “Bad loans” triple. - Findings: - Aggregate NBCOs have sufficient equity to absorb deterioration except in extreme scenarios (3 and 5). - Impact concentrated in NBCOs with high lower-quality portfolios or limited equity. - In extreme scenarios a few NBCOs would have equity wiped out, likely accounting for ~6 percent of market share. - Policy considerations for transition of oversight to NBM: - Regulate NBCOs distinctly from banks to preserve credit provision to underserved customers. - Avoid imposing identical strict credit approval/reporting requirements as banks. - Tailor capital and provisioning requirements to NBCO risks. - Align macroprudential and systemic risk requirements between banks and NBCOs. - Address AML/CFT concerns by analyzing UBOs and sources of foreign funding. - Prioritize on-site supervision post-pandemic with emphasis on governance and risk management. - Consolidation and inclusion: - Consider consolidation to 30–40 large players to increase resilience and simplify supervision while maintaining competition. - Consolidation to top 30 would raise HHI from ~9 percent to ~11 percent (still below 25 percent US threshold cited). *Source: IMF staff analysis as presented in the chapter.* ### 1. Remittances and Monetary Transmission ________________________________________________ 9 ### 1. Remittances and Monetary Transmission ### A. Monetary policy framework and main instruments - The National Bank of Moldova (NBM) formally adopted IT in 2013 after a three-year transition period. - NBM’s first objective: keep inflation at 5 percent with a variability range of ±1.5 percentage points, at a horizon of 18–24 months. - NBM’s secondary objective: promote growth and employment. - Policy rate: official main instrument; transmitted through open market operations in the market for NBM certificates. - Interest rates on standing facilities establish a symmetric corridor at ±3 percentage points around the policy rate. - Other policy tools influencing monetary conditions: reserve requirements (RRs) and foreign exchange interventions (FXIs). - Since 2020, FXIs are guided by a formal FXI strategy that specifies intervention criteria indicating disorderly market conditions and/or excess exchange rate volatility. ### B. Performance of IT and perceived transmission effectiveness - After IT introduction, both average inflation and inflation volatility fell substantially and are in line with a group of peer countries. - Authorities expressed concerns about the effectiveness of the policy rate in steering inflation; the paper seeks to identify frictions and propose policy recommendations. ### C. Frictions and impediments to monetary transmission in LLMICs (and relevance for Moldova) - General channels available for transmission in LLMICs: interest rate channel, credit channel, exchange rate channel (asset and expectations channels often weak). - Conditions required: - Interest rate channel: pass-through of policy rate changes to market rates relying on deep securities and interbank markets. - Credit channel: well-regulated and competitive financial sector with high-quality and transparent lending opportunities. - Exchange rate channel: open capital account and a freely floating exchange rate. - Empirical evidence points to several frictions likely to disrupt MTM in Moldova. ### D. Key quantitative and structural constraints in Moldova - Supply shock vulnerability: - Share of food in CPI in 2020: almost 43 percent. - Large external inflows reduce scope for stabilization: - Remittances inflows: around 16 percent of GDP. - Foreign aid: 2 percent of GDP. - Dollarization and trade openness: - Loan dollarization: 33 percent. - Deposit dollarization: 41 percent. - Import share: 55 percent of GDP. - Exchange rate management and FXIs: - Frequent NBM interventions in the FX market; in 2020 benchmarking shows a moderate footprint, but comparing volumes vs interbank transactions indicates NBM’s footprint was large in most months with only a few months with no or minor interventions. - Frequent and large FXIs and restricted capital mobility can prevent the exchange rate from adjusting to policy rate changes. - Financial account restrictiveness: - Moldova’s financial account remains restrictive, especially on outflows; Financial Account Restrictiveness Index (FARI) based on the 2018 AREAER: inflows-side restrictiveness in line with peers but substantially higher than in Advanced Economies; outflows-side deemed very restrictive even compared to peer countries. - Shallow interbank and securities markets: - Interbank market in local currency largely inactive and completely dried up after the 2014 banking crisis. - Primary and secondary T-Bill markets increased over the last two years but remain small compared to peers; overall stock and securities markets are extremely shallow. - Weak financial intermediation and lending: - Despite middle-range bank capitalization among peers, lending indicators are particularly low in Moldova. - High remittances inflows tend to increase deposits that are often not matched by equivalent growth in credit. - Bank concentration is at the higher end among peers. - Excess liquidity legacy of the 2014 banking fraud: - NBM injected sizeable liquidity after the 2014 banking fraud; excess liquidity has only partially been mopped up by increases in RRs. - Excess liquidity impairs interbank lending and pass-through of policy rate changes to bank rates and loan supply; as credit increases, liquidity slowly comes down but normalization remains a long way off. - Growth of externally funded non-bank credit organizations: - Regulatory arbitrage after the banking crisis led to growth of non-bank credit organizations funded predominantly from abroad. - Domestic monetary policy can only marginally influence non-banks’ lending conditions, potentially reducing the scope of monetary policy. ### E. Remittances: channels through which they affect monetary transmission (Box 1) - Dutch-Disease-like effects: - Persistent remittance inflows can appreciate the real exchange rate (REER), weakening tradable sector competitiveness and possibly driving frequent, non-coordinated FX purchases. - Remittances allow higher spending on tradable and non-tradable goods; prices for tradables are set internationally so only non-tradable prices increase, expanding non-tradable supply and shifting resources away from tradables. - Institutional and financial market effects: - High remittance inflows associated with weaker institutional environment, undermining financial market development. - Remittances provide a stable inflow not sensitive to interest rates, narrowing the scope of monetary policy. - Remittances increase bank deposits but are often not matched one-to-one by credit increases, generating excessive liquidity on banks’ balance sheets and reducing interbank market activity. - Moldovan evidence: - Deposit-to-GDP ratio increased in the early 2000s and remained relatively stable since then; loan indicators have been going down since the early 2010s. - The banking fraud was responsible for a large part of the decrease in credit to the private sector, but high remittances inflows might have contributed as well. ### F. Empirical analysis of monetary transmission (overview) - Empirical approach: VAR to identify monetary policy shocks and estimate inflation response; sample period 2010–2020 split into two subperiods to account for structural break after the banking fraud became public. - Overall finding: - Effectiveness of transmission of policy rate changes to inflation weakened after the banking fraud. - First subperiod (2010–14): in response to an increase in the policy rate by one percentage point, prices decreased by around 0.4 percent after three quarters (statistically insignificant). - After the banking fraud: peak response weakened to a decrease of 0.2 percent after five quarters. ### G. Identified most impaired channels and contributing frictions - Exchange rate channel: - Likely impaired by frequent FXIs, restricted financial account (especially outflows), dollarization, and the economy’s sensitivity to exchange rate swings given 55 percent import share and medium dollarization levels (33 percent loans, 41 percent deposits). - Interest rate channel: - Likely impaired by collapsed interbank market, shallow securities markets, and excess liquidity from post-2014 interventions. - Credit channel: - Likely impaired by subdued bank intermediation, high remittances increasing deposits without commensurate credit growth, bank concentration, legacy of banking fraud, and growth of externally funded non-bank credit organizations. _Italic: Prepared by Julia Otten; empirical details and methodology are described in the technical annex._ ### 13. For the interest rate channel, short-term transmission from the policy rate to market ### 13. For the interest rate channel, short-term transmission from the policy rate to market rates weakened after the banking fraud, while long-term transmission remains solid. ### Interest rate channel — causation and pass-through estimates - Granger causality: the policy rate affects money market, T-Bill, deposit, and lending rates, but not vice versa. Money market rates affect lending and deposit rates, but not vice versa. - ADL regression estimates (example period and magnitudes): - Between 2010 and 2014, a change in the policy rate by 100 basis points led to: - deposit rate: change of 6 basis points in the following month, long-term change of 4 basis points. - lending rate: short-term pass-through of 24 basis points. - money market rate: short-term pass-through of 48 basis points. - T-Bill rate: short-term pass-through of 60 basis points. - Data limitation: Money market rate data is only available for 2010M1 to 2015M4. - Evolution: - After 2014, short-term pass-through from policy rate changes to market interest rates remained low. - Long-term transmission improved (possibly linked to slowly deepening financial markets). - Results for lending and deposit rates are statistically insignificant and very sensitive to the sample period analyzed. - Possible cause of persistent low short-term pass-through: excess liquidity injected into the system, reducing interbank activity and banks’ incentives to immediately pass on interest rate changes. ### Credit (bank lending) channel - Definition and identification: the bank lending channel captures the effect of policy rate changes on banks’ loan supply; identification uses bank-level lending and liquidity data and interaction terms between policy rate changes and a liquidity indicator. - Findings: - First subperiod: evidence of an operative bank lending channel (positive sum of coefficients on the interaction term), driven mainly by small banks (likely liquidity constrained). - Second subperiod: no evidence for an active bank lending channel (negative sum of coefficients on the interaction term). - Likely drivers of breakdown: - Injection of liquidity and efforts to clean up the banking sector in the aftermath of the banking fraud. ### Exchange rate channel and FX interventions (FXIs) - Aggregate VAR-based finding: Transmission via the exchange rate channel is weak, reflecting restricted capital mobility. - Subperiod differences: - First subperiod: CPI response would have been weaker and slower without the exchange rate — exchange rate channel was active but with only a small impact. - After 2015: CPI response would have been stronger without the exchange rate channel — exchange rate was effectively undermining transmission of the policy rate. - Degree of exchange rate management: - At quarterly frequency: Moldova’s exchange rate management is moderate and aligned with IT emerging economies. - At daily (high) frequency: exchange rate management is much stronger, suggesting active high-frequency FXIs that may over-manage lower-frequency movements. - Drivers of FXIs: - Exchange rate volatility is the dominant driver of FXIs; deviations from medium-term exchange rate level play only a minor role. - Empirical likelihood of interventions (average over whole sample, weekly basis): - likelihood of purchases: about 40 percent, - likelihood of no intervention: 45 percent, - likelihood of sales: 15 percent. - In crisis times, likelihood of FX sales increases. - Asymmetry and effectiveness of FXIs: - NBM purchases much more often than it sells; average purchase volume is smaller than average sale volume. - Event-window analysis (three weeks before to three weeks after interventions): - FX purchases happen in weeks after the exchange rate appreciates relative to average by about 0.3 percent; after purchase the difference comes down to not different from zero. - FX sales occur in weeks after the exchange rate depreciates relative to average by about 0.8 percent; after sale the difference comes down to not different from zero. - Interpretation: FXIs are effective in influencing the exchange rate in the very short term. One-sided, repeated interventions can have more permanent effects. - Coordination with interest rate policy: - FXIs are not always aligned with interest rate policy and can counter it, weakening monetary transmission. - Example episodes: - Crisis years 2015/16: policy rate increase accompanied by large FX sales, which stemmed depreciation and fostered monetary tightening. - 2017/18: NBM increased policy rates while purchasing FX, preventing appreciation and thereby impairing tightening effects on inflation. - Mechanisms of harm: - One-sided FX purchases can prevent exchange rate appreciation when policy rates rise. - FX purchases increase liquidity (if only partially sterilized), potentially adding inflationary pressure. - Conflicting signals from FX purchases during monetary tightening can blur messaging and undermine IT credibility. - Strategy and practice since 2020: - In 2020, NBM adopted a FXI strategy to foster two-way exchange-rate flexibility; strategy specifies intervention criteria for disorderly conditions and excess volatility. - Daily intervention data suggests the strategy was broadly followed: most interventions occurred on days when at least one criterion was satisfied; some interventions happened on days without criteria but with very small volumes (NBM reports some of these were preemptive). - The more intervention criteria fulfilled, the larger the average intervention size. ### International experience on FXIs in IT frameworks (Box 2) — selected country examples and lessons - General: Many LLMICs with IT regimes use FXIs to curb excess exchange rate volatility; FXIs can coexist with IT if narrowly defined and targeted, but can entail costs (hindering market development, undermining credibility). - Peru: - Adopted a hybrid framework (IT with managed float and de-dollarization strategy). - Dollarization outcomes cited: - 26 and 35 percent of loan and deposit dollarization, respectively, at the beginning of 2021; - over 60 and 70 percent, respectively, at the beginning of the 1990s. - Result: sterilized FXIs helped guard against exchange-rate-related financial risk and reduced dollarization over time, but interventions constrained financial development (e.g., low credit-to-GDP ratio, limited FX hedging instruments). - Ukraine: - Adopted IT in 2016; FXI strategy implemented from the start. - NBU’s FXI objectives include: - (i) accumulating international reserves; - (ii) smoothing out excess exchange rate volatility; - (iii) supporting the transmission of the policy rate “when it is not efficient enough”. - Research caveat: FXIs are warranted primarily when there are currency mismatches and shallow FX markets; FXIs can encourage moral hazard, disincentivize hedging, and risk undermining credibility of the policy rate in IT frameworks. ### Policy recommendations - Strengthen IT framework credibility and policy consistency: - The policy rate should always be NBM’s primary instrument to achieve its inflation target. - Any additional instruments (reserve requirements for liquidity management, FXIs for disorderly FX conditions) must be coordinated with the policy rate to avoid undermining IT credibility and effectiveness. - Improve public communication on outlook, objectives, and instruments to help anchor inflation expectations and improve the monetary transmission mechanism (MTM). - Gradually reduce NBM’s presence in the FX market to facilitate external adjustment and improve exchange rate channel transmission: - Rationale: three factors associated with large and frequent FXIs harm monetary transmission, especially during tightening: - persistent and one-sided FXIs can prevent exchange rate appreciation when policy rates are hiked; - FX purchases increase liquidity and can add upward pressure on inflation if only partially sterilized; - conflicting signals from FX purchases during tightening can blur messaging and undermine IT credibility. - Continue evaluating and possibly improving the parameterization of the 2020 FXI strategy to gradually improve exchange rate flexibility. - Communicate the main objectives of the strategy to banks so they can make informed risk-pricing decisions. - Foster FX market development: - Improve pricing of exchange rate risks and develop hedging instruments. - Given a still less-open capital account relative to peers (limiting exchange rate shock absorption), review FX market liberalization policies to identify and eliminate obstacles. - National de-dollarization strategy: - Implement a national de-dollarization strategy to limit balance sheet exposure to exchange rate shocks (see Box 2 for international experience). *Source: IMF staff analysis as presented in the chapter.* ### 23. Facilitating financial market development and reducing excess liquidity is essential to ### 23. Facilitating financial market development and reducing excess liquidity is essential to ### Financial market development, excess liquidity, and monetary transmission - Key objectives: - Foster transmission via the interest rate and credit channels. - Foster lending to households and small and medium enterprises without compromising financial stability. - Safeguard progress in reforming the banking sector and continue progress in financial sector reforms. - Policy recommendations and priorities: - Strengthen regulation and supervision of non-bank financial institutions (NBFIs) to eliminate opportunities for regulatory arbitrage and ensure financial stability while addressing risks and increasing consumer protection without depriving NBFIs’ customers of access to lending. - Leverage the forthcoming transfer of supervision from NCFM to NBM to use NBM’s expertise in supervising NBFIs. - Adopt a strategy to foster financial inclusion and capital market development to broaden the financial sector’s customer base and widen NBM’s scope of action. - As credit to the economy grows and excess liquidity is reduced to more normal levels, normalize reserve requirements (RRs) to reduce banks’ funding costs. - Move away from the passive full-allotment system towards a proactive liquidity forecasting and management system when excess liquidity is reduced. - Strengthen budgetary liquidity and debt management procedures and promote development of domestic securities. ### Technical Annex A — VAR analysis of policy rate transmission to inflation - Data used: - Quarterly data from 2010Q1 through 2020Q3 of: policy rate, GDP, CPI, broad money, credit to the economy, and the nominal MDL-USD exchange rate. - Methodology: - Vector Autoregressions (VARs) estimated as Yt = A1 Yt−1 + … + Ap Yt−p + ut, with ut having variance-covariance matrix Ω. - Structural form: Yt = B0 Yt + B1 Yt−1 + … + Bp Yt−p + εt, with E[εt εt′] = I and residuals ut = A0 εt, Ω = A0 A0′. - Identification: Cholesky ordering assumed that a monetary policy shock affects variables in the order: exchange rate, interest rate, broad money, CPI, GDP. - Impulse responses derived from Yt = [I − A(L)]−1 A0 εt. ### Technical Annex B — Interest rate pass-through, ADL and Granger causality - Data used: - Monthly data of the policy rate, deposit rate, lending rate, money market rate, and T-Bill rate from 2010M1 to 2020M12. - Granger causality: - Pairwise Granger causality analysis uses two lags. - Example result: in the first row of the Granger analysis table on page 11, the hypothesis that the deposit rate does not Granger cause the policy rate cannot be rejected, but the hypothesis that the deposit rate does not Granger cause the policy rate can be rejected — implying one-way Granger causality from the money market rate to the deposit rate and not the other way. - ADL modelling: - An Autoregressive Distributive Lag (ADL) model is estimated for every relevant pair of variables: - Δxt = α Δxt−1 + β Δxt−2 + γ Δyt + δ Δyt−1 + ε Δyt−2 + ηt. - Short-term effect given by estimate of coefficient γ̂. - Long-run effect given by (γ̂ + δ̂ + ε̂) / (1 − α̂ − β̂). ### Technical Annex C — Credit channel analysis - Data used: - Monthly bank-level data from 2010M1 to 2021M1 on lending volumes and liquid-assets-to-total-assets ratio for 11 banks (balanced panel). - Approach and identification: - Microeconometric pooled regressions estimate the effect of (excess) liquidity at the bank level on lending supply. - Variables: - Lit is lending volume. - Bit−1 is ratio of liquid assets to total assets (capturing (excess) liquidity and balance sheet strength). - Mt is the policy rate. - Yt is a business cycle indicator. - Interpretation: - An expansionary monetary policy shock should reduce the number of credit-constrained banks if the bank lending channel is active; this would be reflected by a positive sum of coefficients on the first interaction term and its lags (per Kashyap and Stein, 2000). ### Technical Annex D — FXI analyses - Data used: - Weekly data of NBM FX sales and purchases and the nominal MDL-USD exchange rate; aggregated from daily data. - FXI index construction: - Index = σ_t^FXI / (σ_t^FXI + σ_t^E), where σ_t^FXI is the standard deviation of FXIs and σ_t^E is the standard deviation of the exchange rate. - For the index at quarterly frequency: weekly data aggregated to quarterly frequency, then a 12-quarter rolling window is used to compute standard deviations. - For the index at daily frequency: the standard deviation of daily data over one quarter is computed for both FXIs and the exchange rate. - Ordered logit model for intervention probability: - Determinants: eee_t = percentage deviation of the exchange rate at time t from its average in a 12-week rolling window centered around t; vvvvv_t = proxy for volatility = standard deviation of the exchange rate in a 12-week moving window centered around t. - Latent specification: yt* = α + β eee_t + γ vvvvv_t + εt. - Observed outcomes mapped to cut-offs κ1 and κ2 with outcomes: FXI purchases, no intervention, FXI sales. - Event-window analysis specification: - Exchange rate log-differences (ee) multiplied by dummy variables S (1 in periods of FX sales) and P (1 in periods of FX purchases) are used to assess FX sales/purchases effects on exchange rate changes. ### Public enterprises, governance, and macroeconomic vulnerabilities (overview) - Public enterprise sector context in Moldova: - Moldova’s public enterprise sector comprises over 900 state-owned companies. - The paper analyzes financial performance and risks for 326 public enterprises for which data are available (roughly one-third of all public enterprises). - Governance challenges and economic implications: - Weak governance frameworks manifest as unclear state ownership objectives, weak legal/regulatory umbrellas, poor institutional setups, frail checks and balances, and undue state intervention. - Consequences include fiscal and quasi-fiscal risks, corruption and mismanagement, misallocation of resources, weakened market competition, reduced private investment, and higher fiscal costs when governments continue funding poorly performing enterprises. - Structure of the public enterprise governance analysis: - Section B: links between governance weaknesses, corporate performance, fiscal costs, and macro-vulnerabilities. - Section C: summary of Moldova’s public enterprise sector size, scope, coverage, legal framework, and organizational/institutional set-up. - Section D: international best practices and comparison with Moldova’s framework. - Section E: in-depth assessment of financial performance and risks of 326 public enterprises. - Section F: stress tests for individual enterprises and a cohort with the largest liabilities to explore links between corporate risks, fiscal costs, and macroeconomic vulnerabilities. - Section G: framework and strategy to reform governance of public enterprises in Moldova and key recommendations. - Key governance role: - A strong governance framework clarifies mandates, management structures, boards of directors, legal setup, reporting and audit requirements, transparency and public disclosures — thereby shaping enterprise size, activities (commercial vs non-commercial), financial viability, and the degree to which enterprise performance feeds into public sector or private sector risks. *Source: Chapter/section "23. Facilitating financial market development and reducing excess liquidity is essential to" from the provided IMF content unit.* ### 7. Corporate performance and the complex financial interlinkages with the state are a ### 7. Corporate performance and the complex financial interlinkages with the state are a direct source of fiscal risks and costs to the state ### Fiscal risks from corporate–state linkages and feedback loops - Public enterprises rely on the central government for grants, loans, subsidies, extension of sovereign guarantees, and other transfers; such dependence creates contingent fiscal liabilities and poses fiscal costs and risks to the state, especially when firms have weak financial performance. - Fiscal policy actions by the state (for example, cuts to publicly funded companies’ capital allocations due to countercyclical policy needs) can: - Complicate execution of companies’ budget plans; - Slow down operations; - Raise firm-level risks to profitability, liquidity, or solvency; - Produce adverse feedback loops to the state via lower tax, dividend, and other transfers. - Elevated, persistent firm-level risks can require prompt government intervention; inaction can result in sector-wide systemic risks that are costlier to address and have macro-fiscal implications. ### Quantified corporate vulnerabilities and direct financial flows to the state - Total debt and arrears outstanding on companies for which data is available amounted to 7.3 percent and 3.3 percent of GDP, respectively, as of end-2018. - A sizable share of debt is held by a handful of large firms in key sectors: energy, public works, telecommunications, and transportation. - On-lending to public enterprises appears to have added little value to their performance, based on elevated risk profiles for four key companies in strategic sectors (energy and transportation). - Net financial flows from government to public enterprises can be high and constitute a large fiscal drain; example cited where inter-governmental financial transactions added 6 percent of GDP to general government expenditures over a five-year period (2014–18) in another country. ### Size, scope, and fiscal footprint of Moldova’s public enterprise sector - Over 900 public companies in the sector. - One-third operate at the central government level; the remainder at sub-national (municipal or local) level. - Combined assets valued at about 21 percent of GDP (US$2.5 billion), concentrated in four key sectors: public works, energy, telecom, and transportation. - In 90 percent of enterprises operating at the central level, the controlling share of the state was 100 percent; only 44 companies had state participation below 50 percent. - Companies at the central level hired almost 40 thousand employees (as of 2018)—about 4 percent of total national employment. - This dominance in employment and shares is reportedly more than twice the average in other Eastern European countries in the region. - Fifteen companies reportedly listed on the stock exchange (as of 2016). - The largest five enterprises (State road administration, Termoelectrica, Moldtelecom, Moldova Railways, and Red-Nord) had assets (and outstanding liabilities) constituting roughly 6–7 percent of total assets and total liabilities out of the 326 companies for which data is available. - Companies at sub‑national level are much smaller; only three enterprises had liabilities exceeding MDL 100 million (liabilities of the largest company are below MDL 500 million). ### Financial performance, equity, and solvency concerns - Total net worth improved in 2019 driven by growth in equity in most central government sectors, though net worth as a share of GDP dropped slightly from 2018 to 2019. - Net worth grew 5 percent annually (aggregate measure cited). - One-fifth of all companies suffered negative equity; the majority of these at the local level. - Many companies incurred sustained net losses that eroded retained earnings and shrank overall equity positions. - Five enterprises had the value of net assets lower than their statutory capital (contrary to the law on Joint-Stock Companies’ requirements), plus another 28 state-owned enterprises at the central government level, and 92 other enterprises operating at the sub-national government level. ### Organization, legal framework, and governance arrangements - State ownership follows a centralized model supervised by the Public Property Agency (PPA). - Legal forms: state-owned enterprises (SOEs), joint-stock companies (JSCs), and limited liability companies (LLCs). - Principal laws: Law 246 (2017) on state owned enterprises and municipal enterprises; Law 1134 (1997) on Joint Stock Companies; Decision 902 (2017) on organization and functioning of the PPA. Other relevant laws include Law 149 (2012) on Insolvency, Law 183 (2012) on Competition, and Civil Code 1107–XV (2002). - Four management bodies govern public enterprises: the founder (PPA), the board of directors (BoD), the administrator (executive body), and the board of auditors (BoA, or commission of censors). Key institutional features (as specified in existing law) - PPA powers include approving statutes, regulating BoDs, appointing BoAs, appointing/revoking BoD members, establishing monthly remuneration for senior management, deciding on annual deductions from net profits to be transferred to state/local budgets, approving distribution of annual net profits, raising/lowering share capital, and agreeing to pledge state assets as collateral for bank loans. - BoD functions for SOEs: approve business plans, set performance indicators, approve annual finances including personnel and salary fund, ensure transparency of procurement, select the administrator (by competition) and audit entity, and present proposals to the PPA to improve management and streamline activity. - Administrator functions: daily operations, executing PPA and BoD decisions, present quarterly financial reports to the BoD, and submit annual financial statements and audit report to the PPA and BoD. Administrators appointed for up to five years. - BoA functions: bi-annual audits and unannounced audits, present audit reports to administrator and BoD, members appointed for up to 2 years and must be qualified in accounting, finance, or economics/jurisprudence. ### Identified governance weaknesses affecting corporate performance - Governance shortcomings documented as drivers of weak performance: - High perceptions of corruption due to mismanagement of corporate portfolios; - Poor transparency and disclosure practices; - Political interference in company decisions and forced policy mandates; - Inadequate supervision and opaque lines of authority; - Prevalence of monopolistic / oligopolistic practices introducing market inefficiencies and weakening the private sector. - These challenges are linked to a lack of clarity on the state’s overall ownership policy and weak control over adherence to institutional, legal, and operational requirements by public corporations. ### International best practices in governance of public enterprises (principles) - Broad objectives for SOE governance: (i) professionalise the state as an owner; (ii) make SOEs operate with similar efficiency, transparency and accountability as good practice private enterprises; and (iii) ensure competition between SOEs and private enterprises is on a level playing field. - Key principles and measures: - Ownership rationale: governments should provide a rationale for the creation of public companies and ensure it is enshrined in legislation, including facilitating termination when rationale ceases. - Criteria for SOE classification: whether an entity meets definition of an institutional unit, charges economically significant prices, and depends on regular financial assistance from the government. - Grant companies full autonomy in executing operations; respect independence and avoid opaque intervention or redefinition of objectives. - Ensure the ownership entity is accountable and endowed with competencies/capacity to carry out duties effectively. - Clarify BoD responsibilities: boards should have clear legislative mandate, set strategy, oversee management, and exercise independent, objective judgment. - Hold BoDs accountable for financial performance via impartial performance assessments and transparency over board participation in other companies. - Combat corruption and mitigate conflicts of interest through strong internal controls, ethics, compliance measures, transparent and merit‑based board selection, and appropriate remuneration. - Strengthen corporate disclosure and transparency: regular disclosure and reporting requirements, independent external audits annually, and publication of high-quality annual reports by the ownership entity; mandate disclosures of BoD and senior management remuneration and governance structures. - Safeguard a level playing field with private sector firms: avoid exemptions, preferential access to finance or inputs, and ensure competitive, transparent procurement and access to legal recourse for shareholders. - References made to OECD, World Bank Group, IMF guidance (GFSM 2014, IMF Georgia TA report July 2020) in framing these principles. *Source: Chapter excerpt from IMF country report material on Moldova (text and boxes provided in the source content).* ### 18. Moldova has elements of a public corporate governance regime. It follows a centralized ### 18. Moldova has elements of a public corporate governance regime. It follows a centralized ### Governance framework and institutional setup - Moldova follows a centralized model with one state body (the PPA) exercising ownership rights on behalf of the state and has adopted laws over time to help guide the regime governing the operations of public corporations. - According to a self-assessment survey, Moldovan legal and institutional framework fares above average on an SOE governance index relative to other countries in central, eastern and south-eastern European economies. - Moldova ranks 6th highest (out of 20 countries in the region) in a composite index measuring SOE governance that covers ownership policy, financial oversight, and fiscal and policy interactions. ### Identified legal, institutional, and operational shortcomings - A thorough review of the existing governance framework is needed (Annex 2). - Specific shortcomings highlighted: - An absence of a comprehensive state-ownership policy and strategy document that ultimately defines the objective for state ownership and the rationale behind establishment and termination of public corporations (including the criteria to determine their legal form as public corporations versus other general government units, criteria for extension of state aid, etc.). - Weak enforcement and less-than-ideal implementation of laws and regulations in practice, despite appearing to be comprehensive on paper. - Example: audits — there appears to be three layers of audit (internal company audit, external expert audit, and audit performed by the court of audit) that are not applicable to all public companies. - Legislation that does not clearly safeguard independence of company boards. Heavy participation by the state on company boards also calls into question their independence. - Poorly defined organizational arrangements and institutional set-ups that blur the respective roles and responsibilities between the state (MoF, line ministries), the PPA, and company boards as far as oversight, managerial, and ownership functions intersect, and between enterprises and the MoF, external donors, and other levels of government, adding to confusion over the appropriate lines of authority and hindering accountability. - Overly stretched mandate of the PPA amid capacity constraints that undermine its ability to execute its functions efficiently and effectively. - Limitations on the quality and availability of data on all public enterprises at all levels of government undermine a solid understanding of the complex financial transactions between enterprises and the state and prohibit a comprehensive assessment of fiscal costs (risks). Data constraints also reaffirm concerns over the lack of proper record keeping and accounting standards, fueling perceptions of corruption and poor transparency and disclosure practices. ### Financial performance assessment: scope and methodology - The SOE health check toolkit developed by the IMF’s Fiscal Affairs Department (FAD) is used to assess the financial performance of 326 public enterprises in Moldova in 2019, based on official data from financial statements provided by the authorities (income statements and balance sheets). - The toolkit assesses performance through standard financial ratios covering profitability, liquidity, and solvency metrics on a company level and on aggregated basis. It estimates assigned risks in each of these metrics using pre-set benchmarks tailored to Moldova (Annex 1). ### Profitability findings - Overall profitability was very poor: 60 percent of public companies incurred net losses and seven sectors faced recurring losses. - Net income was negative in 195 companies out of the 326 under study, dragged down by over 150 companies that posted losses at the sub-national level (particularly in water, other services and agriculture activities). - The sector as a whole suffered a minor loss in 2019 (compared to a slight net profit in 2018) because net profits at the central government level were wiped out by losses at the sub-national level. - Out of a total of 16 sectors, four at the central government level and 8 at the sub-national level posted overall losses, and seven sectors carried over persistent losses from 2018. - ROA averaged 5 percent overall but exhibited considerable variation across sectors and government levels. - ROA averaged 5 percent at the central government level and recorded negative 2 percent for the sub-national companies. - ROE was low at 1.3 percent for companies at the central government level, but averaged negative 10 percent at the sub-national level. - Cost recovery: - The cost recovery ratio hovered only mildly above “1” in 10 sectors. - Six sectors had poor cost recovery ratios: significantly below 1 in agriculture and water activities. - More sectors at the local level struggled to achieve cost recovery relative to those at the central level. - Labor costs: - A third of the 326 enterprises pay salaries that exceed 75 percent of their total expenses, with 47 enterprises paying salaries comprising over 90 percent of their total expenses. - Labor costs absorb almost 15 percent on average of total enterprise revenue; 14 companies are burdened with salary costs that exceed their revenue intake. - The share of salaries to expenses and to revenue at the sub-national government level are over 60 percent and over 40 percent, respectively, compared to 36 percent and 15 percent at the central government level. - State subsidies: - Total subsidies extended to only four public enterprises (operating at the central government level) reached 1.5 percent of GDP, of which the overwhelming amount was granted to one state company managing roads (the state roads administration). - That subsidy exceeded the enterprise’s operating revenue by 20 percent; two of the other companies had realized losses despite being directly subsidized by the state. - Depreciation and amortization costs in key sectors dampen profitability prospects; three sectors bear the brunt of depressed profit margins due to significant debt amortization and depreciation costs. ### Liquidity findings - Net liquid assets position was broadly adequate overall but masks discrepancies. - All sectors appear broadly liquid except for telecommunications and water. - Net liquid assets position at the sub-national level was negative, much weaker than at the central government level. - Current ratio: - On average, all sectors had current ratios above “1” except four sectors, implying good short-term liquidity for most. - Some sectors had current ratios exceeding “2” (transportation, construction, industry, trade); telecommunications, agriculture, tourism and water activities had insufficient liquidity. - Turnover indicators: - Debtor turnover averaged 2 months on average (for enterprises with available data). - Creditor turnover came in worse at over 4 months on average. - Forty-six enterprises recorded debtor or creditor turnover exceeding 1,000 days. ### Solvency findings - Solvency is a major problem at all levels of government. - Debt-to-assets: - Manageable overall (less than 0.5 for most sectors), but higher in water services, agro-industry, electricity and agriculture at the central government level, and in construction and telecommunications at the sub-national level. - 9 companies had debt exceeding their assets; among these, two companies had debts three-to-four times the size of their assets. - Debt-to-equity: - Elevated and concerning, especially in companies with negative equity. - Many sectors suffered excessive (double digit) ratios: agro-industries, water sector, retail, agriculture and construction. - The picture is notably worse at the sub-national level. - Debt-to-profits (EBITDA): - Extremely high and particularly worrying in loss-making enterprises. - Indebtedness in five sectors was 20–50 times the size of their earnings, highlighting severe financial instability and potential insolvency. ### Overall financial risk assessment - Overall, Moldova’s 326 public enterprises under study face a “High” overall level of financial risk, both at the central and sub-national level of government. - Profitability risk: “Moderate” overall, but “High” at the sub-national level. - 200 enterprises faced “high” or “very high” risks to their profitability; three-quarters operated at the sub-national level. - Liquidity risk: “High” overall, but “Very high” at the sub-national level. - 102 enterprises suffered from “high” or “very high” liquidity risks; two-thirds of these were at the sub-national level. - Solvency risk: “High” overall and at all levels of government. - 128 enterprises faced “high” or “very high” solvency risks; three-quarters operated at the sub-national level. *Source: Republic of Moldova — IMF staff analysis and PPA data (chapter content).* ### 29. A striking feature is the large share of ### 1mdaea2022003 - 29. A striking feature is the large share of ### Scale and concentration of risks in Moldova’s public enterprises - 250 companies face combined or multiple financial risks, representing over three-quarters of the sample. - 44 companies form a “core” group facing “high” or “very high” risks across profitability, liquidity, and solvency. - Of these 44 core companies: - 14 are at the central level of government, including 3 companies operating in the energy sector (production, supply, transmission, distribution, etc.), which are also among the largest by outstanding liabilities. - 30 are at the sub-national level, including 1 company operating in the water and sanitation sector that is also large by liability size. - The core group’s liabilities make up over 13 percent of the 326 public companies’ liabilities, or 1 percent of GDP. - Prior to COVID-19, over 15 percent of public enterprises showed worsening risk profiles: - 40 companies reduced overall financial risk in 2019 relative to 2018 (example: Cricova wine company). - 50 companies saw riskiness rise in 2019 (including two electricity companies). - Data for 2020 is unavailable; the pandemic likely intensified vulnerabilities and could contribute to deeper scarring and further fiscal and macroeconomic risk buildup. ### Stress testing approach and scenario design - Focus: 15 companies with the largest liabilities (all had at least one “high” or “very high” risk except the largest—State Road Administration). - State Road Administration liabilities equivalent to 2.1 percent of GDP. - Stress testing tool: IMF Fiscal Affairs Department SOE Fiscal Risks and Stress Test tool. - Two modeled shocks applied for 2021: - Growth shock: assumes a 1.5 percentage points weaker economic growth recovery from the pandemic in 2021 (relative to the baseline) coupled with a weaker leu (10 percent). - Liquidity shock (added to growth shock): liquidity constraint defined as a 30 percent share of receivables potentially materializing. - Results: immediate deterioration in profitability indicators, weaker liquidity positions, and worsening debt indicators for the 15-company group, with losses that can be permanent beyond the immediate year. ### Macroeconomic repercussions and contingent fiscal liabilities - Worst-case modeled scenario: government assumes the liability burden of the 15 largest enterprises. - Adds about 8 percentage points of GDP to public debt in the year of the shock (2021). - Adds 16 percentage points of GDP to public debt over the next 5 years. - Additional burden of absorbing economic losses estimated at: - 0.7 percent of GDP in 2021. - 1.4 percent of GDP cumulatively over the medium term. - These estimates exclude other forgone financial flows (e.g., tax and dividend transfers, tariff collections), and are described as conservative. - Mechanisms of macroeconomic transmission highlighted: - Higher fiscal costs → unsustainable public debt trajectories → financial market destabilization (squeezing private credit, raising government borrowing costs). - Government financing constraints → cuts to productive spending or abrupt revenue increases → loss of confidence. - Supply shocks from enterprise failures → more expensive imports, foreign currency losses, exchange rate pressures. - Labor market costs (unemployment benefits, severance, retraining) and broader social costs. ### Case study: State Roads Administration (Box 4) - The State Roads Administration: - Manages state assets worth almost 6 percent of GDP. - Employs about 170 staff. - Receives government subsidies amounting to 1.4 percent of GDP. - Reported net income of MDL 0.8 million in 2019, inclusive of state direct subsidies. - Removing direct state support in stress tests reveals: - Underlying operational weaknesses; would have faced a huge loss, hampering cash flow and ability to meet obligations (imminent liquidity and solvency risks). - Possible responses include asset sales, expensive local borrowing, or tapping reserves (which would have depleted/wiped out 40 percent of its equity), ultimately requiring costly capital injection via state intervention. - Policy questions raised for such enterprises include: rationale for state ownership and subsidies, opportunity cost of large public funds to one company, efficiency of state asset management, market competitiveness, and managerial oversight. ### Case study: National Railway Company (Box 5) - National Railway Company profile and recent performance: - Second largest by asset size, managing state assets worth 1.6 percent of GDP. - Employs around 7,600 staff. - Recipient of on-lent external financing from the state. - 2019: sales revenues increased and cost of goods sold declined, but company incurred a large cumulative net loss for a second consecutive year and saw a reduction in equity (net worth). - Profitability: positive rate of return on assets but negative ROE — high profitability risk. - Liquidity: moderate risk due to strong current and quick ratios and a positive net liquid asset position. - Solvency: very low insolvency risk as debt remained subdued relative to assets and equity. - March 2021: salary arrears reported and staff advised to take voluntary leave with less than full remuneration, reflecting COVID restrictions on travel demand. - Stress test results: - Growth shock: moderate vulnerabilities in profitability and liquidity. - Combined liquidity-macro shock: more pronounced vulnerabilities and stronger risks to company performance. - Both shocks highlight permanent nature of losses beyond the immediate year. - Governance recommendations: strengthen executive board oversight, set clear forward-looking financial targets and KPIs, implement prudent risk management, and enhance managerial accountability for public funds (including on-lent funds). ### Reform roadmap and policy recommendations - A clear reform roadmap anchored in five interrelated areas (aligned with international best practices) is recommended to strengthen governance at all levels of government. - At the state-wide level (MoF, line ministries, etc.): - Adopt an ownership rationale and policy strategy document specifying public policy objectives, rationale and criteria for existence/termination, classification criteria, legal forms, commercial vs non-commercial mandates, sector choices, etc. - Strengthen MoF’s roles in fiscal risk oversight and fiscal cost and risk management; expand the Fiscal Risk Statement (FRS) to cover at least the 44 “core” companies and other systemically important enterprises and those with large liabilities. - Improve coverage, monitoring, inter-agency coordination, and reporting of contingent liabilities. - Clarify state aid rules and intervention criteria: define fiscal/quasi-fiscal activities, identify net financial flows, specify when intervention is allowed/prohibited, and set rules on bail-out clauses and forms of fiscal support (subsidies, transfers, grants, loans, on-lending, guarantees, etc.). - Reduce fragmentation of roles/responsibilities across MoF, line ministries, legislature, and other bodies; clarify lines of authority, managerial responsibility, accountability, and oversight splits. - Strengthen oversight, accountability, and capacity (human and financial) of the PPA. - Create a level playing field between public enterprises and private corporations to raise market efficiency and competition, including: - Review and amend legislation on privileged access to government resources (land, utilities, infrastructure), transparency in public procurement, and extension of loans/grants/support. - Review and amend laws to enhance domestic and foreign private investment. - Review anti-monopoly legislation to curb harmful monopolistic/oligopolistic practices. - Balance economic vs social returns in public companies vis-à-vis private counterparts. - At the PPA level: - Comprehensive review of institutional arrangements governing public enterprises; identify management structures, address opaqueness, and establish clear codes of conduct to ensure accountability. - Comprehensive review of legal and regulatory frameworks to: - Root out corruption and conflicts of interest; close loopholes and enforce stringent penalties. - Review and clarify remuneration rules for boards and dividend payout policies. - Strengthen rules safeguarding policy independence and reduce heavy presence of state representatives on boards; report reform proposals to MoF. - Strengthen operational efficiency and incentives: - Require companies to set clear forward-looking performance targets (KPIs) and limits on liabilities to enable scenario analysis, stress testing, and adequate contingencies in budgets. - Enforce accounting standards and robust data/record keeping. - Hire and retain BoD and BoA members via transparent, merit-based processes; ensure remuneration is commensurate with expertise. - Enhance transparency and disclosure: - Strengthen the annual consolidated report to include employment information; key changes to company ownership policy; synopsis of intended reforms for selected enterprises; clearer breakdown between commercial and non-commercial enterprises; website links to companies; and reporting of mismanagement instances and proposed remedies. - Adopt a strategy to strengthen accountability over use and management of public funds, including cost/benefit analysis to assess efficiency of state asset use and to improve asset management. *Source: 1mdaea2022003 - 29. A striking feature is the large share of (excerpt).* ### 37. Proper sequencing of reform objectives and priorities, in addition to political will and ### 1mdaea2022003 - 37. Proper sequencing of reform objectives and priorities, in addition to political will and ### Sequencing, ownership, and political prerequisites for SOE reform - Proper sequencing of reform objectives and priorities, in addition to political will and stability, and ownership of the reform process will be key to its success (Annex 2). - A comprehensive assessment of the financial position of all SOEs operating at all levels of government is a key first step to identify public corporations at a risk of generating fiscal costs. - Results of the assessment should feed into and inform the development and adoption of a time-bound and explicit state-ownership strategy. - The state-ownership strategy will identify enterprises chosen to undergo restructuring by means of reorganization (mergers and acquisitions, or recapitalization), privatization or liquidation, as well as plans to strengthen their governance structures. - Given the long tenure associated with reforming SOEs and the scope of reforms likely to be needed in Moldova (see Annex 2), a stable political environment and government that owns the reform process and oversees its completion is vital. ### International experience and expected gains - Moldova stands to gain from leveraging international best practices and learning from reform experiences in other countries, including on mitigating measures (Annex 3). - Empirical experience noted: gains to productivity of between 6-14 percent in strategic sectors and labor cost savings of up to 2.5 percent as a result of public corporate governance reforms. - Reform policies with socio-economic implications (e.g., relocating furloughed or overstaffed workers, redesigning tariffs or abolishing administered or below-market prices) are often preceded by an upfront detailed assessment and review of potential effects of restructuring public companies to mitigate costs and ensure adequate safety nets. ### Additional recommendations for future consideration - Update the assessment using latest data for 2020 for a more comprehensive analysis and to gauge the impact of the COVID-19 pandemic on enterprise performance, both at the firm and sectoral levels (sensitivity analysis). This paper highlighted the extent to which the performance of many enterprises was already weak prior to the pandemic (2020), which is expected to lead to even deeper scarring of their operations. The anticipated build-up of further fiscal risks during 2020 reiterates the urgency of governance reforms in these enterprises. - Strengthen disclosure of critical missing data. Reliability of the assessment depends on the quality of underlying data and accounting standards. Strengthening disclosure on missing data gaps is needed, such as having comprehensive data on employment in all companies at all levels (central and municipal; both on staff figures and remuneration); data on all enterprises in which the state claims less than 50 percent of ownership, value added by sector, enterprise’s shares held by majority vs. minority shareholders, cross ownership across companies/sectors, and data on all other enterprises in general (the remaining 600 or so companies not analyzed). - Conduct further deeper analysis of the net financial flows from the government to public enterprises in Moldova, given the data gaps at the time this paper was drafted. Strong financial interlinkages related to indebtedness, guarantees, externally funded projects, transfers, subsidies—to name a few—were not reported/available for all enterprises. This undermines a true understanding of risks to profitability, liquidity, and solvency, and the overall financial viability and performance of these companies. An assessment of such missing information for all enterprises at all levels of government is vital to help identify the degree of dependence of such companies on government funding, and the role of government in their financial management. - Strengthen the assessment by reviewing and understanding intra-company debts, which could pose serious financial risks across sectors (e.g., between companies providing a service and others providing a good as input to such a service, e.g., the telecommunications sector). - Undertake a comprehensive Public Sector Balance Sheet (PSBS) approach in the future to provide a complete assessment of the financial performance and net worth of all public enterprises operating at various levels of government in Moldova, in addition to all other entities that the government controls (e.g., pension funds) to give a full picture of public wealth (or fiscal solvency). ### Key methodological definitions used in benchmarking - Return on assets = earnings before interest, tax, depreciation, amortization (EBITDA) / assets - Return on equity = net profit (loss) after tax / equity - Cost recovery = total revenue / (cost of goods sold + other expenses) - Current ratio = current assets / current liabilities - Quick ratio = (current assets - inventory) / current liabilities - Debtor turnover (days) = trade receivables*365 / total revenue - Creditor turnover (days) = trade payables*365 / cost of goods sold - Debt to assets = liabilities / assets - Debt to equity = liabilities / equity - Debt to EBITDA = liabilities / EBITDA ### Risk thresholds applied in the analysis (thresholds as assumed) - Profitability — Return on Assets: Very Low Risk greater than 8.00; Low Risk 4.00; Moderate Risk 0.00; High Risk -5.00 - Profitability — Return on Equity: Very Low Risk greater than 15.00; Low Risk 8.00; Moderate Risk 0.00; High Risk -10.00 - Profitability — Cost Recovery: Very Low Risk greater than 1.50; Low Risk 1.25; Moderate Risk 1.00; High Risk 0.75 - Liquidity — Current Ratio: Very Low Risk greater than 2.00; Low Risk 1.50; Moderate Risk 1.25; High Risk 1.00 - Liquidity — Quick Ratio: Very Low Risk greater than 1.20; Low Risk 1.00; Moderate Risk 0.80; High Risk 0.70 - Liquidity — Debtor Turnover (days): Very Low Risk less than 30.00; Low Risk 40.00; Moderate Risk 50.00; High Risk 75.00 - Liquidity — Creditor Turnover (days): Very Low Risk less than 30.00; Low Risk 60.00; Moderate Risk 90.00; High Risk 120.00 - Solvency — Debt to Assets: Very Low Risk less than 0.25; Low Risk 0.50; Moderate Risk 0.75; High Risk 1.00 - Solvency — Debt to Equity: Very Low Risk less than 0.50; Low Risk 1.00; Moderate Risk 1.50; High Risk 2.00 - Solvency — Debt to EBITDA: Very Low Risk less than 1.50; Low Risk 2.00; Moderate Risk 3.00; High Risk 5.00 - Interest Coverage: Very Low Risk greater than 2.00; Low Risk 1.50; Moderate Risk 1.20; High Risk 1.00 - Cash Interest Coverage: Very Low Risk greater than 3.00; Low Risk 2.00; Moderate Risk 1.50; High Risk 1.00 - Debt Coverage: Very Low Risk greater than 0.50; Low Risk 0.30; Moderate Risk 0.20; High Risk 0.10 ### Annex II — comparative assessment of Moldova’s governance framework against best practices (high-level findings) - Adoption of an explicit ownership policy: under development; alignment to best practices yet to be fully determined (see IMF, March 2021 report). - Rationale for establishing and terminating SOEs: legal framework clearer on termination than on rationale for establishment; reorganization/dissolution governed by Law 246 and JSC law provisions, with specifics on liquidator appointment and grounds for forced dissolution. - Aggregate reporting and audits: Law on SOEs mandates disclosure of annual financial reports and audit reports (published on enterprises’ webpages and PPA website), but it is unclear if mandates are met in practice. Audit arrangements and exemptions (e.g., Court of Accounts audited entities) may weaken audit objectivity. - Institutional fragmentation: involvement of the PPA, the MoF, and line ministries blurs lines between ownership and regulation. - Several areas remain unknown/unclear from legislation, including separation of competitive from non-competitive activity, fiscal and regulatory treatment akin to private companies, clear basis for financing decisions, access to non-preferential debt financing, and direct state support calibrated to public policy costs. - Board professionalization and governance: some minimum qualifications exist for audit committees but not consistently for boards of directors; rules on remuneration, regular board evaluations, and explicit risk management systems are not clearly specified in the laws reviewed. - Asset management and conflict-of-interest rules: Law on SOEs contains provisions on asset management but leaves operational aspects to other legislation; both the Law on SOEs and the Law on JSCs contain provisions addressing conflicts of interest and certain disclosure requirements. *REPUBLIC OF MOLDOVA — INTERNATIONAL MONETARY FUND* ### Annex III. Cross Country Experiences at SOE Governance Reforms ### Annex III. Cross Country Experiences at SOE Governance Reforms ### Cross-country experiences and lessons - Cross-country experiences in reforming public enterprises focused on targeted policies to: - strengthen governance and financial performance, - deepen private sector participation, and - mitigate externalities from reforms. - Many countries sought clarity on their ownership policy (rationale) following a comprehensive review of all enterprises. - Governance measures included: - strengthening legal frameworks and management, - separating management from supervision, - enhancing oversight and audit functions, financial reporting and oversight, and continuous monitoring, - reducing state interference and strengthening independence of company boards to uproot corruption, - improving internal controls and transparency (disclosure) requirements. - Financial and operational measures included: - addressing commercial viability through operational revenue enhancements and cost-effectiveness (e.g., structural reduction in expenditure financed through central government funding or support), - risk mitigation measures such as better monitoring and management of fiscal risks and liabilities, - improvements in asset management and putting state assets to better productive use to earn higher returns. - Competition and market-level reforms included: - introducing a stronger role for competition councils to address anti-competitive behavior, - restructuring enterprises via mergers and divestments, selling state assets, and increasing private sector participation. - Empirical evidence: - Improvements in corporate governance can enhance SOE performance, both operationally and institutionally (example: Lithuania during 2012-13). - Other studies assert strengthened governance is instrumental to improvements in public corporate efficiency and performance. ### Country-level reform priority areas (selected) - Morocco - Restructure SOEs through two draft laws, aiming at eliminating enterprises “deemed no longer essential” and merging others operating within their sector to exploit synergies. - Introducing a “National Agency responsible for the valorization and strategic management of SOEs, as well as the continuous monitoring of their performance”. - Efforts to strengthen the national anti-corruption strategy and to reform public administration. - A stronger role for the Competition Council to address anti-competitive behavior. - Uzbekistan - Improve governance through restructuring of non-financial SOEs. - Strengthen and clearly separate management and supervision. - Potential set-up of an Agency to “monitor and provide financial oversight of all SOEs”. - Barbados - Improve commercial viability and strengthen monitoring and oversight. - Mergers and divestment. - SOE reforms implemented include staff layoffs at SOEs, renegotiation of supplier contracts, an increase in some tariffs (bus fares, water rates), and new levies on sanitation, health services, and tourism. - Poland - Improvements in asset management; putting state assets to the better productive use to earn higher returns. - Ukraine - Introduced governance reforms such as a Management by objective framework in railway sector to enhance supervision and strengthen and centralize internal audit functions. - Sold real estate assets and created a company tasked with property development. - Eventually restructured failing railways through cost-cutting measures (labor shedding), followed by legal reforms to (a) establish a holding company, and (b) allow private sector participation in provision of selected rail services. - Slovak republic - Corporate governance reform that focused on strengthened management practices via a private investment in a large state-owned monopoly (aluminum). - Serbia - Reform areas advised: adopting an ownership policy document; appointing permanent professional management; publishing a comprehensive list of SOEs; expanding capacity to analyze fiscal risks from SOEs. - Belarus - Advised: systematic, risk-based assessment of SOEs’ viability, followed by an actionable plan to guide restructuring (including through strengthened corporate governance, reduced transfers, and better monitoring of contingent fiscal risks). - Enhanced social safety nets to cushion the impact of restructuring on vulnerable groups. - Improving the business environment to ensure a level playing field between public and private companies and facilitate private sector activity. - Georgia - Finalize an SOE Governance Law. - Strengthen the fiscal risk statement to better cover SOE risks, including by reviewing the sectoral classification of SOEs and including a risk analysis of the top 10 SOEs. - Determine the extent to which restructuring of SOEs (including recapitalizations) are reflected in the fiscal envelope. - South Africa - Address SOE weaknesses (e.g., indebtedness, structural inefficiencies and governance-related) via an improved governance framework and promoting competition and private sector participation. - Example: reforms to the macro-critical national electricity company, including clean up efforts for the company's Board and Management. ### Relevance for the Republic of Moldova - Strengthening governance of public enterprises in Moldova is particularly relevant given: - the systemic underperformance of the majority of enterprises, - their elevated financial risk profiles, - dominance across all levels of government and sectors, - the magnitude of assets invested in these companies and managed by the state. ### Non-Bank Credit Organizations and financial inclusion (Moldova) - Overview - Moldova lags many peers in important metrics of financial inclusion, affecting consumers and SMEs. - The high-profile 2014 banking fraud saw a retreat of the banking sector from the credit market; the role of nonbanks, especially Non-Bank Credit Organizations (NBCOs), as a source of credit for consumers and for SMEs has significantly increased since 2014. - Planned transfer of regulation and supervision of Non-Bank Financial Institutions (NBFIs) to the National Bank of Moldova (NBM) provides an opportunity to make regulation and supervision more consistent across banks and NBCOs, while ensuring this reset does not inappropriately tighten credit provision to consumers and SMEs. - Streamlining supervisory framework offers an opportunity to strengthen financial consumer protection as it applies to credit provision. - Challenges to private sector access to credit - Credit extended by commercial banks as a share of GDP at around 22 percent of GDP is the lowest amongst regional peers. - Limited access applies to low volume of lending to both households and SMEs relative to the size of the economy. - Legacy of the 2014 banking fraud led to significant tightening of lending standards, including more stringent loan application processes, submission of quarterly financial reports by commercial borrowers to banks, minimum requirements for loan amortizations, and limitations on refinancings. - Increased cost of credit administration and the need for borrowers to provide collateral and proof of financial history are key barriers for consumers’ and SMEs’ access to bank credit after 2014. - SME-specific burdens: - Failure to comply with quarterly financial reporting results in reclassification of affected loans to a reduced quality category requiring 30 percent provisioning. - NBM’s prudential provisioning requirements are about one-third higher than those stipulated by the International Financial Reporting Standards (IFRS). - SMEs face disproportionate costs preparing financial statements and higher interest rates relative to larger companies. - Consumer-specific burdens: - Onerous paperwork and a complex and lengthy loan approval process. - Online onboarding remains cumbersome and difficult to operationalize. - Low financial literacy among low-income consumers limits access to financial services. - High informality exacerbates credit access challenges: - Transactions by microenterprises and consumers are mostly cash-based and outside formal financial structures. - Tax compliance is low, accounting practices are weak, and record-keeping on income and earnings is limited, impeding assessment of creditworthiness. - The 2014 banking fraud episode (Box 1) - In 2014, US$1 billion disappeared from Moldova’s banking system following coordinated fraudulent financing to related entities; much of these funds were moved outside the country. - The total loss from this incident was estimated to be equivalent to 12 percent of Moldova’s GDP at the time. - The NBM responded by tightening lending standards: requiring quarterly financial statements, commercial loans to start amortizing no later than three months after disbursement, prohibiting refinancing existing loans through a new loan with the same lender, and making the credit application process more stringent. - While appropriate, these actions had unintended consequences: with closure of three banks at the center of the scandal, lending to corporates declined notably. - “Commercial bank lending as a share of GDP went from above 30 pecent to below 15 in recent years.” - NBCO trends and drivers - NBCOs became an increasingly important source of financing, especially for consumers and SMEs. - NBCO lending grew by an average of around 20 percent annually over the last decade, accelerating in 2019 before slowing in the wake of the COVID-19 pandemic. - Between 2010 and 2020, NBCOs’ share of Moldova’s credit market rose from around 7 percent in 2010 to almost one-fifth at end-2020. - At the end of 2019, 570,000 individuals and firms in Moldova had established a credit relationship with an NBCO (about 20 percent of Moldova’s citizens). - NBCOs’ market-share gains were driven largely by consumer lending: NBCO lending to consumers as a share of total lending to consumers grew from under 2 percent in 2010 to around one-third in 2020. - Commercial lending by NBCOs grew from a 2 percent share of all commercial lending in 2010 to 6 percent in 2020. - Structure and concentration of NBCOs - Almost 200 NBCOs on record; 144 NBCOs have equity amounting to less than MDL 10 million (USD 569,000). - Of these, 63 have less than MDL 1 million (USD 56,900) in equity, and 12 have negative equity. - Fewer than 40 NBCOs have assets exceeding MDL 30 million; among those below this threshold, total assets average less than MDL 5 million. - Market concentration: the 30 NBCOs with the largest market shares account for around 90 percent of the market; the remaining NBCOs have market shares below 1 percent, with an average smaller than 0.1 percent. - Around 40 NBCOs had zero loan assets at the end of the first quarter of 2021. - Demand-side drivers for NBCO growth - Growth of NBCO lending was spurred by increasing credit-financed consumption spending amid constrained bank lending and GDP per capita growth averaging 4 percent during 2009–2020. - NBCOs cooperated with retailers to originate consumer loans at point of purchase. - NBCOs offered relatively small-sized loans: average loan size MDL 20,800 (about USD 1,200, or a quarter of GDP per capita) in 2020. *Annex III. Cross Country Experiences at SOE Governance Reforms — Source: IMF (content extracted from the provided PDF).* ### 10. The supply of NBCO credit was driven ### 10. The supply of NBCO credit was driven ### Profitability and market expansion - Banks’ returns on assets and on equity during 2017–19 averaged around 2.5 percent and 14.6 percent, respectively. - NBCOs reported returns on assets and on equity of 9.3 percent and 27.8 percent, respectively, during 2017–19. - Interest income was the key cause of profitability differences: interest rates on NBCO loans averaged almost four times the average rate on bank loans as of early 2021. - NBCOs’ profitability also benefited from high fees, including large penalties for late payments compared to those charged by banks. - High profit margins attracted additional funding, with much of it coming from non-residents. ### Business model, risk-taking, and regulatory arbitrage - Banks and NBCOs are supervised by different regulatory bodies: National Bank of Moldova (NBM) for banks; National Commission for Financial Markets (NCFM) for NBCOs. - Until 2020, loan loss provisioning rates that applied to NBCOs were appreciably lower than those that obtained for banks. - NBCOs targeted the underbanked population and enterprises lacking collateral and proof of income, using simplified application and approval procedures. - Most consumer loans are extended without collateral; some NBCOs use proprietary credit scores; some consumer loan applications can be processed in as little as 10 minutes. - NBCOs have shown greater willingness than banks to extend foreign currency-indexed loans; NBCOs’ share of the foreign currency-linked credit market reached as much as one-fifth of the total foreign currency-linked market in 2020. - Plans are currently underway to transfer supervision of non-banks from the National Commission of Financial Markets to the National Bank of Moldova; the transition is expected to be completed in 2023. ### Consumer protection and financial inclusion concerns - NBCOs have targeted consumers with low financial literacy, raising consumer protection concerns. - Reported abusive practices include not informing consumers of the full cost of loans, burying costs in the ‘fine print’, and not disclosing penalty interest rates and fees for late or non-payments. - Examples include advertising zero interest rates without disclosing all-in loan costs, contrary to legal requirements. - Debt collection practices include visits to family members and threats outlining consequences that have no legal basis. - Banks generally exhibit close compliance with legal provisions governing financial consumer protection (FCP). - Moldova lacks a national financial inclusion strategy. - Moldova’s population declined from 2.98 million in 1992 to 2.66 million in 2019, with a disproportionate decrease outside the greater Chișinău area, contributing to regional financial inclusion gaps. - NBCOs have filled credit provision gaps in rural and informal segments where banks are less active. ### Funding structure and key statistics - Borrowing is the single largest source of funding for NBCOs, equivalent to roughly 50 percent of NBCOs’ assets over the last ten years. - NBCO borrowing rose from MDL 832 million in 2010 to almost MDL 6.1 billion at end-2020. - Non-residents have been the largest source of lending to NBCOs, with an average share of around 60 percent of the total; banks have been the second largest source with an average of 24 percent. - NBCOs cannot accept deposits from the general population (as prescribed by the law that came into effect in early 2020). ### Regulation, capital, and provisioning - A law regulating NBCOs was enacted in 2018; amendments effective April 2020 strengthened the NCFM’s powers. - Minimum capital requirement increased from MDL 100,000 to MDL 1,000,000 from 2021 onwards. - As a result, the number of active NBCOs reduced to 144 as of the end of the first quarter of 2021. - Key provisions include: - Lending by individuals to NBCOs restricted to existing shareholders; such lending must be subordinated to other debt, have a maturity of at least three years, and be for a minimum of MDL 600,000. - The total payments—interest plus fees—that a consumer makes on debt with an original maturity of two years or less, or an original principal value of MDL 50,000 or less, cannot exceed the original loan principal. - Marketing materials and loan documentation must contain the all-in cost of a loan, inclusive of all fees. - NBCOs must hold capital of at least 5 percent of assets, or a 10 percent minimum if they borrow from banks. - If information from credit bureaus is not used in the provision of credit, a loan must be classified as substandard from the outset. - Loans 30 days or more overdue are subject to higher provisioning rates than for banks. ### Supervision fragmentation and systemic visibility - Fragmented supervision across different entities limits visibility over systemic vulnerabilities, including true borrower leverage. - Financial stability measures covering banks, NBCOs, and insurers must be agreed by the Financial Stability Council, which has led to implementation delays in the past. - Supervisory fragmentation created space for regulatory arbitrage. - Oversight responsibilities for consumer protection are distributed across the NCFM and the Consumer Protection Agency (CPA); the CPA lacks capacity to address complex financial complaints and overlap has led to duplicative and uncoordinated efforts. ### Credit bureaus and overindebtedness risk - Three credit bureaus operate in Moldova, with a fourth slated to be approved for operation soon. - Each bank must report to at least one bureau; NBCOs and SCAs have been required to do the same since April 2021. - Credit bureaus are supposed to exchange information with each other since July 2021, but it is unclear whether full and timely information sharing is taking place. - NBCOs are not required to consider information from credit bureaus, and no overall limits on debt-to-income ratios exist, implying risk of overindebtedness that could have systemic implications during a downturn. ### Asset quality, buffers, and vulnerabilities - NBCO asset quality deteriorated during 2020: the share of loans between 31 and 90 days past due increased by more than 10 percentage points since March 2020 through end-2021Q3. - Shares of loans reflecting delinquencies above 90 days have increased but remained relatively contained. - NBCOs tend to extend loans with relatively short maturities—between 1 and 2 years—limiting prolonged exposure. - Aggressive collection practices leveraging family connections have helped limit loan losses. - At the aggregate level, total buffers (including equity capital and loan loss provisions) account for around half of outstanding loans, though differences exist across companies. - With NBCOs unable to accept deposits, a potentially important source of systemic risk from deposit-funded risky NBCOs is removed. ### Funding concentration, FX exposure, and contagion risks - Heavy reliance on borrowing and especially on non-resident funding exposes NBCOs to maturity mismatch risks and vulnerability to changes in global financial conditions (“risk on, risk off”). - Nonresident funding is concentrated: a handful of entities receive the lion’s share of nonresident lending. - Large flows of non-resident funds into NBCOs could give rise to AML/CFT concerns; while the NCFM has received UBO information from NBCOs, concerns remain. - NBCOs’ reliance on nonresident funding potentially exposes them to currency mismatch risks, partially mitigated by their growing share of foreign currency-indexed credit. - Bank lending to NBCOs is small in aggregate—under 2 percent of the banking system’s total assets as at the end of the first quarter of 2021—but concentrated: - In three banks, lending to NBCOs constitutes 17.6, 14.9, and 10.4 percent of their regulatory capital respectively (as of end-2020Q1). - 93 percent of bank lending to NBCOs goes to the top 14 NBCOs by assets. - In five NBCOs, loans from a single bank exceed 60 percent of their bank borrowing. - The bilateral complexity and FX risks imply supervisors for banks and NBCOs should cooperate when assessing institution-specific and systemic risks. ### Stress-test findings and scenarios - A stress test suggests that most NBCOs have sufficient capital to weather a number of downside scenarios. - Among the largest NBCOs accounting for around 80 percent of total market share, significantly deteriorated loan portfolios appear concentrated in a handful of players. - Some NBCOs will potentially fail during a downturn, with some ending up with their entire equity wiped out; these institutions are estimated to account for around 6 percent of the total market share. - Adverse market events can be expected to increase leverage at the industry level drastically, suggesting increased vulnerability during prolonged downturns. *Source: 1mdaea2022003 — Chapter 10, "The supply of NBCO credit was driven".* ### Box 2. NBCOs: A Stress Test ### Box 2. NBCOs: A Stress Test ### Stress-test design - Due to data limitations, dynamic multiperiod simulations of adverse events affecting NBCOs are not feasible in the context of this paper. The exercise instead considers the potential impact of different degrees of deterioration of NBCOs’ loan portfolio under selected downside scenarios. - Definition used: loans delinquent for a duration exceeding 90 days are referred to as “bad loans”. This category was chosen since loans in this category require 100 percent provisioning compared to loans subject to shorter periods of delinquency. - Higher levels of “bad loans” reduce NBCOs’ equity because they require increased loan loss provisioning. - The stress test focused for the most part on the 15 and 30 largest NBCOs, which accounted for around 80 and 90 percent of NBCOs’ total market share, respectively, as of at the end of the first quarter of 2021. - The impact on individual banks’ equity is estimated and aggregated. Charts present: NPLs > 90 Days Under Various Scenarios (In percent of equity); NPLs Exceeding 90 Days Under Various Scenarios (In percent of equity); Equity After Provisioning Under Various Scenarios (In millions of lei); Debt-Equity Ratio Under Various Scenarios (In percent). ### Downside scenarios considered - Scenario 1: 25 percent of loans delinquent for 60 to 90 days turn into “bad loans” - Scenario 2: 50 percent of loans delinquent for 60 to 90 days turn into “bad loans” - Scenario 3: 75 percent of loans delinquent for 60 to 90 days turn into “bad loans” - Scenario 4: “Bad loans” double in level - Scenario 5: “Bad loans” triple in level ### Key findings from the stress test - At the aggregate level, NBCOs in Moldova have sufficient equity to absorb further deterioration in asset quality, with the exception of extreme losses (scenarios 3 and 5). - The impact of the downside scenarios is mainly concentrated in NBCOs with high levels of lower-quality loan portfolio or limited equity currently. - In the most extreme scenarios, a few NBCOs would see their equity wiped out, but these NBCOs are likely to account for a relatively small portion of the market share (around 6 percent). - While NBCOs at the aggregate level are expected to be able to withstand most economic downturns, extreme shocks would increase leverage at the industry level measurably. ### Policy considerations and recommendations (transition of oversight and regulation) - Plans for making the NBM as the sole regulator of banks and non-banks mark a step in the right direction: a single regulator can increase consistency and speed in implementation of macroprudential measures, improve information sharing, and better pool data and IT resources. - Forthcoming changes to the regulatory and supervisory regime governing NBCOs should not lead to reduced access to credit for financially underserved consumers and firms; subjecting NBCOs to the same regulations governing banks could jeopardize their business model and risk driving borrowers to shadow lenders. - NBCO supervision and regulation should be tailored to that particular sector, distinct from the regime that governs banks. ### Key principles for the transition of NCFM’s oversight responsibilities to the NBM - Regulate and supervise NBCOs distinctly from banks, to preserve their positive impact on credit provision. - Do not subject NBCOs’ borrowers to the same strict credit approval and reporting requirements that some bank borrowers are subject to. - Develop capital and provisioning requirements that reflect the higher risks of NBCO loans. - Develop macroprudential and systemic risk-related requirements that are aligned between banks and NBCOs. - Develop procedures that help to track and analyze certain NBCO-specific risks, including risks caused by FX- and maturity-misalignments between loans and funds, sudden stop risks from foreign funding, as well as risks emanating from the NBCO-bank nexus. - AML/CFT concerns should be addressed by analyzing UBO and the ultimate source of foreign funding of NBCOs. - Prioritize on-site supervision once pandemic-related restrictions have been removed. Supervision should put appropriate emphasis on governance and risk management of the institutions. ### Consolidation and financial inclusion - Further consolidation of the NBCO sector can be considered to increase resilience and efficiency: many NBCOs are undercapitalized and have inadequate capital and negligible assets, which may strain regulators with limited resources. - Consolidating the market into a smaller number of players can simplify supervision and ensure borrowers are served by NBCOs with adequate resources, expertise, and capacity. - A significant reduction in the number of NBCOs—for example to the 30-40 large players that account for 90 or more percent of market share—can be done while maintaining a healthy level of competition. - Note on concentration: Consolidating the industry to the top 30 players would have a limited impact on market concentration as measured by the Herfindahl Hirschman Index, which currently stands at around 9 percent and would increase to 11 percent, still significantly below, for instance, the 25 percent threshold used in the US in assessing oligopolistic conditions. - The authorities should develop a targeted approach on FCP that should be integrated into a national financial inclusion strategy. Data on indebtedness should be compiled and analyzed to allow sharing with the CPA or other agencies that partake in consumer protection. Common requirements on information sharing across banks, NBCOs, SCAs, and credit bureaus, as well as caps on debt-to-income ratios, could become more easily enforced under a common supervisor. *Source: Box 2. NBCOs: A Stress Test* --- _Source: https://www.imf.org/-/media/files/publications/cr/2022/english/1mdaea2022003.pdf_