## _cr07258

## Source details

**Canonical URL:** [_cr07258](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2007/_cr07258.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2007/_cr07258.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2007/_cr07258.pdf.json)

---

### Macroeconomic management of scaled-up foreign aid — objective and context
- Objective: assess macroeconomic challenges of managing scaled-up foreign aid in Mozambique, focusing on measures and policies to absorb and efficiently use higher aid flows to help achieve the Millennium Development Goals (MDGs).
- Contextual facts:
  - Aid inflows in Mozambique are "about 15 percent of GDP."
  - Three approaches to preparing scaling-up scenarios considered; chapter adopts approach suited to Mozambique given absence of explicit MDG costing and large, rising aid flows.
- Paper organization:
  - (i) analyze past impact of aid in Mozambique;
  - (ii) present illustrative scaling-up scenarios;
  - (iii) identify challenges for Mozambique.

### Methodology for assessing fiscal effects of aid
- Fiscal response estimated using a fiscal response model (FRM) within a vector autoregression (VAR) framework following Osei, Morrissey, and Lloyd (2005).
- VAR specification and data:
  - Variable ordering: foreign aid (AID), government expenditure (Expenditures), tax revenue (Tax), change in net credit to the government from the banking system (NCG).
  - Model variants: aid disaggregated into grants and loans; aid split into project aid and budget support (BUDSUP); expenditures disaggregated into capital and current components.
  - Sample and measurement: quarterly data over the period 1996Q1 to 2006Q3; all variables in constant 2004 prices expressed in millions of new meticais (MT).
- Time-series properties and estimation choices:
  - Variables are I(1) except for aid and NCG which are stationary I(0).
  - Cointegration tests did not identify a significant cointegrating vector; series are first differenced for estimation.
  - Granger causality tests indicate aid Granger-causes fiscal variables but fiscal variables do not Granger-cause aid—supporting interpretation of aid shocks as exogenous.
  - Structural shocks recovered using Cholesky decomposition of the VAR residual covariance matrix.

### Key empirical findings on fiscal effects of aid
- Magnitudes and timing:
  - A one standard deviation shock in aid corresponds to about MT 1500 million (this would represent about 1 percent of GDP in 2004).
  - The cumulative impulse response to a one standard deviation aid shock shows a full pass-through to total expenditures within four quarters.
  - Budget support initially reduces NCG and then begins to pick up with a slight lag of a quarter or two, leading to full spending within 5–6 quarters.
- Composition of spending:
  - Public expenditures tend to increase more than the original shock to aid.
  - The aid shock results in more capital expenditures; only about one-third of the aid finances current spending within a year.
  - Caveats: donor-financed projects are classified as capital spending in fiscal accounts although a significant share may be on wages and goods and services; lack of quarterly functional spending breakdown limits deeper composition analysis.
- Revenue and financing:
  - Aid inflows are associated with an increase in domestic taxation.
  - The increase in government taxation counters concerns that grants substitute for domestic revenues and dampen revenue collection efforts.
  - The impact of aid on NCG is less clear, possibly reflecting expenditure smoothing and/or issuance of domestic debt.
- Robustness:
  - Results largely unaltered when considering loans versus grants, though some differences in the NCG response emerge.
  - Alternative data treatments (detrending) yielded similar results to first-differencing.

### Implications for macro policy from aid dynamics
- Aid tends to be spent rapidly and can finance both capital and current expenditures, with a stronger effect on capital spending in the observed period.
- Increased aid is associated with higher domestic tax mobilization rather than revenue displacement.
- Budget support dynamics differ from project aid: initial reduction in NCG followed by absorption over 5–6 quarters.
- Policy attention required on:
  - Public financial management and classification of expenditures to correctly interpret capital vs. current content of donor projects.
  - Domestic debt management and smoothing behaviors affecting NCG.
  - Measures to ensure aid supports productive investments while avoiding macroeconomic absorption problems such as inflation, crowding-out of the private sector, and loss of competitiveness.

### Aid spending, real exchange rate (REER), and exports — findings
- Government spending (both current and total expenditures) has tended to appreciate the fundamental equilibrium real exchange rate, although some effects mitigated by trade liberalization and negative terms of trade movements.
- Mozambique’s total export shares in world trade increased mainly due to megaproject exports; nonmegaproject exports’ share of world trade and ratio to GDP remained roughly constant.
- Econometric results suggest the REER may have been overvalued at times of tight exchange rate management; attribution to aid response is difficult.
- Supply-side effects of aid (health and infrastructure investments) could mitigate REER appreciation, but visible evidence that these supply-side effects are strong enough is lacking.

### Short-run dynamics and aid absorption — VAR and DSGE evidence
- Aid absorption definition: IMF (2005a) defines aid absorption as extent to which a country’s nonaid current account deficit widens in response to an increase in aid inflows; full aid absorption equivalent to unchanged level of international reserves under a closed capital account.
- Six-variable VAR (Q1 1996–Q3 2006) ordering: foreign aid, real GDP (or real nonmegaproject exports), inflation, reserves, domestic debt, nominal exchange rate; structural shocks recovered via Cholesky.
- Impulse responses (quarterly):
  - Aid shocks (mostly spent) initially resulted in nominal exchange rate appreciation, and possibly lower output.
  - Inflation declines initially but picks up gradually.
  - Using real nonmegaproject exports shows only modest contractions of the export sector in response to aid shocks.
  - Most foreign exchange associated with aid inflows appears to have been sold by the central bank within five quarters of the initial aid shock, facilitating a “spend and absorb” response.
  - Central bank smoothed foreign exchange sales in short term through temporary net domestic debt issuance (sterilization), helping keep prices at bay and minimize crowding-out of private sector credit.
  - Pattern contrasts with most other non-CFA SSA countries that relied more heavily on sterilization, contributing to unfavorable debt dynamics and probably greater crowding-out.

### Illustrative scaling-up scenarios and growth prospects
- Purpose: identify macro implications and key measures to absorb higher aid and ensure efficiency.
- No full MDG costing; considers arbitrary but sustained increases of external finance rather than MDG-targeted scenarios.
- Focus: macro implications and policy trade-offs of spending and absorbing further scaling-up of foreign aid amounting to more than 15 percent of GDP at present.
- Growth prospects from growth accounting and studies:
  - about 7 percent growth in the medium term (baseline in staff report)
  - 5–6 percent in the long-term, a level sufficient to halve the poverty rate by 2015, assuming continuation of strong productivity growth.
- Upside from higher public investment depends on ratio of current-to-capital spending and on crowding-in of private investment and spillovers from infrastructure and human capital accumulation.
- Possible negative consequences from further scaling-up:
  - adverse effects on private investment reflecting diminishing returns,
  - challenges to human capital quality amid rapid aid-financed expansion of the school network.
- Recent evidence: primary education completion rates and quality relatively low; high pupil-to-teacher ratios and growing number of untrained teachers.

### DSGE scaling-up simulations — scenarios and assumptions
- DSGE model simulated (Peiris and Saxegaard (2007)) for a persistent aid shock with autocorrelation coefficient of 0.7 raising aid by 2 percent of steady-state GDP.
- All scenarios assume full spending and absorption of foreign aid.
- Three scenarios:
  - Scenario 1: no productivity spillovers from public investment.
  - Scenario 2 (baseline): modest productivity spillovers from public investment.
  - Scenario 3: high productivity spillovers from public investment.
- Productivity function follows Pratti and Tressel (2006): productivity is an increasing function of the size of public investment expenditure; higher productivity can arise from greater spillovers or a decrease in the ratio of current-to-capital spending.

### Policy implications and trade-offs from scaling-up simulations
- Spending and absorbing aid is generally preferred, but can entail significant loss of competitiveness and lower exports and real GDP growth if unmitigated.
- Priority reforms and analyses include:
  - improving PFM systems to support effective use of aid and enable supply-side responses,
  - ensuring public spending composition favors capital formation and productivity-enhancing investment (infrastructure, human capital),
  - careful management of foreign exchange flows to limit short-term REER misalignment while avoiding excessive sterilization that could harm debt dynamics and crowd out private credit.

### Productive public investment and Dutch-disease mitigation
- Productive government investment can mitigate Dutch-disease effects by reducing REER appreciation pressures and eliciting stronger export performance (impulse responses in Figure 5).
- Consumption and output are unambiguously higher with greater productivity spillovers.
- Inflation pick-up following an initial decline would be milder with greater productivity spillovers.
- Outcomes depend critically on composition of government expenditure and efficiency of public service delivery.
- Modeling caveat: analysis assumed productive government spending affects transitional dynamics but not the steady-state; permanent increases in aid affecting steady-state not modeled.

### Priority sectors, scaling-up experience, and agriculture
- Since 2000, about 65 percent of scaled-up spending allocated to priority sectors in PARPA I (education, health and infrastructure), contributing to:
  - doubling the number of children in primary school,
  - reducing infant and maternal mortality,
  - beginning provision of Anti-Retro-Viral (ARV) treatment for HIV infection.
- Large human capital and infrastructure gaps remain; need to expand secondary education and address communication and transportation infrastructure.
- Agriculture and rural development priorities:
  - security of land-tenure,
  - adoption of new technologies associated with cash crops,
  - diversification into profitable cash crops and contract farming,
  - use of productivity-enhancing inputs (fertilizers, seeds, irrigation),
  - development of rural credit and input markets,
  - improved road infrastructure, storage facilities and extension services.
- Mozambique has potential for high-value vegetable crops and floriculture for export to Europe with better quality standards and marketing infrastructure.

### Public financial management (PFM) and SISTAFE rollout
- Strengthening PFM and public sector reforms vital to ensure resources reach productive priority sectors, including subnational level.
- Government transitioning towards program-based budgeting; pilot with a few line ministries in the 2008 budget suggested.
- e-SISTAFE rollout and projected outputs:
  - e-SISTAFE rolled out to most line ministries at central and provincial level by end-year 2006.
  - e-SISTAFE extension to remaining ministries and state organs by end-December 2007; phased rollout to districts (initially to 27 districts by end-October 2007).
  - New e-SISTAFE functionalities in 2007: Phase II of budget formulation, payroll and pension, revenue collections modules.
  - Integrated payroll database to enable salary payments via direct budget execution in e-SISTAFE by end-June 2007.
  - Projected outputs (2006–2009) include: financial and budget execution in operation in all ministries and levels; payroll and pension module implemented; budget formulation module implemented (Phase II); asset management, procurement interface, ATM revenue management, debt management, internal audit modules implemented; CPD and Project Implementation Unit operating as effective and sustainable units.

### Exit strategy and fiscal sustainability under scaling-up
- Objective: gradually increase domestic revenue to enable orderly exit from donor dependence and finance at least recurrent spending from own resources over next eight years (post-2015 donor reversion or decrease scenario).
- Illustrative fiscal framework: assumes spending of an additional US$200 million from 2008–15 compared to baseline; increased expenditures assumed divided between current and capital spending in a 1:2 ratio.
- Projected revenue and sustainability:
  - Projected revenue ratio of about 19½ percent of GDP in 2015 (increasing from 14 percent of GDP in 2006) would cover scaled-up recurrent spending and a portion of public investment under an annual US$200 million scaling-up.
  - Even if all scaled-up spending financed through concessional external borrowing, external debt-to-export ratio would remain below indicative performance-based debt burden thresholds (e.g., NPV of debt-to-exports ratio of 150 percent for Mozambique).
  - Low level of domestic public debt: about 8 ½ percent of GDP.
- Fiscal policy implications:
  - Spending plans need regular reassessment in light of growth and Dutch-disease effects and realized revenue outcomes.
  - Any nonconcessional financing of large infrastructure projects should be considered case by case based on economic return, impact on debt sustainability, and effects on donors’ financing decisions.
  - A prudent external debt management strategy of avoiding non-concessional external borrowing is assumed to continue.

### Revenue mobilization and tax administration reforms
- Varsano et al. (2007) estimated Mozambique’s tax effort at about 50 percent (ratio of actual collection to taxing capability), suggesting scope for raising revenues up to nearly 22 percent of GDP.
- Schenone (2004) found the difference between potential and actual collection due to noncompliance (evasion) is much larger than that due to tax law features (exemptions and nonassessments).
- Policy implication: greater gains likely from measures that increase tax administration efficiency rather than from raising tax rates.
- Next phase of revenue administration reforms (2007–10) will focus on implementing the central revenue authority (ATM) strategic plan:
  - ATM establishment stages: transition period to end-2007; gradual integration of tax and customs agencies during 2008; further strengthening and consolidation in 2009 and 2010.

### Natural resources, megaprojects, and fiscal trade-offs
- Megaprojects in mining and petroleum have placed Mozambique on global FDI map, but fiscal revenues to date low due to generous fiscal incentives used during recovery from civil war.
- Authorities recognize fiscal contribution could be substantially increased by reducing generous tax exemptions.
- New fiscal regime for mining sector under preparation; capacity building to negotiate effectively with multinationals using financial models and model contracts.
- Present pipeline projects, including Moatize coal-mine project, could yield revenues in the range of 2–4 percent of GDP beyond 2010.
- Further gas and oil discoveries would result in significant long-term revenue contributions.

### Petroleum sector developments and fiscal implications
- Pande-Temane gas field installations, processing facilities, and gas export pipeline to South Africa were commissioned in 2004 amounting to some US$1 billion FDI in all.
- Government secured more than US$300 million in petroleum exploration work commitments from the Rovuma Basin round, adding to perhaps US$200 million in planned drilling and evaluation activities on earlier contract areas.
- Pande-Temane project currently in cost recovery and high debt service; government revenues modest but set to increase as production ramps up and investors recover development costs.
- Joint venture partners considering possible 50 percent expansion of annual gas sales volume to around 180mmGj.
- Under Petroleum Production Agreement, joint venture partners allowed 25 percent depreciation, reducing income tax revenues in early years in addition to tax exemptions.
- Authorities reviewing petroleum fiscal regime intending to put a comprehensive fiscal regime in general tax law to avoid case-by-case negotiations.

### Aid scaling-up outcomes and policy responses
- Mozambique has fully spent and mostly absorbed scaled-up foreign aid ranging between 10–20 percent of GDP over the last decade.
- Outcomes achieved during scaled-up aid period include:
  - Doubling of the number of children in primary school.
  - Reduction in infant and maternal mortality.
  - Beginning provision of ARV treatment for HIV infections.
  - Sustaining economic growth of eight percent per year on average.
  - Reducing the poverty headcount index from 69 percent in 1997 to 54 percent in 2003.
- A modest scaling-up of foreign aid inflows (about 2 percent of 2008 GDP) could continue to be spent and absorbed aimed at achieving the MDGs.
- Risks of scaling-up:
  - Spending increases tend to appreciate the REER and potentially hurt export sector and long-run growth (Dutch-disease).
  - If scaling-up is significantly larger or less persistent, or microeconomic capacity constraints bind, macroeconomic variables and volatility may be too large for consumer welfare or long-term growth.
- Recommended policy responses:
  - Reach productive priority sectors (agriculture, education, health, infrastructure in PARPA II) to elicit supply response and mitigate Dutch-disease.
  - Strengthen PFM systems and public sector reform to improve efficiency and service delivery.
  - Complement productive public investments with second-generation reforms to strengthen business environment and agricultural production.
  - If necessary, smooth expenditure pattern and save part of aid in international reserves to spend and absorb later.

### Financial sector status, weaknesses, and reform priorities
- Financial sector development is a main objective of PARPA II’s economic development pillar.
- By end-2002 banking system dominated by foreign banks and highly concentrated: five largest banks accounted for 96 percent of total deposits at end-2002.
- System increasingly dollarized: foreign currency deposits and loans accounted for 51 percent and 70 percent of their respective totals in 2002.
- Nonperforming loans (NPL) high at 21 percent at end-2002.
- Real lending rates in Meticais in 1999–2002 averaged 17.5 percent.
- Causes of high lending rates and poor intermediation: high NPLs, weak supervision, inadequate reporting, high overheads, high reserve requirements, limited competition.
- Legal and institutional impediments: weak debt-enforcement mechanisms, incomplete collateral and bankruptcy laws, lack of comprehensive credit information.
- Mozambique deepening reforms by strengthening institutional regulatory architecture and addressing financing constraints identified in 2003 FSAP.

### High and volatile real treasury bill rates — drivers and institutional roots
- Drivers: significant supply shocks, large fiscally-induced demand shocks (unanticipated influx of external grants and loans), weaknesses in monetary framework and liquidity management.
- Monetary framework weaknesses:
  - Lack of transparency, insufficient BM communication, inconsistent use of instruments.
  - BM reported positive capital and operating profits but net worth and profits negative once valuation adjustments accounted for.
  - BM losses mainly from accumulated foreign exchange valuation losses on external liabilities and operating losses from sterilization not transferred to budget.
  - BM’s fragile financial position risked operational independence.
- Contract enforcement and securitization problems: inefficient courts, weak property rights, costly procedures to register and transfer property.
- Credit information shortfalls: public credit registry covered about 11,500 borrowers with only three full-time staff; borrowers increased to about 99,000 in 2006 but quality remained below regional best practices.

### FSAP policy recommendations and FSTAP strategy (selected)
- FSAP recommendations (key items):
  - Strengthen banking supervision and reporting standards; adopt IFRS; modernize supervisory functions; review bank resolution framework.
  - Ensure consistency between instruments and monetary targeting; introduce repos/reverse repos; enhance central bank capacity for liquidity forecasts; improve central bank communication and operational independence arrangements.
  - Strengthen public debt management by improving reporting and formulating a debt management strategy; restructure Public Debt Unit.
  - Institutional lending environment reforms: amend BM law and Banking Law; modernize Notary Office, Public Commercial Registry, Civil Procedure Code; consider specialized commercial court.
  - Capital market, pensions, insurance: review capital controls; build supervisory capacity; improve accounting and disclosure for insurance and pensions; separate supervisory and administrative functions of INSS.
  - Microfinance: review microfinance legislation.
- Mozambique’s FSTAP goals:
  - Strengthen BM and supervisory capacity; improve financial accountability and IFRS adoption; strengthen public debt management; improve money and government bond market efficiency and depth; support micro and rural finance.
- FSTAP implementation: one project with ten components, PIU established; donor coordination across IDA, AfDB, DFID, SIDA, KfW/GTZ, IMF and others.
- Monitoring and evaluation: PIU prepares quarterly monitoring tables and progress reports; results framework and Performance Assessment Framework (PAF) used for evaluation; medium-term review expected end-2008.

### Banking restructuring, supervision, and reform impact
- Bank privatization process in mid-1990s lacked due diligence; recapitalizations of Banco Austral and Banco Comercial de Moçambique amounted to 6 percent of GDP during 2000–02.
- Government committed to return banks to private ownership; only minority shareholdings remain.
- Comprehensive diagnostic reviews by international firms (IFRS) completed in 2005 endorsed overall soundness.
- New Financial Institutions Law implemented in 2004 strengthening central bank supervisory powers; IFRS adoption by BM in 2006 and by commercial banks in 2007.
- IMF technical assistance priorities include: making monetary policy more effective; strengthening supervisory and regulatory framework for full compliance with international standards; developing interbank foreign exchange and government debt market capacity.
- Monetary policy and public debt management reforms: BM adopted money targeting using base money; planned introduction of repos in 2007; ongoing work to separate temporary and structural liquidity; need to ensure operational independence via memorandum of understanding with Ministry of Finance expected in 2007.

### Financial soundness, microfinance, and access to finance
- Reforms contributed to improvements: NPLs declined markedly; bank profitability recuperated; decline in financial dollarization and vulnerability observed.
- Real lending rates and spreads on a declining trend since 2002; private credit growth picking up, particularly in domestic currency.
- MFIs’ outstanding loan portfolio and clients rose about five- and seven-fold from 2000 to 2005; consolidation: three institutions control more than 68 percent of active loan portfolio.
- Microfinance outreach limited in rural areas: about 62 percent of MFIs and 50 percent of microborrowers concentrated in Maputo-Matola corridor; rural areas account for less than 12 percent of borrowers and only 5 percent of total active portfolio.
- Access to affordable credit remains a top constraint: more than 70 percent of firms do not have access to a bank loan or overdraft facilities (higher for SMEs).

### Improving access to affordable finance — priorities
- Strengthen lending environment and outreach of banks, MFIs, and private financial services to SMEs and rural areas.
- Identify quick wins: expand credit registries, explore leasing and factoring agreements.
- Capacity building to strengthen monetary policy formulation and deepen public debt markets; implement RTGS and repos to deepen government securities market.
- Near-term priorities: introduce risk-based supervision, roll out IFRS standards to corporate sector, secure sustainability of national pension fund through actuarial study, strengthen supervisory framework for nonbank financial sector.

### Export performance, megaproject dominance, and competitiveness
- Mozambique’s average share of world exports more than doubled in 2001–06 relative to 1991–00.
- Terms of trade improved on average by nearly 7 percent a year in 2001–06, up from 0.5 percent in 1991–2000, mainly due to aluminum price increases from 2003.
- Megaproject dominance:
  - All megaproject exports (electricity, aluminum, gas) accounted for more than 70 percent of total exports in 2006; aluminum alone accounted for 58.6 percent of total exports.
  - Traditional exports grew by 4 percent a year; mega-project exports grew by 10.4 percent.
  - Since 2000 (Mozal full capacity), megaproject-related exports surged; in 2004 Mozal’s exports accounted for 72 percent of total exports.
- REER assessment: REER relatively stable over past decade; econometric results suggest potential overvaluation of about 10 percent with statistical uncertainty.
- Megaprojects benefit from favorable policies: tax exemptions, EPZ status, duty-free imports, on-site customs and tax incentives.
- Protection examples:
  - Export tax on raw cashew: 18 percent.
  - Import tax on sugar: 25 percent.
- Tariff regime:
  - Four nonzero bands: 2.5, 5, 7, and 20 percents.
  - Trade-weighted average tariff: 9.1 percent; simple average tariff: 12.1 percent.
  - Mozambique reduced maximum tariff from 35 percent in 1999, to 25 in 2002, to 20 percent in 2006.
- Trade openness increases:
  - Imports to GDP: 1991-00: 14.5; 2001-06: 29.2.
  - Exports to GDP: 1991-00: 38.1; 2001-06: 43.4.
- Enterprise survey indicators (percent of sample):
  - Exporting firms: 2002: 11.5; 2006: 13.9.
  - Importing firms: 2002: 43.2; 2006: 63.9.
  - Innovate: 2002: 25.5; 2006: 23.4.

### Trade policy, market access, and competitiveness strategy
- Mozambique has preferential market access:
  - United States under the African Growth and Opportunity Act.
  - European Union under the Cotonou Agreement and Everything but Arms initiative.
  - Preferential access to South Africa under SADC Trade Protocol.
- Exports destination structure (percent of total exports):
  - Exports to industrial countries: 1991-01: 57.5; 2002-05: 66.2.
  - Exports to European Union: 1991-01: 40.1; 2002-05: 63.9.
  - South Africa absorbs about 16 percent of Mozambique’s exports.
- Mirror trade discrepancies suggest customs weaknesses and smuggling on some borders.
- Summary: external competitiveness gains concentrated in a few protected sectors (megaprojects); non-megaproject businesses face low tariff and minimal nontariff barriers relative to SSA peers but weak business environment constrains competitiveness, especially for SMEs.
- Policy priority: make business environment more friendly, reduce burden of trade regulations and customs procedures, and complement broad-based MFN liberalization with stronger regulatory environment to allow firms to exploit market access.

### Literature on liberalization and Mozambique’s strategy
- Literature: broad-based multilateral liberalization generally superior to discriminatory liberalization; overlapping RTAs increase administrative costs.
- Yagci and Aldaz-Carroll (2004) estimate SSA would lose 0.7 percent of GDP if all SSA countries negotiated bilateral FTAs with Australia, Canada, EU, Japan, New Zealand, and the United States.
- Mozambique’s market access strategy: membership of SADC only; suspended participation in COMESA in 1995 to avoid overlapping agreements.
- Revenue risks of trade agreements:
  - Alfieri et al (2006) estimate potential revenue losses: 22 percent of revenues on adoption of SACU; 28 percent by joining SADC; 39 percent by achieving MFN status.
- Policy recommendation: prioritize business environment reforms over pursuing complex regional trade integration when business environment is binding constraint.

### Lessons learned and near-term challenges in financial sector reform
- Lessons:
  - Accurate diagnostics and prioritization enabled targeted interventions and political capital for reformers.
  - Institutional arrangements that ensure government ownership and coordination reduced transaction costs.
  - Capacity building and measurable performance indicators essential for sustaining reforms.
- Near-term challenges:
  - Accelerate reforms to improve institutional lending environment, rural outreach by banks and MFIs, and reduce cost of credit for SMEs and agriculture.
  - Deepen interbank, secondary debt, and foreign exchange markets to improve monetary policy transmission.
  - Strengthen capacity of regulators for pension and insurance sectors and finalize outstanding legal and institutional reforms (property registries modernization, Insolvency and Bankruptcy Law, revised Civil Procedure Code).

*Source: excerpts from IMF staff report content unit _cr07258.*

### 2007. The views expressed in this document are those of the staff team and do not necessarily reflect

### MACROECONOMIC MANAGEMENT OF SCALED-UP FOREIGN AID IN MOZAMBIQUE

### Introduction and objective
- The chapter assesses the macroeconomic challenges of managing scaled-up foreign aid in Mozambique, with focus on measures and policies to absorb and efficiently use higher aid flows to help achieve the Millennium Development Goals (MDGs).
- Contextual points from the source:
  - Aid inflows in Mozambique are "about 15 percent of GDP."
  - Three approaches to preparing scaling-up scenarios are discussed; the chapter adopts an approach suited to Mozambique given the absence of an explicit MDG costing and the large, rising aid flows.
- The paper is organized to: (i) analyze the past impact of aid in Mozambique; (ii) present illustrative scaling-up scenarios; and (iii) identify challenges for Mozambique.

### Methodology for assessing fiscal effects of aid
- Fiscal response estimated using a fiscal response model (FRM) within a vector autoregression (VAR) framework following Osei, Morrissey, and Lloyd (2005).
- VAR specification and data:
  - Variable ordering: foreign aid (AID), government expenditure (Expenditures), tax revenue (Tax), change in net credit to the government from the banking system (NCG).
  - Model variants: aid disaggregated into grants and loans; aid split into project aid and budget support (BUDSUP); expenditures disaggregated into capital and current components.
  - Sample and measurement: quarterly data over the period 1996Q1 to 2006Q3; all variables in constant 2004 prices expressed in millions of new meticais (MT).
- Time-series properties and estimation choices:
  - Variables are I(1) except for aid and NCG which are stationary I(0).
  - Cointegration tests did not identify a significant cointegrating vector; series are first differenced for estimation.
  - Granger causality tests indicate aid Granger-causes fiscal variables but fiscal variables do not Granger-cause aid—supporting the interpretation of aid shocks as exogenous to the system.
  - Structural shocks recovered using Cholesky decomposition of the VAR residual covariance matrix.

### Key empirical findings on fiscal effects of aid
- Magnitudes and timing:
  - A one standard deviation shock in aid corresponds to about MT 1500 million (this would represent about 1 percent of GDP in 2004).
  - The cumulative impulse response to a one standard deviation aid shock shows a full pass-through to total expenditures within four quarters.
  - Budget support initially reduces NCG and then begins to pick up with a slight lag of a quarter or two, leading to full spending within 5–6 quarters.
- Composition of spending:
  - Public expenditures tend to increase more than the original shock to aid.
  - The aid shock results in more capital expenditures; only about one-third of the aid finances current spending within a year.
  - Caveats: donor-financed projects are classified as capital spending in fiscal accounts although a significant share may be on wages and goods and services; lack of quarterly functional spending breakdown limits deeper composition analysis.
- Revenue and financing:
  - Aid inflows are associated with an increase in domestic taxation.
  - The increase in government taxation counters concerns that grants substitute for domestic revenues and dampen revenue collection efforts.
  - The impact of aid on NCG is less clear, possibly reflecting expenditure smoothing and/or issuance of domestic debt.
- Robustness:
  - Results are largely unaltered when considering loans versus grants, though some differences in the NCG response emerge.
  - Alternative data treatments (detrending) yielded similar results to first-differencing.

### Implications highlighted in the chapter (linked to macro policy)
- Aid tends to be spent rapidly and can finance both capital and current expenditures, with a stronger effect on capital spending in the observed period.
- Increased aid is associated with higher domestic tax mobilization rather than revenue displacement.
- Budget support dynamics differ from project aid: initial reduction in NCG followed by absorption over 5–6 quarters.
- These fiscal dynamics imply that scaling-up aid requires attention to:
  - Public financial management and classification of expenditures (to correctly interpret capital vs. current content of donor projects).
  - Domestic debt management and smoothing behaviors that affect NCG.
  - Measures to ensure aid supports productive investments while avoiding macroeconomic absorption problems such as inflation, crowding-out of the private sector, and loss of competitiveness.

*Source: Selected Issues — Republic of Mozambique (Prepared by Victor Lledó, Shanaka J. Peiris (AFR), and Eteri Kvintradze (PDR); Approved by the African Department; June 1, 2007).*

### 9. The spending of foreign aid could undermine competitiveness, although such

### 9. The spending of foreign aid could undermine competitiveness, although such effects can be ameliorated depending on the nature of the supply side response over the long-term

### Aid spending, real exchange rate, and exports — key findings
- Government spending (both current and total expenditures) has tended to appreciate the fundamental equilibrium real exchange rate, although some effects have been mitigated by trade liberalization and negative terms of trade movements.
- Mozambique’s total export shares in world trade have increased mainly as a result of strong growth in megaproject exports; nonmegaproject exports’ share of world trade and ratio to GDP has remained roughly constant.
- Econometric results suggest the REER may have potentially been overvalued at times of tight exchange rate management; it is difficult to infer whether deviations from equilibrium could be attributed to the response to aid.
- Supply-side effects of aid (through health and infrastructure investments) could mitigate real exchange rate appreciation, but visible evidence that these supply-side effects are strong enough is lacking — possibly due to lagged (and possibly nonlinear) effects and the rudimentary state of PFM systems in Mozambique until recently.

### Short-run dynamics and aid absorption — VAR and DSGE evidence
- The short-run response to aid depends on a number of factors, including the degree of aid absorption (IMF (2005a) defines aid absorption as the extent to which a country’s nonaid current account deficit widens in response to an increase in aid inflows; full aid absorption is equivalent to an unchanged level of international reserves in response to an aid shock under a closed capital account).
- An identified six-variable VAR was estimated for Q1 1996 to Q3 2006 to investigate transmission of aid shocks.
  - Ordering of variables used for identification: foreign aid, real GDP (or real nonmegaproject exports), inflation, reserves, domestic debt, and the nominal exchange rate.
  - Structural shocks recovered using Cholesky decomposition of the variance-covariance matrix.
  - Time-series properties: variables are I(1) except aid which is I(0); series were detrended for estimation.
- Impulse responses (quarterly) indicate:
  - Aid shocks (that have been mostly spent) initially resulted in a nominal exchange rate appreciation, and possibly lower output.
  - Inflation declines initially but picks up gradually.
  - Using real nonmegaproject exports instead of the real GDP proxy shows only modest contractions of the export sector in response to aid shocks.
  - Most of the foreign exchange associated with aid inflows appears to have been sold by the central bank within five quarters of the initial aid shock, facilitating a textbook “spend and absorb” response.
  - The central bank appears to have smoothed foreign exchange sales in the very short term through temporary net domestic debt issuance (sterilization), which may have helped keep prices at bay and minimized crowding-out of private sector credit.
  - This pattern contrasts with most other non-CFA SSA countries that relied more heavily on sterilization, contributing to unfavorable debt dynamics and probably greater crowding-out of the private sector.

### Illustrative scaling-up scenarios — setup and motivation
- Purpose: identify potential macroeconomic implications and key measures and policies to absorb higher aid and ensure efficiency.
- No full MDG costing for Mozambique; paper considers arbitrary but sustained increases of external finance for illustrative purposes rather than MDG-targeted scenarios.
- Focus: macroeconomic implications and policy trade-offs involved in spending and absorbing a further scaling-up of foreign aid amounting to more than 15 percent of GDP at present.

### Growth prospects and possible channels of impact
- Growth accounting literature indicates physical capital accumulation and total factor productivity have been fundamental to post-conflict growth acceleration.
- Studies confirm the possibility of achieving:
  - about 7 percent growth in the medium term (baseline in staff report)
  - 5–6 percent in the long-term, a level sufficient to halve the poverty rate by 2015, assuming continuation of strong productivity growth.
- Upside potential from higher public investment depends on the ratio of current-to-capital spending, and through crowding-in private investment and spillovers from infrastructure and human capital accumulation.
- Possible negative consequences from further scaling-up include:
  - adverse effects on private investment reflecting diminishing returns,
  - challenges to human capital quality (e.g., maintaining education quality amid rapid aid-financed expansion of the school network).
- Recent evidence: primary education completion rates and quality of education are relatively low, probably reflecting high pupil-to-teacher ratios and a growing number of untrained teachers.

### DSGE scaling-up simulations and scenarios
- Simulated the estimated DSGE model for Mozambique (Peiris and Saxegaard (2007)) in response to a persistent aid shock with autocorrelation coefficient of 0.7 which raises aid by 2 percent of steady-state GDP.
- All scenarios assume full spending and absorption of foreign aid.
- Three scenarios considered:
  - Scenario 1 (unbroken line): no productivity spillovers from public investment.
  - Scenario 2 (thick broken line, baseline): modest level of productivity spillovers from public investment.
  - Scenario 3 (light broken line): high level of productivity spillovers from public investment.
- Note on productivity function: follows Pratti and Tressel (2006) where productivity is an increasing function of the size of public investment expenditure; higher productivity can arise from greater spillovers (e.g., infrastructure) or a decrease in the ratio of current-to-capital spending.

### Policy implications and trade-offs
- Spending and absorbing aid is generally the preferred macro response, but can entail significant loss of competitiveness and lower exports and real GDP growth if unmitigated.
- Importance of prioritizing reforms and further analysis of absorptive capacity, including:
  - improving PFM systems to support the effective use of aid and enable supply-side responses,
  - ensuring the composition of public spending favors capital formation and productivity-enhancing investment (infrastructure, human capital) to generate spillovers,
  - careful management of foreign exchange flows to limit short-term REER misalignment while avoiding excessive sterilization that could harm debt dynamics and crowd out private credit.

*Source: IMF staff report chapter entitled "9. The spending of foreign aid could undermine competitiveness, although such effects can be ameliorated depending on the nature of the supply side response over the long-term."*

### 15. Greater productivity-enhancing public investment expenditure could ameliorate

### 15. Greater productivity-enhancing public investment expenditure could ameliorate or even reverse Dutch-disease effects of additional government spending in Mozambique (Figure 5)

### Impulse-response findings on productive public investment
- Productive government investment expenditures can mitigate Dutch-disease effects of aid-financed government spending by reducing real exchange rate appreciation pressures and eliciting a stronger export performance (impulse responses in Figure 5 on a quarterly time interval).
- Consumption and output are unambiguously higher with greater productivity spillovers.
- The pick-up in inflation following an initial decline would be milder with greater productivity spillovers.
- None of the outcomes of productivity spillovers are assured and will depend critically on the composition of government expenditure and efficiency of public service delivery.

### Caveats and modelling assumptions
- The analysis assumed that productive government spending affects transitional dynamics but not the steady-state for simplicity and tractability.
- This assumption is consistent with fairly persistent but stationary aid shocks observed in the past in Mozambique, but permanent increases in aid that affect steady-state variables were not modeled and remain an area for future research.
- Further analyses could consider absorptive capacity (e.g., through the World Bank’s Marquette for MDG simulations), particularly relevant as Mozambique is one of the few countries in the range of saturation points for aid identified in the literature.

### Priority sectors and scaling-up experience
- Since 2000, about 65 percent of scaled-up spending has been allocated to priority sectors identified in PARPA I (education, health and infrastructure), contributing to:
  - doubling the number of children in primary school,
  - reducing infant and maternal mortality,
  - beginning the provision of Anti-Retro-Viral (ARV) treatment for HIV infection, partly financed by HIPC resources.
- Large human capital and infrastructure gaps remain; comparisons with fast-growth Asian economies indicate the importance of expanding secondary education and addressing acute infrastructure gaps, particularly communication and transportation networks.

### Agriculture and rural development
- Improving agricultural productivity is central to growth prospects and poverty reduction (Fox et al 2007).
- Much crop income growth has come from area expansion and use of available labor; basic food crops grown by smallholders have exhibited relatively stagnant yields.
- Policy and investment priorities include:
  - security of land-tenure,
  - adoption of new technologies associated with cash crops,
  - diversification into profitable cash crops and contract farming,
  - use of productivity-enhancing inputs (fertilizers, seeds, irrigation),
  - development of rural credit and input markets,
  - improved road infrastructure, storage facilities and extension services.
- Mozambique has potential for high-value vegetable crops and floriculture for export to Europe through better quality standards and marketing infrastructure.

### Strengthening the business environment
- Mozambique’s manufacturing base remains nascent due to poor competitiveness and low “factory-floor” productivity explained by high indirect costs, output losses, burdensome regulations, obsolete equipment, and absence of modern management techniques.
- Core reform priorities to lower the costs of doing business:
  - (i) streamlining burdensome regulatory practices;
  - (ii) reducing the costs of hiring and firing workers;
  - (iii) removing infrastructure bottlenecks;
  - (iv) financial sector reforms to reduce cost of and expand access to financial services;
  - (v) a governance agenda to minimize expropriation risks.

### Public financial management (PFM) and public sector reforms
- Strengthening PFM systems and wider public sector reforms are vital to ensure resources efficiently reach productive priority sectors, including at the subnational level.
- Enhancing administrative capacity is essential to absorb scaled-up donor assistance.
- Government is transitioning towards program-based budgeting; pilot with a few line ministries in the 2008 budget is suggested.
- CFMP requires strengthening through better costing of policies and more comprehensive sector strategies.
- State-owned or public-participating enterprises should be consolidated into budgetary accounts or at minimum better monitored.
- Donor coordination: include donor-financed projects in the single treasury account and e-SISTAFE to improve timely fiscal reporting on development expenditures.

### PFM reforms in Mozambique (Box I.1) and projected SISTAFE outputs
- e-SISTAFE rolled out to most line ministries at the central and provincial level by end-year 2006.
- Medium-term fiscal framework (CFMP) aligned with PARPA II and approved by the Council of Ministers in 2006.
- APB milestones and e-SISTAFE rollout:
  - e-SISTAFE extension to remaining ministries and state organs by end-December 2007; phased rollout to districts and municipalities (initially to 27 districts by end-October 2007).
  - New e-SISTAFE functionalities in 2007: Phase II of budget formulation, payroll and pension, revenue collections modules.
  - Integrated payroll database to enable salary payments via direct budget execution in e-SISTAFE by end-June 2007.
  - Limited separate foreign currency accounts within the single treasury account in 2007 to facilitate inclusion of donor-financed projects.
  - National Decentralization Strategy, including intergovernmental fiscal relations review, to be prepared by June 2008.
  - Continued implementation of the new procurement system up to the district level and capacity increases for internal and external audit bodies with milestones set for end-2007.
- Projected SISTAFE Outputs, 2006–2009 (selected):
  - Financial and budget execution in operation in all ministries at central, provincial and district levels, and in municipalities and public enterprises.
  - The module of payment of salaries and pensions implemented.
  - The budget formulation module implemented (Phase II).
  - The asset management module and procurement interface implemented.
  - The ATM revenue management module implemented.
  - Program budgets under preparation by ministries.
  - The debt management module implemented.
  - The internal audit module implemented.
  - The Data Processing Centre (CPD) operating as an effective and sustainable unit.
  - Project Implementation Unit for SISTAFE operating as an effective and sustainable unit.

### Exit strategy and fiscal sustainability under scaling-up
- Objective: gradually increase domestic revenue to enable an orderly exit from donor dependence and finance at least recurrent spending from own resources over the next eight years (post-2015 donor reversion or decrease scenario).
- Illustrative fiscal framework: assumes spending of an additional US$200 million from 2008–15 compared to the baseline; increased expenditures assumed divided between current and capital spending in a 1:2 ratio.
- Projected revenue and sustainability results:
  - Projected revenue ratio of about 19½ percent of GDP in 2015 (increasing from 14 percent of GDP in 2006) would cover both the scaled-up level of recurrent spending and a portion of the public investment program under an annual US$200 million scaling-up.
  - Even if all scaled-up spending were financed through concessional external borrowing, the external debt-to-export ratio would remain below indicative performance-based debt burden thresholds (e.g., NPV of debt-to-exports ratio of 150 percent for Mozambique).
  - Low level of domestic public debt: about 8 ½ percent of GDP.
- Fiscal policy implications:
  - Spending plans need regular reassessment in light of growth and Dutch-disease effects of initial spending expansion and realized revenue outcomes.
  - Any nonconcessional financing of large infrastructure projects should be considered case by case based on economic return, impact on debt sustainability, and effects on donors’ financing decisions.
  - A prudent external debt management strategy of avoiding non-concessional external borrowing is assumed to continue.

### Revenue mobilization and tax administration reforms
- Varsano et al. (2007) estimated Mozambique’s tax effort at about 50 percent (ratio of actual collection to taxing capability), suggesting scope for raising revenues up to nearly 22 percent of GDP.
- Schenone (2004) found that the difference between potential and actual collection due to noncompliance (evasion) is much larger than that due to tax law features (exemptions and nonassessments).
- Policy implication: greater gains likely from measures that increase tax administration efficiency rather than from raising tax rates.
- Next phase of revenue administration reforms (2007–10), supported by a multi-donor common fund, will focus on implementing the central revenue authority (ATM) strategic plan:
  - ATM establishment stages: transition period to end-2007; gradual integration of tax and customs agencies during 2008; further strengthening and consolidation in 2009 and 2010.

*Source: _cr07258 - 15. Greater productivity-enhancing public investment expenditure could ameliorate or even reverse Dutch-disease effects of additional government spending in Mozambique (Figure 5)._*

### 19. Mozambique’s vast natural resources and present boom in commodity prices

### 19. Mozambique’s vast natural resources and present boom in commodity prices

### Natural resources, FDI, and fiscal trade-offs
- Megaprojects in the mining and petroleum sectors have placed Mozambique on the global FDI map, but fiscal revenues to date have been low due to generous fiscal incentives put in place to attract private investment during the recovery from the civil war.
- Authorities recognize that fiscal contribution of new projects could be substantially increased by reducing generous tax exemptions.
- Mozambique is putting in place a new fiscal regime for the mining sector and building capacity of personnel to negotiate effectively with multinationals through the use of financial models and model contracts.
- Present mining and petroleum projects in the pipeline, including the multibillion dollar Moatize coal-mine project now under negotiation, could yield revenues in the range of 2–4 percent of GDP beyond 2010.
- Further gas and oil discoveries would result in significant revenue contributions over the long term.

### Petroleum sector developments and prospects
- Following agreements negotiated in 2000, the Pande-Temane gas field installations, processing facilities, and gas export pipeline to South Africa were commissioned in 2004 amounting to some US$1 billion FDI in all.
- The government has secured more than US$300 million in petroleum exploration work commitments from the recent Rovuma Basin round, adding to perhaps US$200 million in planned drilling and evaluation activities on contract areas awarded earlier.
- The Pande-Temane project (including the pipeline) is currently in a period of cost recovery and high debt service; government revenues have been modest, comprising mainly petroleum production tax and a small amount of corporate income tax, but these are set to increase significantly as production ramps up and joint venture investors have recovered their initial development costs.
- Sufficient additional reserves have been identified that the joint venture partners are considering a possible 50 percent expansion of the annual gas sales volume to around 180mmGj.
- Under the Petroleum Production Agreement the joint venture partners are allowed 25 percent depreciation, resulting in reduced income tax revenues in the early years of the project in addition to numerous other tax exemptions.
- The current combination of economic terms for exploration and production concession contracts (EPCCs), if properly administered, should yield a share of revenues for the government that is competitive by international standards.
- Authorities are reviewing the petroleum fiscal regime and intend to put in place a comprehensive fiscal regime embodied in the general tax law to avoid case-by-case negotiation of petroleum tax and production sharing terms.

### Fiscal implications, policy direction, and megaprojects
- Generous fiscal incentives granted earlier to attract investment have limited fiscal revenues to date.
- Reforms under way aim at reducing tax exemptions and improving contract negotiation capacity to increase fiscal returns and economic linkages while minimizing environmental damage and social dislocation.
- Revenues from present and planned mining and petroleum projects could help sustain long-term fiscal sustainability and provide an exit strategy from aid dependence, enabling at least recurrent spending to be financed from own resources.

### Aid scaling-up scenarios and macroeconomic management
- Mozambique has fully spent and mostly absorbed scaled-up foreign aid ranging between 10–20 percent of GDP over the last decade.
- Outcomes achieved during the period of scaled-up aid include:
  - Doubling of the number of children in primary school.
  - Reduction in infant and maternal mortality.
  - Beginning the provision of ARV treatment for HIV infections.
  - Sustaining economic growth of eight percent per year on average.
  - Reducing the poverty headcount index from 69 percent in 1997 to 54 percent in 2003.
- A modest scaling-up of foreign aid inflows (about 2 percent of 2008 GDP) could continue to be spent and absorbed aimed at achieving the MDGs.
- Risks of scaling-up:
  - Spending increases tend to appreciate the real exchange rate and potentially hurt the export sector and long-run growth (Dutch-disease effects).
  - If scaling-up is significantly larger or less persistent, or if microeconomic capacity constraints bind, policymakers may view macroeconomic variables and volatility as too large for consumer welfare or long-term growth.
- Policy responses recommended:
  - Effectively reach the most productive priority sectors (e.g., agriculture, education, health, infrastructure as defined in PARPA II) to elicit a supply response and mitigate Dutch-disease effects.
  - Strengthen public financial management (PFM) systems while embarking on public sector reform to improve efficiency and public service delivery.
  - Complement greater productive public investments with second-generation reforms to strengthen the business environment and buttress agricultural production.
  - If necessary, smooth the expenditure pattern and save part of the aid in international reserves to be spent and absorbed at a later date.

### Debt relief, fiscal space, and revenue targets
- Debt relief under the HIPC initiative and MDRI has reduced Mozambique’s debt levels and provided fiscal space to maintain relatively high expenditures (about 25-30 percent of GDP) financed through concessional external borrowing and foreign grants.
- The authorities’ strategy of absorbing foreign aid through sales of foreign exchange and avoiding nonconcessional external borrowing has helped create a conducive environment for scaled-up expenditures.
- Given the low tax-to-GDP ratio and need to guard against aid volatility and reduce donor dependence, an annual average revenue increase of 0.5 percent of GDP should continue to be targeted through the CFMP, achieved by widening the tax base and improving revenue administration.
- Beyond the PARPA II period, strong growth and increased fiscal revenues from megaprojects—including the development of the multibillion dollar Moatize coal-mine project and Pande-Temane Gas Project—would help Mozambique maintain long-term fiscal sustainability and possibly allow it to start tapping international capital markets.

### Financial sector status and reform priorities (summary from subsequent chapter)
- Financial sector development is a main objective of PARPA II’s economic development pillar.
- By end-2002 the financial system had become dominated by foreign banks and highly concentrated: the five largest banks accounted for 96 percent of total deposits at end-2002.
- The system was increasingly dollarized: foreign currency deposits and loans accounted for 51 percent and 70 percent of their respective totals in 2002.
- Nonperforming loans (NPL) were high at 21 percent at end-2002.
- Real lending rates in Meticais in the period 1999–2002 averaged 17.5 percent.
- High lending rates and poor financial intermediation reflected large intermediation spreads driven in part by:
  - High levels of NPLs made to state-owned enterprises and other non-compliant borrowers after bank privatization.
  - Lack of mandate and capacity to effectively supervise banks and inadequate reporting standards for monitoring soundness of bank lending.
  - High overhead costs (partly reflecting the small size of the financial system), high reserve requirements, and substantial profit margins reflecting bank concentration and limited competition.
- Legal and institutional impediments constraining intermediation include weak debt-enforcement mechanisms for unsecured loans, incomplete collateral and bankruptcy laws, and lack of comprehensive credit information on borrowers—factors that have led banks to adopt conservative lending policies, impose stringent collateral requirements, and limit access to credit for firms without credit histories.
- Mozambique is deepening reforms by strengthening the institutional regulatory architecture for financial sector development and addressing critical financing constraints identified in the 2003 Financial Sector Assessment Program (FSAP).

*Source: _cr07258 - 19. Mozambique’s vast natural resources and present boom in commodity prices*

### 28.      High and volatile real treasury bill rates were also partly affected by weaknesses

### _cr07258 - 28.      High and volatile real treasury bill rates were also partly affected by weaknesses

### Drivers of high and volatile real treasury bill rates
- Significant supply shocks (e.g., related to weather and terms of trade) and large fiscally-induced demand shocks (i.e., unanticipated influx of external grants and loans) contributed to instability in Mozambique’s macroeconomic environment.
- Weaknesses in the monetary framework and liquidity management operations prevented monetary authorities from cushioning shocks in a systematic, credible, and transparent way, causing substantial volatility in both prices and nominal interest rates.
- Real interest volatility was further exacerbated by shallow foreign exchange and public debt markets.
- Empirical reference: "Boyd, Levine, and Smith (2001) have shown that low levels of financial intermediation are partially explained by high volatility in interest rates and monetary aggregates..."

### Institutional roots of poor financial intermediation
- Contract enforcement and securitization problems:
  - Inefficient courts with a high number of procedures, long delays, and high costs to enforce commercial contracts impeded lending.
  - Pledging real property as collateral was difficult because of weak property rights and complex, costly procedures to register and transfer property.
- Credit information shortfalls:
  - The public credit registry run by the central bank covered only about 11,500 borrowers (firms and individuals) and had only three full-time staff.
  - Though borrowers increased to about 99,000 in 2006, overall quality still fell short of regional best practices.
- Presentation of comparative indices and indicators appears in the source (World Bank 2003; World Bank 2007).

### Weaknesses in monetary policy management (Box II.2)
- Monetary framework lacked transparency:
  - Market participants did not fully understand the goals or operational framework of monetary policy.
  - Insufficient BM communication and inconsistent use of instruments sent conflicting signals on policy stance.
- Bank of Mozambique (BM) net financial position:
  - BM reported positive capital and declared operating profits, but net worth and profits were negative once valuation adjustments were accounted for.
  - BM losses originated mainly from:
    - accumulated foreign exchange valuation losses on BM external liabilities to foreign lenders, and
    - operating losses stemming from the purpose of sterilization (costs of monetary policy) not transferred to the budget.
- Consequence: BM’s fragile financial position risked its operational independence.

### Access to affordable finance and outreach (FSAP findings)
- Financial savings mobilization was constrained by Mozambique’s low-income levels and a thin bank branch network, particularly in rural areas.
- Microfinance sector was relatively small and limited in outreach.
- Constraints to providing financial services in rural areas included very low population densities, weak or nonexistent markets, poor physical and communication infrastructure, and local unavailability of skilled staff.
- Table II.2 entries as presented in source:
  - "Mozambique20023.085,64511.0"
  - "South Africa3,09334.516,84220.2"
  - "Tanzania27139.0173,99811.6"
  - "Zambia35413.866,33917.7"
  - "India448338.414,88824.3"
  - "Indonesia683114.112,54727.1"

### Nonbank financial institutions (NBFI) risks
- NBFIs were small, nontransparent and subject to limited institutional and regulatory oversight.
- Supervisory capacity was weak and diluted between the Ministry of Finance (insurance) and the state pension scheme, INSS (pensions), which had both administrative and prudential mandates.
- NBFIs could mask sizeable contingent fiscal liabilities, especially in insurance and pension sectors.

### FSAP policy recommendations (Box II.3) — key items
- Banking Supervision and Reporting Standards:
  - Establish closer links with foreign shareholders and their supervisors.
  - Bring the loan classification and loan loss provisioning systems in line with international practices.
  - Strengthen and modernize supervisory functions through capacity building and reorganization.
  - Define a program towards full adoption of IFRS.
  - Review the legal and regulatory framework for bank resolution.
- Monetary Policy and Public Debt Management:
  - Ensure consistency between instruments and the monetary targeting framework.
  - Introduce new instruments (repos/reverse repos) for short-term liquidity management and direct foreign exchange interventions to smooth volatility.
  - Enhance central bank capacity for timely forecasts of short-term liquidity flows in coordination with Ministry of Finance.
  - Improve central bank communication by establishing a monetary policy strategy document and better dissemination of decisions.
  - Ensure BM’s operational independence by:
    - Transfering medium- and long-term foreign (HIPC) debts to treasury; and
    - Reviewing accounting and profit transfer practices to shift monetary policy costs to the Ministry of Finance.
  - Strengthen public debt management by improving reporting and formulating a debt management strategy; restructure the Public Debt Unit of the Ministry of Finance.
- Institutional Lending Environment:
  - Enact proposed amendments to BM law and the Banking Law; review Commercial Code; modernize Notary Office, Public Commercial Registry, and Code of Civil Procedure; establish a special division of Maputo City Court; update secured transactions framework; consider a specialized commercial court.
- Capital Market, Pensions, Insurance:
  - Review capital controls regulation and build supervisory capacity; improve accounting and disclosure standards for insurance and pensions; separate supervisory and administrative functions of INSS.
- Microfinance:
  - Review draft legislation on microfinance.

### Mozambique’s financial sector reform strategy: FSTAP
- Goals of FSTAP:
  - Strengthen banking sector and BM capacity in supervision, banking system management, and financial infrastructure including real time gross settlement system.
  - Improve financial accountability and transparency via IFRS adoption (commercial banks first, corporate entities second).
  - Strengthen public debt management capacity: develop a debt management strategy; reorganize and strengthen the public debt department of the Ministry of Finance.
  - Improve money and government bond market efficiency and depth, including introduction of a central securities depository.
- IMF role: provide technical assistance in coordination with government and other donors.

### FSTAP components, financing, and donor roles
- Project structure: one project with ten components supported by multiple donors; an administrative component established a project implementation unit (PIU).
- Donor involvement summary (as presented in source Table II.3):
  - Agencies (F = financier, T = technical assistance provider) across components included IDA, AfDB, DFID, SIDA, KfW/GTZ, and IMF.
  - Note on component 5: Provided in collaboration with the International Fund for Agriculture Development outside FSTAP under the Rural Finance Support Program.
- Component highlights (Box II.4):
  - Component 1: Strengthening BM’s Institutional Capacity (Basel I implementation, inspections manuals, prudential regulations, bankruptcy law drafting for financial institutions, supervisory framework for NBFIs, IFRS transition for central bank accounting, research capacity, divest government shareholdings).
  - Component 2: Improving Financial Accountability and Transparency (IFRS adoption, central securities depository, private debt registration procedures).
  - Component 3: Strengthening Public Debt Management (debt management strategy, liquidity management, restructuring Public Debt Unit, consolidating debt database).
  - Component 4: Improving Money and Debt Markets and Financial Infrastructure (improve primary and secondary government debt market; develop the repo market).
  - Component 5: Micro and Rural Finance (support MFIs; promote rural expansion and MFI integration).
  - Component 6–9: Insurance/pensions legislation and supervision; social security system reforms and actuarial analysis of INSS; legal/judicial reforms for financial transactions including bankruptcy law, commercial tribunals, property registries, credit registry enhancements; establishment of a Financial Intelligence Unit.
  - Component 10: Support for Project Implementation (finance consultancies, advisors, personnel, training, equipment).

### Implementation, coordination, monitoring, and capacity building
- Implementation arrangements:
  - PIU headed by a project coordinator reporting to the Ministry of Finance; responsible for procurement guidance, monitoring and evaluation, and overall reform thrust.
  - Arrangement aims to reduce transaction costs and promote donor coordination in line with Paris Declaration for Aid Effectiveness.
- Monitoring and evaluation:
  - Reform strategy monitored by measurable performance indicators and systematically evaluated by government and development partners.
  - An action plan (results framework matrix) for each FSTAP component spans the program’s five years.
  - Quarterly monitoring tables and progress reports are prepared by the PIU and discussed by a technical working group (TWG) jointly chaired by the PIU coordinator, its procurement officer and beneficiary government agency representatives.
  - Actions identified as high priority are included in the Performance Assessment Framework (PAF) Matrix used by government and donors to evaluate progress on poverty reduction strategy annually.
  - A medium-term review of the strategy is expected to take place at the end-2008, including an FSAP update and reappraisal of obstacles to financial system soundness and depth.
- Capacity building:
  - Central element of FSTAP; success depends on local capacity to analyze, redesign, and implement strategies.
  - IMF provides technical assistance to improve BM’s liquidity management, monetary and exchange rate policy operations, and financial sector supervisory capacities.
  - A technical assistance strategy specifying priority areas and delivery modes has been prepared.

*Source: _cr07258 - 28.      High and volatile real treasury bill rates were also partly affected by weaknesses (IMF PDF).*

### 38.      Efforts since the 2003 FSAP have focused on restructuring and strengthening

### Efforts since the 2003 FSAP have focused on restructuring and strengthening the supervision of Mozambique’s banking system

### Banking restructuring, privatization, and supervision
- Bank privatization process in the mid-1990s was plagued by lack of due diligence, resulting in costly government recapitalizations during 2000–02.
- Recapitalizations related to Banco Austral and Banco Comercial de Moçambique amounted to 6 percent of GDP.
- Over the last four years the government has demonstrated commitment to return banks to private ownership and remains a minority shareholder of only one bank for which intentions to withdraw have already been publicly announced.
- To dispel uncertainties after restructuring, the government allowed comprehensive diagnostic reviews of the major banks by international firms according to IFRS; these reviews were concluded in 2005 and endorsed the overall soundness of the banking system.
- A new law strengthening the BM’s supervisory and enforcement mandates was implemented in 2004, and capacity building has been provided by the IMF, World Bank, and other donors.
- Key measures adopted after the FSAP (Box II.6):
  - Comprehensive diagnostic reviews of the major banks by international firms according to IFRS (completed in 2005).
  - Implementation in 2004 of a new Financial Institutions Law strengthening the central bank’s supervisory powers and introducing more strict penalties for noncompliance violations.
  - Continued progress to strengthen banking supervision practices in line with Basel Core Principles (BCP) by: (i) building on- and off-site supervisory capacity; (ii) requiring the preparation of financial statements of Mozambique’s financial institutions and corporate sector in compliance with international [standards]; and (iii) adopting prudential regulations in line with international best practices.
  - Adoption of IFRS by the BM in 2006 and by commercial banks in 2007.
- IMF technical assistance priorities (Box II.5) include: (i) making monetary policy more effective; (ii) further strengthening the supervisory and regulatory framework to foster the sound expansion of the financial sector and achieve full compliance with international standards in accounting, anti-money laundering, and banking supervision; and (iii) developing interbank foreign exchange market and government debt management capacity to support financial deepening and smooth monetary policy transmission.
  - These priorities are supported by annual multi-topic TA missions and an FSAP update in 2008.

### Monetary policy operations and public debt management
- The BM has taken steps to strengthen monetary policy operations by increasing the use of foreign exchange sales to sterilize excessive liquidity and allowing issuance of treasury-bills for monetary policy purposes.
- Ongoing work to improve liquidity management practices and forecasting, ensure operational independence, and deepen foreign exchange, money, and debt markets (Box II.7).
- Monetary policy framework and operational changes:
  - BM adopted a money targeting framework using base money as the operational target.
  - BM approved a long-term monetary policy strategy document clarifying the money targeting framework, defining a new format for the monetary policy committee, and specifying its communication policies.
  - Work is ongoing to translate annual and quarterly monetary targets into a liquidity management plan separating short-term “temporary” and long-term “structural” liquidity.
  - The BM will introduce repos in 2007 to fine-tune “temporary” liquidity variations.
  - BM is working to ensure an “appropriate” mix of foreign exchange sales and net treasury bill issuance to sterilize structural liquidity.
- Central bank operational independence:
  - Despite recent recapitalization agreements, central bank operational independence in conducting monetary policy has yet to be ensured through a comprehensive agreement to shift corresponding costs to the Ministry of Finance; a memorandum of understanding expected to be signed by both parties in 2007 is being discussed.
- Central bank liquidity management:
  - BM is refining liquidity monitoring and forecasting with Fund TA; further TA is needed to enhance short-term liquidity forecasting and use of appropriate instruments.
  - Critical importance of improved treasury information on daily cash flows and introduction of repos.
- Public debt management:
  - Reform is still in early stages; government needs a strategy for management of domestic debt and restructuring/training of the public debt department at the Ministry of Finance.
  - Treasury bill and money markets: removal of interest caps on T-bill auctions is progress, but further steps needed to foster an active and liquid government securities market, including enforcement of legal framework governing transfer of securities and implementation of an RTGS system; draft law on a new national payment system has been submitted and an analysis of obstacles to money and secondary debt markets expected to be concluded by end-2007.
  - Foreign exchange markets: important steps on the interbank foreign exchange market (MCI) with gradual removal of price caps and measures to promote firm quotations; further reforms needed to complete removal of remaining price caps and adopt an appropriate collateral system.

### Institutional lending environment, NBFIs, and rural finance
- Institutional lending reforms: recent achievements include implementation of the new Commercial Code and establishment of commercial sections of judicial tribunals in Maputo, Beira, and Nampula; the latter will become operational in 2007 and is expected to accelerate debt recovery of unsecured lending once equipment, facilities, and capacity building are provided.
- A revised Civil Procedure Code with streamlined judicial procedures is expected to further accelerate debt recovery once enacted.
- Nonbank financial institutions (NBFIs) and social protection:
  - BM has started to license and supervise deposit-taking MFIs.
  - A new law on social protection submitted to the Assembly will help strengthen social security and supplementary pension system.
  - As part of INSS restructuring, an actuarial study to determine sustainability of current contribution levels will be completed before the end of 2007.
  - Insurance Law is being reviewed in line with international best practices; revised prudential and solvency requirements for Mozambican insurers are being developed and new IFRS-compliant Chart of Accounts for insurers will be designed.
  - New supplementary pension funds regulations are being prepared; future focus needs to shift to strengthening capacity of industry regulators.
- Rural financing and outreach:
  - Government launched initiatives to enhance physical presence of banks in rural districts, targeting reduced operational costs through fiscal incentives, infrastructure improvements, and relaxing legal reserve requirements in rural branches to include cash in vault.
  - Transport costs related to constitution of legal reserves and cash in vault minimized with opening of BM branches in selected districts.
  - Fiscal incentives in the form of grant and loan resources are partially provided under the Rural Finance Support Program (RFSP) financed by the AfDB and IFAD.

### Reform impact on financial soundness and intermediation
- Financial sector reforms appear to have contributed to improvements in bank soundness: nonperforming loans have declined markedly, and bank profitability has recuperated (Figure II.2).
- Decrease in financial dollarization and lower financial vulnerability observed.
  - The strong decline in credit in foreign currency was particularly driven by a prudential measure to provision 50 percent of foreign-currency denominated loans to nonexporters introduced in July 2005 (Notice 5/2005); this decrease was accompanied by an increase in applications for private external financing.
- Positive spillovers to financial intermediation (Figure II.3):
  - Real lending rates and spreads have been on a declining trend since 2002.
  - Private credit growth started to pick up more recently, particularly in domestic currency.
- Microfinance sector trends (Box II.9):
  - MFIs’ outstanding loan portfolio and clients increased about a five and seven-fold increase from 2000 to 2005.
  - Consolidation: three institutions out of about thirty-five control more than 68 percent of the active loan portfolio.
  - Institutional growth: in 1998 there were only two operators with more than 1,000 clients; by 2005 half the MFIs had more than 1,000 clients.
  - Improvement in performance: percentage of MFIs with recovery rates considered acceptable (90 percent or above) increased from 16 percent in 1997 to about 30 percent in 2005.
  - Outreach remains limited in rural areas: about 62 percent of MFIs and 50 percent of microborrowers concentrated in the Maputo-Matola urban corridor; rural areas account for less than 12 percent of borrowers and only 5 percent of the total active portfolio.
  - Preliminary 2005 informal sector data: of the nonagricultural informal sector activities, 42 percent of those involved in trading and 48 percent of those involved in services (excluding transport) are located in Maputo/Maputo City areas.

### Access to finance and remaining constraints
- Access to affordable credit remains a top constraint:
  - Recent business surveys identify high cost of finance as the top constraint to private sector growth; access to financial services is among the top three constraints.
  - More than 70 percent of firms do not have access to a bank loan or to overdraft facilities (higher for SMEs), according to the 2003 ICA and 2006 CTA-MPD survey.
  - Credit to agriculture has shrunk as a share of total credit, and microfinance outreach in rural areas remains poor—indicating binding constraints to financial intermediation still need to be tackled.
- Reform implementation unevenness:
  - Emphasis has been placed correctly on finalizing banking restructuring and improving supervision, but prevalence of high cost and limited access to credit among SMEs and limited credit growth in agriculture suggest need to step up reforms to improve the institutional lending environment and expand outreach by banks and MFIs in rural areas.
  - Pace of reforms in monetary and public debt management has been slower than envisaged in areas of liquidity and public debt management; interbank and secondary markets remain shallow, limiting effectiveness of monetary policy and pace of capital market development.

### Lessons learned and near-term challenges
- Lessons learned from Mozambique’s financial sector reform strategy:
  - Getting diagnostics and priorities right is an important starting point: use of recent diagnostics and FSAP-proposed prioritization supported targeted and more effective interventions, built political capital for reformers, and sustained reform momentum.
  - Close attention to institutional arrangements that ensure government ownership and coordination is critical: clear division of labor and financing responsibilities among donors, adoption of a common implementation unit accountable to a single government ministry but responsive to a committee of beneficiaries, helped donor coordination and reduced transaction costs.
  - Capacity building is essential to sustain reform: inclusion of capacity building measures in each FSTAP component was important.
  - Measuring performance is key: availability of easily measurable outcomes is essential to gauge reform progress, identify bottlenecks, and ensure accountability; Mozambique’s results framework proved useful.
- Near-term challenges:
  - Accelerate reforms to improve institutional lending environment, rural outreach by banks and MFIs, and measures to reduce cost of credit for SMEs and agriculture.
  - Deepen interbank, secondary debt, and foreign exchange markets to improve monetary policy transmission and capital market development.
  - Strengthen capacity of industry regulators for pension and insurance sectors and finalize outstanding legal and institutional reforms (e.g., property registries modernization, Insolvency and Bankruptcy Law, revised Civil Procedure Code).

*Source: IMF staff report excerpt — “Efforts since the 2003 FSAP have focused on restructuring and strengthening the supervision of Mozambique’s banking system”*

### 48.      Improving access to affordable finance remains a key challenge. With the

### Improving access to affordable finance remains a key challenge. With the

### Access to affordable finance and financial intermediation
- With the soundness of the banking system restored, reforms to improve financial sector intermediation and access to financial services are now central.
- Policies and reforms should strengthen the lending environment and facilitate outreach of banks, MFIs and other privately provided financial services to SMEs and rural areas.
- The recently announced rural financing strategy tackles impediments to bank branch development, including less stringent reserve requirements.
- Importance of identifying new quick wins, including:
  - Expanding credit registries’ scope and ensuring coverage increases in line with outreach growth.
  - Exploring leasing and factoring agreements as alternatives.

### Capacity building, monetary policy, and public debt markets
- Capacity building and human resource mobilization must keep pace with reforms to strengthen monetary policy formulation and deepen monetary and public debt markets.
- Accelerate reforms to improve the efficiency of primary debt markets and reduce segmentation in the interbank market with the introduction of repo operations.
- Implementing an RTGS system and these measures would help deepen the market for government securities and establish a benchmark for future capital market development.
- At the BM, permanent mobilization of resources toward policy and operations is required to:
  - (i) systematically assess the monetary stance and select the appropriate mix of instruments,
  - (ii) conduct repo transactions, and
  - (iii) communicate and justify BM’s actions to market participants in a transparent way.
- At the Ministry of Finance, reorganize the public debt department and train staff to strengthen information and analysis reporting.

### Supervision, accounting standards, and nonbank sector oversight
- Strengthen supervision and accounting standards beyond the banking sector to ensure the soundness of an expanding financial system.
- Near-term priorities:
  - Introduce risk-based supervision and international reporting standards in the banking sector to facilitate monitoring of capital adequacy during strong credit growth.
  - Roll out IFRS standards to the corporate sector to reduce credit risk by improving evaluation of loan portfolio quality.
- Secure sustainability of the national pension fund through restructuring based on an actuarial study.
- Strengthen the supervisory framework for the growing nonbank financial sector, with special attention to microfinance and institutions catering to the poor and unbanked.
- Regulatory and supervisory framework must balance soundness and sustainability without compromising growth and outreach.

### Institutional arrangements, coordination, and FSTAP
- Preserve positive institutional features of the reform strategy to maintain reform momentum.
- Ensure FSTAP remains the umbrella under which financial sector reforms are conducted in Mozambique.
- Government and donors should rely on FSTAP implementation arrangements when pursuing new initiatives.
- Rather than transferring the Micro and Rural Finance component from the FSTAP to the IFAD project, consider increasing coordination and information sharing between the two projects.
- Integrate the IFC-MCC project on credit registry to the FSTAP to avoid duplication of implementation and monitoring arrangements and rising transaction costs.

### Selected FSTAP targets and timetabled outcomes (2007–2010) — exact targets preserved
- 50 percent of banks in compliance with improved prudential regulations; and 80 percent of banks in compliance with improved prudential regulations.
- 100 percent of banks in compliance with improved prudential regulations; and BM compliant with 16 out of 23 Basel Core Principles.
- BM compliant with all Basel Core Principles.
- Timetable for the implementation for IFRS adopted; and 90 percent of banks' financial reporting IFRS compliant.
- 100 percent of banks' financial reporting IFRS compliant; and 75 percent of large enterprises' financial reporting IFRS compliant 2/; and New regulations on assessment, classification and provision, and on integral risk management becomes effective.
- IFRS starts to be extended to large enterprises. 2/
- 20 percent of medium enterprises financial reporting IFRS compliant.
- Strengthen staff complement, consolidate debt database; Annual publication of public debt reports including comprehensive debt inventory; and Public debt report published; and Regular debt forecasts first published.
- Semiannual publication of debt forecasts.
- Quarterly publication of debt forecasts.
- 4) Improved money and government bond market efficiency and depth:
  - Implementation of a master repurchase agreement and program.
  - Treasury security auctions held on a monthly basis.
- Gap analysis for MFI training conducted and training strategy in place; New regulation for the operation and supervision of MFIs developped; Development and introduction of innovative financial products.
- Microfinance Association operational; Number of Licensed microfinance institutions; 3/
- IFRS adopted by all large MFIs (microbanks); and Access of all large MFIs to payment system; and FinScope study initiated.
- FinScope study concluded.
- 6) Strengthened regulatory and supervisory regime for the insurance industry:
  - New prudential regulations and a supervisory framework implemented.
  - Insurance companies adopt IFRS.
  - Insurance premiums as a percentage of GDP at 3 percent.
- New pension fund legislation approved; and Recommendations on improving social security coverage implemented; and Number of registered participants of INSS reaches 90 percent; and Actuarial study completed.
- New organization structure for INSS adopted.
- Number of active contributors of INSS reaches 60 percent of participants by 2010.
- New laws on insolvency and property registries enacted; and Credit Registry expanded to include all providers of credit; and New commercial sections created in the Judicial Tribunals of the City of Maputo, and Provinces of Beira and Nampula.
- All property registries computerized and electronically linked.
- New law on the Financial Information Center of Mozambique enacted; and Financial Information Center established and fully functional.

*Source: excerpts from the IMF staff report content unit _cr07258.*

### 60.      Mozambique’s world export shares and terms of trade have grown markedly in

### Mozambique’s world export shares and terms of trade have grown markedly in

### Recent export performance and terms of trade
- Mozambique’s average share of world exports more than doubled in 2001–06, more than the average for sub-Saharan African (SSA) and ASEAN 4 countries.
- Mozambique’s terms of trade improved on average by nearly 7 percent a year in 2001–06, up from 0.5 percent in 1991–2000, mainly because prices for aluminum rose more than prices on imports beginning in 2003.
- Table III.1 (select figures):
  - Share in World’s Exports: Mozambique 1991-00: 0.01; 2001-06: 0.02
  - Terms of Trade Growth: Mozambique 1991-00: 0.5; 2001-06: 6.9
  - Comparative figures: SSA Share in World’s Exports 1991-00: 1.5; 2001-06: 1.6; SSA Terms of Trade Growth 1991-00: -0.2; 2001-06: 4.9
  - ASEAN 4 Share in World’s Exports 1991-00: 3.5; 2001-06: 23.5; ASEAN 4 Terms of Trade Growth 1991-00: 10.6; 2001-06: 0.4

### Composition of exports and megaproject dominance
- Mozambique’s strong export performance almost exclusively reflects megaproject-related exports, most notably aluminum.
- Revealed comparative advantage: aluminum, gas, and wood articles.
- Megaproject export shares and growth:
  - All megaproject exports (electricity, aluminum, and gas) accounted for more than 70 percent of total exports in 2006; aluminum alone accounted for 58.6 percent of total exports.
  - Traditional exports grew by an average 4 percent a year; mega-project exports grew by 10.4 percent.
  - Megaproject-related imports as a share of total imports hovered around 20 percent.
  - Since 2000 (Mozal full capacity), megaproject-related exports have surged; in 2004 Mozal’s exports accounted for 72 percent of total exports.
- Table III.2 (select figures, percent unless noted):
  - All products: Share in Total 2000: 100; 2004: 100. Share in World Exports 2000: 0.00; 2004: 0.60.
  - Aluminium: Share in Total 2000: 16.5; 2004: 27.2. Share in World Exports 2000: 0.87; 2004: 2.87. RCA 2004: 71.02 (RCA = Revealed comparative advantage).
  - Fish: Share in Total 2000: 27.3; 2004: 6.0. Share in World Exports 2000: 0.90; 2004: 0.25. RCA 2004: 0.58.
  - Mineral Fuels, Gas: Share in Total 2000: 20.9; 2004: 4.0. Share in World Exports 2000: 0.01; 2004: 0.01. RCA 2004: 0.63.
  - Wood articles: Share in Total 2000: 3.9; 2004: 2.6. Share in World Exports 2000: 0.02; 2004: 0.02. RCA 2004: 1.05.

### Competitiveness assessment and REER
- An updated macro assessment based on REER (Box 1 of the accompanying staff report) shows Mozambique’s REER relatively stable over the past decade.
- Econometric results suggest the REER may be overvalued by about 10 percent, though with statistical uncertainty; external competitiveness may have been stronger had REER aligned with underlying equilibrium.

### Policy environment favoring megaprojects
- Megaprojects benefit from favorable domestic policies: favorable trading terms, tax exemptions, and qualification as export-processing zones (EPZs) which allow duty-free imports, on-site customs facilities, and tax incentives.
- Example: Mozal (an EPZ) is exempt from customs duties on imported inputs, real property tax, value-added tax, and specific consumption taxes.
- Nineteen manufacturing factories benefit from EPZs, including Mozal and Moma.

### Protective measures for specific industries
- Protection for sugar and cashew:
  - Export tax on raw cashew: 18 percent.
  - Import tax on sugar: 25 percent.
- Rationale: sugar import tax introduced as condition for large-scale foreign investment in remote sugar estates; cashew export tax to help newly privatized sector adjust and after abolition of price control.
- Effects: increased production capacity and output of sugar mills; export tax penalizes small-size raw cashew exporters and stimulates small and medium-size cashew processing factories.

### Tariff and nontariff regime
- Tariff structure: four nonzero bands (2.5, 5, 7, and 20 percents).
- Trade-weighted average tariff: 9.1 percent; simple average tariff: 12.1 percent.
- Mozambique’s trade barriers (tariff and nontariff) are lower than neighboring countries and most ASEAN countries.
- Reducing the maximum tariff band further could eliminate the higher simple average tariff; a reduction of the maximum tariff rate by 5 percentage points, else equal, has been shown to lower Mozambique’s simple average tariff from 12 to 10 percent.
- Table III.3: Overall ratings (1-10) for tariff and non-tariff barriers (selected):
  - Mozambique: Overall Rating 2; Tariff Barriers Rating 2; Non-Tariff Barriers Rating 1.
  - Selected comparators: South Africa Overall Rating 5; Indonesia Overall Rating 4; Malaysia Overall Rating 4; Thailand Overall Rating 5.
- Table III.4 (Average tariffs, 2006):
  - Trade Weighted Average / Simple Average Tariff: Mauritius 2.2 / 3.1; Angola 5.4 / 7.2; South Africa 5.8 / 7.7; Tanzania 8.1 / 12.7; Mozambique 9.1 / 12.1; Madagascar 9.2 / 13.5; Namibia 9.6 / 7.7; Zambia 9.6 / 13.4; Botswana 11.1 / 7.7; Zimbabwe 18.0 / 16.2.
- Mozambique has gradually reduced its maximum tariff from 35 percent in 1999, to 25 in 2002, to 20 percent in 2006.

### Non-tariff barriers
- Mozambique’s nontariff barriers are less restrictive than bordering neighbors; import permits and import licensing are the most common nontrade barriers.
- Table III.5 (Types of Non-Tariff Barriers and Ratings, 2006):
  - Mozambique: Not significant (rating 1).
  - Examples: Indonesia—Affect 10% imports, 40% non-oil exports, 30% production; Export tax on rubber, import restrictions and licensing (rating 2). Thailand—Licensing for 26 categories of goods (rating 2).

### Trade liberalization and firm-level competitiveness
- Trade openness increased:
  - Imports to GDP: Mozambique 1991-00: 14.5; 2001-06: 29.2.
  - Exports to GDP: Mozambique 1991-00: 38.1; 2001-06: 43.4.
  - Comparative averages: SSA Imports to GDP 1991-00: 28.3; 2001-06: 32.5. ASEAN 4 Imports to GDP 1991-00: 53.8; 2001-06: 63.7.
- Despite liberalization, firm-level competitiveness gains limited:
  - Enterprise survey 2002–06 (percent of sample):
    - Exporting firms: 2002: 11.5; 2006: 13.9.
    - Importing firms: 2002: 43.2; 2006: 63.9.
    - Diversified production: 2002: 78.6; 2006: 60.1.
    - Innovate: 2002: 25.5; 2006: 23.4.
    - Capacity utilization improvement: 2002: 51.2; 2006: 59.1.
  - Note: 192 firms in 2002 survey; 158 firms in 2006 survey.

### Business environment and regulatory costs
- Mozambique’s business environment is weak relative to top ranks:
  - Governance and Doing Business distance to top rank (selected):
    - Mozambique Governance: 0.53; Doing Business: 0.79.
    - South Africa Governance: 0.59; Doing Business: 0.16.
    - Mauritius Governance: 0.22; Doing Business: 0.18.
  - On average, SSA countries rank worse on Governance and Doing Business than ASEAN countries for the sample.
- Regulatory costs to trade are among the most binding constraints: customs procedures, contract enforcement, and business registration.
- World Bank Doing Business rankings (Table III.9, selected metrics):
  - Mozambique ranks (out of 175 countries) and distances:
    - Ease of Doing Business: 140
    - Enforcing Contracts: 168
    - Starting a Business: 153
    - Trading Across Borders: 141
    - Change relative to 2005: +30 (worsening rank) in aggregate.
  - Small- and medium-size firms find customs procedures burdensome and nontransparent.

### Trade data quality and mirror statistics
- Mirror trade discrepancies indicate ineffective customs procedures:
  - Table III.10 (Imports from bordering countries as reported by Mozambique; Exports to Mozambique as reported by bordering countries; Share not recorded at Mozambique's border):
    - Malawi: Imports reported by Mozambique 29.4; Exports to Mozambique reported by Malawi 17.2; Share not recorded at Mozambique's border -0.71.
    - South Africa: 701.9 vs 991.9; Share not recorded 0.29.
    - Tanzania: 4.4 vs 6.9; Share not recorded 0.37.
    - Zimbabwe: 17 vs 19.5; Share not recorded 0.13.
- Interpretation: discrepancies suggest contraband, smuggling, and substantial duty-free entry of goods; countries with lower trade-weighted average tariffs (Angola, South Africa, Mauritius) have lower discrepancy between simple and trade-weighted average tariffs.

### Summary findings
- Mozambique’s external competitiveness gains concentrated in a few protected sectors (megaprojects) with widespread protection (tariff and tax exemptions, EPZ privileges).
- Non-megaproject businesses face relatively low tariff and minimal nontariff barriers compared with SSA peers, but a weak business environment continues to constrain competitiveness, especially for small- and medium-size enterprises.
- Improvements in customs procedures and regulatory environment would particularly benefit small- and medium-size enterprises.

### Market access
- Mozambique has preferential market access:
  - To the United States under the African Growth and Opportunity Act.
  - To the European Union under the Cotonou Agreement and Everything but Arms initiative.
  - Under the SADC Trade Protocol, preferential access to the South African market.
- Exports destination structure (Table III.11, percent of total exports):
  - Exports to World 1991-01: 100.0; 2002-05: 100.0.
  - Exports to industrial countries: 1991-01: 57.5; 2002-05: 66.2.
  - Exports to United States: 7.9; 1.5.
  - Exports to European Union: 40.1; 63.9.
  - Exports to Africa: 29.4; 23.1.
  - Exports to South Africa: 16.2; 15.8.
- South Africa absorbs about 16 percent of Mozambique’s exports and accounted for more than 11 percent of FDI in Mozambique in 2005.

### Strategic observation on market access strategies
- Historically SSA (including Mozambique) pursued regional or bilateral preferential trading arrangements, while ASEAN 4 moved toward broad multilateral liberalization following MFN principles; the MFN principle extends equal market access regardless of origin, whereas regional/bilateral arrangements extend preferential access only to participating countries.

*Source: IMF staff estimates and projections and accompanying staff report as contained in the supplied chapter content.*

### 74.      The economic literature has usually regarded broad-based liberalization as the

### _cr07258 - 74.      The economic literature has usually regarded broad-based liberalization as the

### Literature findings on liberalization and RTAs
- Broad-based multilateral liberalization is considered the best strategy in the economic literature.
- Vamvakidis (1998) and Yagci and Aldaz-Carroll (2004) find economies trade more and grow faster after broad-based multilateral liberalization than after discriminatory liberalization.
- Overlapping trade agreements (“spaghetti-bowl of bilateral and regional trading agreements”) increase complexity and can impose administrative costs on trade.
- Yagci and Aldaz-Carroll (2004) estimate SSA would lose 0.7 percent of its GDP if all SSA countries negotiated bilateral free trade agreements (FTA) on manufacturing products with Australia, Canada, the EU, Japan, New Zealand, and the United States.
- Yang and Gupta (2005) highlight three additional cost sources:
  - (i) not all countries want to negotiate bilateral FTAs with African countries;
  - (ii) complex rules of origin can be and have been used to restrict market access;
  - (iii) SSA countries negotiating and implementing multiple bilateral FTAs face capacity constraints.

### Mozambique: market access strategy and preference erosion
- Given developments in Mozambique’s traditional export sector and other arguments, pursuing regional or bilateral agreements should be reconsidered as Mozambique’s main market access strategy.
- Reconsideration is recommended because of expected negative impact on exports from preference erosion resulting from further trade liberalization, either via more regional/bilateral FTAs or further multilateral liberalization (e.g., WTO Doha round).
- Mozambique has simplified its trading arrangements relative to most SSA countries:
  - Member of one regional trade agreement (SADC) and has avoided confusion from overlapping trade agreements.
  - Decision to suspend participation in COMESA in 1995 cited as rationale.

### Trade liberalization and poverty
- Arndt (2005) used a general equilibrium model and household survey covering 2002–03 to evaluate price-change effects on household consumption from trade liberalization.
- Conclusion: the Doha trade liberalization scenario has insignificant implications for poverty, particularly given the high value of domestic products consumption in the consumption distribution.
- Note: Mozambique’s tariffs are zero in the Doha scenario.

### Revenue and administrative constraints of trade agreements
- Market access decisions must consider potential revenue losses and administrative capacity constraints.
- Alfieri et al (2006) estimate Mozambique could lose:
  - 22 percent of its revenues on adoption of the Southern African Customs Unions (SACU),
  - 28 percent of revenues by joining the SADC,
  - 39 percent by achieving Most Favored Nation status.
- Composition of revenue losses differs by arrangement (Table 12).
- Mozambique could face administrative constraints, especially in joining SACU, which requires members to have a complex common external tariff of 38 bands and 216 rates.
- Extracts from Table III.12 as presented:
  - Actual Revenues (Millions of Meticais) SADCMFNSACU
  - Total Revenue from International Trade268283922
  - Duties and tariffs986710078
  - Excise2210727
  - VAT148539
  - Source: Alfieri, Cierera, and Rawlinson (2006).
  - (Percent of actual revenues) Estimated Loss

### Business environment, competitiveness, and sectoral data
- Success of any market access strategy depends on the business environment.
- Improving the business environment would:
  - generate long-term gains in overall productivity via more secure property rights, business-friendly regulations, cheaper inputs, adoption of new technologies, and emergence of more efficient firms;
  - better enable Mozambique to exploit current market access opportunities and potential gains from further liberalizations.
- SADC peers often have stronger and less costly business environments despite similar preferential market access and labor cost structures (see Table III.13).
- Selected figures from Table III.13 (labor cost and productivity indicators for garments) as presented:
  - Kenya 0.38 0.18 150 0.47 Yes Yes No
  - Lesotho 0.45 1/0.19 180 0.65 Yes Yes Yes
  - Mauritius 1.25 n.a. n.a. n.a. Yes Yes 1
  - Mozambique n.a. 0.16 110 0.79 Yes Yes Yes
  - Madagascar 0.33 0.16 150 0.85 Yes Yes No
  - China 0.78 3/0.29 22 n.a.000
  - India 0.38 0.17 16 n.a.000
  - Bangladesh 0.39 0.08 4/n.a. n.a.010
  - South Africa 1.38 0.65 150 0.16 Yes Yes No
  - Source: Cadot and Nasir (2001), The World Bank's Doing Business Indicators (2006).
- Mozambique’s business environment is weak even by regional standards with burdensome and ineffective trade regulations and customs procedures among its most binding constraints.

### Policy priorities and recommendations (Conclusion)
- Primary trade policy priority: make the business environment more friendly, including by reducing the burden of trade regulations for small and medium-size enterprises.
- Rationale:
  - Mozambique already has preferential access to profitable markets and a less restrictive trade regime than most SSA countries.
  - Weak business environment undermines ability to exploit market access.
  - Strengthening the business environment would permit Mozambique to better utilize current and prospective market access opportunities.
- Complementary actions:
  - Continue broad-based MFN trade liberalization alongside business environment reforms.
  - Under a better regulatory environment, trade liberalization’s impact is expected to be higher because emerging export-oriented firms would benefit from lower input costs and easier technology transfer.
- Policy priority statement (as summarized in Figure III.3):
  - Market Access: Yes/No (contextual)
  - Competitiveness: Address competitiveness constraints in order to utilize market access opportunities.
  - Outcome (3): Trade is restricted; Preferential access to SADC, E.U., and U.S. markets; Weak business environment including burdensome and ineffective custom procedures.

### Lessons for other SSA countries
- Set clear policy priorities focused on simplification of trade regulations and strengthening the business environment for small- and medium-size enterprises to improve trade.
- Avoid relying on exemptions and special incentives that favor large businesses, which can leave SMEs disadvantaged.
- Prioritize creating a fair business environment for all economic agents over pursuing complex regional trade integration when the business environment is the binding constraint.

### Appendix summary: RTA commitment and group integrity indices
- Individual-country commitment index assumptions:
  - (i) an individual country is 100 percent committed to the RTA if it is not a member of any other RTA;
  - (ii) the sum of members’ commitments to an RTA defines the organizational integrity of this group.
- Individual country commitment indices vary from 25 to 50 percent for many countries; Mozambique has a 100 percent commitment to SADC.
- Group integrity indices for RTAs as presented in Table III.A:
  - Group Integrity Index 0.42 0.40 0.35 0.28 0.34 0.29 0.43......
- The SADC is a more integrated RTA than all the other RTAs except for IGAD.

*Source: _cr07258 - 74.      The economic literature has usually regarded broad-based liberalization as the (excerpt).*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2007/_cr07258.pdf_
