## wpiea2022118-print-pdf ## Source details **Canonical URL:** [wpiea2022118-print-pdf](https://www.imf.org/-/media/files/publications/wp/2022/english/wpiea2022118-print-pdf.pdf) ## Other formats - [Markdown version](/-/media/files/publications/wp/2022/english/wpiea2022118-print-pdf.pdf.md) - [Structured JSON version](/-/media/files/publications/wp/2022/english/wpiea2022118-print-pdf.pdf.json) --- ### Background and motivation - The Ministry of Finance (MOF) of Armenia developed forecasting tools over decades (elasticity-based calculations, linear trend models, univariate regressions, multivariate regressions, error correction models, SVARs) and adopted a financial programming framework in the last 10 years to organize real, external, fiscal, and monetary sectors with satellite models. - MOF and IMF staff developed the Ararat Fiscal Strategy Model (AFSM) under a three-year Technical Assistance project to provide an in-house, structural, dynamic DSGE framework tailored to Armenian needs and fiscal rules. ### Core AFSM features and structure - Model class: New-Keynesian DSGE, small open economy with nominal price and wage rigidities. - Household heterogeneity: - Two finite-life horizon household types: savers (smooth consumption via assets and debt) and liquidity-constrained consumers (consume net income each period). - Finite horizons and liquidity constraints break Ricardian equivalence and allow non-neutral fiscal effects. - Share of liquidity constrained households (f): 0.4. - Production and sectoral structure: - Two sectors: traded and non-traded goods. - Production uses labor, private capital (sector-specific), public infrastructure (public good), and imported inputs. Labor is mobile across sectors. - Public capital accumulates via public investment and is subject to inefficiencies; government investment efficiency (ε_g): 0.6. - Fiscal block: - Spending instruments: consumption spending, transfers to households, public investment. - Revenue instruments: distortionary taxes on labor income, capital income, and consumption. - Three debt types: domestic, external commercial, and concessional debt, each with different interest rates; external commercial debt faces a country risk premium that depends on government indebtedness relative to GDP. - Policy and external environment: - Monetary policy: Taylor rule targeting price stability; weight on inflation stabilization (φ_π): 1.6; persistence of monetary policy rate (ρ_i or ρ_φ): 0.75 (reported variants: 0.75 and 0.7 in different sections). - Exchange rate: fully flexible (de jure and largely de facto flexible with occasional interventions). - Rest-of-world variables: exogenous autoregressive processes, subject to shocks (e.g., remittances persistence ρ_Rem: 0.5; persistence of foreign demand for Armenian exports ρ_Y*: 0.8). - Calibration and validation: - Calibrated to Armenian macro ratios and fiscal multipliers from SVAR estimations. - Structure based on IMF STAMP/FP 2.0 canonical DSGE framework and draws on GIMF experience. - Discount factor (β): 0.9951; Probability of survival (ζ): 0.9932; Steady-state domestic nominal risk-free interest rate (i): 9.4; Steady-state foreign nominal risk-free interest rate (i^*): 9.4. ### Fiscal rules, fiscal speed limit monitor, and institutional use - AFSM embeds Armenia’s fiscal rules framework via a fiscal speed limit monitor that captures non-linearities and multiple debt thresholds in law. - Three CG debt thresholds considered: 40, 50, and 60 percent of GDP. - Rules implemented via three logistic switches (r_n,t for n = 1,2,3): r_n,t = 1 / (1 + exp(—A (b_t — b^n))) with b^n in {40, 50, 60}; parameter A calibrated for rapid switching. - Operational rule mechanics: - If CG debt exceeds 40 percent of GDP: - The overall deficit should not be greater than capital expenditures (Rule 40). - If CG debt is between 50 and 60 percent of GDP: - The growth rate of primary current expenditures is capped by the average nominal GDP growth of the previous 7 years and a debt reduction program must be introduced (Rule 50). - If CG debt exceeds 60 percent of GDP: - Current expenditures growth cap tightened by minus 0.5 percentage points; current expenditures capped by anticipated tax revenue; Government must submit a debt reduction program to Parliament (Rule 60). - Fiscal speed limit monitor functions as a traffic light for de jure constraints; when violated the user must design corrective fiscal interventions and rerun simulations. - Default AFSM main fiscal instrument: consumption spending C_{t,j} is endogenously determined by the fiscal rule; taxes and public investment can alternatively act as main instruments. ### AFSM intended role and limitations - Intended role: quantitative policy scenario analysis to complement forecasting tools and inform budget planning and medium-term fiscal frameworks; not primarily a forecasting tool. - Limitations: satellite model used to capture occasional binding, exclusions, and instrument activations because de jure rules are non-linear and occasionally binding. ### Armenian economy: key structural facts and averages (periods noted where specified) - Population: about 3 million people. - Income: US$4,500 per capita annual income (US$10,000 in PPP terms). - Traded sector: about 53 percent of GDP (2010-2019); Non-traded: about 47 percent of GDP. - Exports composition (of tradable production): Mining 61 percent; Prepared food 20 percent; Agricultural and other 19 percent. - Employment and investment: Traded sector employs 54 percent of labor and attracts 52 percent of investment; Non-traded employs 46 percent and attracts 48 percent of investment. - Market concentration: 69 percent of the market supplied by large firms. - Growth (2010-2019): Average growth 4.5 percent; services contribution 2.7 percent; industry contribution 1.3 percent. - Expenditure averages (2010-2019): - Private consumption: 82.7 percent of GDP. - Private investment: 18.9 percent of GDP. - Government consumption and investment: 16.3 percent of GDP. - Exports: 30.6 percent of GDP. - Imports: 47.7 percent of GDP. - Household consumption composition (2010-2019 averages): Food 46 percent; Services 32 percent; Non-food incl. durables 19 percent; Housing and utilities 13 percent. - Household income shares: Wages 50 percent; Public benefits and remittances 26 percent. - Household savings: Average saving rate 17 percent (2010-2019); at least 40 percent of households not able to save. - International balances (2010-2019): Remittances 12 percent of GDP on average; primary and secondary income inflows about 10 percent of GDP; average current account deficits near 7 percent of GDP. During 2015-2019 trade deficits average 13 percent of GDP and current account deficits average 3 percent of GDP. - Fiscal averages (2010-2019): Overall fiscal deficit 3.0 percent of GDP; Primary deficit 1.5 percent of GDP; Revenues 21.7 percent of GDP (Personal income taxes 5.9 percent; Corporate income taxes 3.2 percent; Consumption taxes 10.6 percent); Primary expenditures 24.5 percent of GDP (Current spending excl. debt service 12.8 percent; Capital spending 3.6 percent; Social and other transfers 7.0 percent); Interest payments about 1.5 percent of GDP. - Government debt dynamics: central government debt rose from 13.9 percent to 54 percent of GDP in 2017 and declined to 50 percent in 2019; debt composition about 80 percent external and most external debt concessional; last Eurobonds issuance in 2019. ### AFSM household sector (structure and implications) - Households live on a unit circle with two groups: - (1 − f) share liquidity constrained households (non-savers) consume current income each period. - f share savers have access to financial assets and smooth consumption. - Aggregation: aggregate consumption C_t = C_t^o + C_t^l; savers’ consumption linear in wealth P_t C_t^o = MPC_t ℧_t where MPC_t can be shocked. - Remittances process: Rem_t^* = (1 − ρ_Rem) Rem̅^* + ρ_Rem Rem_{t-1}^* + ε_t^Rem with ε_t^Rem ~ N(0, σ_Rem^2). - Utility and budget constraints preserve notation and parameters including habit parameter χ: 0.65 and relative risk aversion σ: 1. ### AFSM government sector: rules, processes, and calibration highlights - Monetary policy rule: i_t = ρ_i i_{t-1} + (1 − ρ_i) \bar{i} + φ_π (π_t − π^*) + φ_y \tilde{y}_t + ε_t^i with ε_t^i ~ N(0, σ_i^2). - Fiscal authority budget constraint preserves notation for domestic debt B_{t,j}, foreign commercial B_{t,j}^{fm}, foreign concessional B_{t,j}^{fgm}; total nominal domestic-currency debt B_t = B_{t,d} + S_t B_{t,dm} + S_t B_{t,dgm}. - Primary deficit rule (counter-cyclical, debt-stabilizing): - pd_t = (1 − ρ_pd) pd* + ρ_pd pd_{t-1} − ρ_y \tilde{y}_t − ρ_b (b_t − \bar{b}) + ε_t^{pd} with ε_t^{pd} ~ N(0, σ_{pd}^2). - Weight on business cycle stabilization ρ_y: 0.06; weight on debt stabilization ρ_b: 0.06. - Financing shares: share of fiscal deficits financed domestically φ_f: 0.1789. - Public capital accumulation: G_{t,s} = (1 − δ_g) G_{t-1,s} + ε_g I_{t,g,s} with depreciation of public capital δ_g: 0.02 (quarterly or calibrated consistent with 8 percent annually elsewhere); government investment efficiency ε_g: 0.6. - External financing and risk premia: - 1 + i_t^m = (1 + i_t^*) prem_t with prem_t = ξ [ (B_{t-1}^{gf} − B_{t-1}^{gcf} + B_{t-1}^{p,f}) / GDP_{t-1} − NFD^* ] where ξ measures sensitivity; private net foreign liabilities b^p: 4*0.45. - Key fiscal targets (steady-state shares): - Domestic public debt target b̄_dom: 0.103. - Foreign public debt target b̄_for: 0.078. - Foreign public concessional debt target b̄_fcon: 0.321. - Annual nominal interest rate on concessional loans i_g: 1.2. - Public investment long-run level: 0.02 (long-run level of (non)tradable-sector-specific public capital investment). - Long-run net transfers to liquidity constrained households T̅_l: 0.13; to liquidity unconstrained households T̅_o: −0.06. ### Key calibration and steady-state parameter values (preserved exactly where given) - Exogenous annual growth rate: 4 percent. - Steady-state growth rate, quarterly (g): 0.0125. - Share of public capital in non-tradable and tradable sectors (ψ): 0.1. - Price elasticity of demand for final consumption goods (θ_c): 6; price elasticity of labor demand (θ_n): 6. - Intra-temporal elasticity of substitution of consumption goods (θ_s): 0.75; investment goods (θ_i): 0.75. - Price elasticity of foreign demand for Armenian exports (θ_x): 1.5. - Share of domestic goods in consumption (ω): 0.35; share of domestic traded goods in consumption (κ): 0.3. - Technology levels: A̅_N: 1.5; A̅_T: 1. - Remittances (share of GDP) Rem^*: 0.119. - Depreciation of physical capital in tradable and non-tradable sectors (δ^T, δ^N): 0.01. - Wage adjustment costs (ζ_w): 180; price adjustment costs ζ_p^N, ζ_p^T: 96. - Habit persistence χ: 0.65. - Persistence parameters (selected): ρ_Y*: 0.8; ρ_Rem: 0.5; ρ_tot: 0.5; ρ_i*: 0.7; ρ_π*: 0.4; ρ_A^N, ρ_A^T: 0.95. - Long-run effective tax rates: τ̄_c: 0.13; τ̄_w: 0.15; τ̄_k^T: 0.12; τ̄_k^N: 0.19. - Public investment and transfers steady-state shares: Total public investment 3.8 percent of GDP (split equally between non-traded and traded sector-specific investment); Total net transfers about 10 percent of GDP; Transfers per household group T̅_l / GDP = 0.13; T̅_o / GDP = −0.06. - Great ratios (model steady state column, percent of GDP): - Private consumption expenditures: 80.3 (domestic goods 57.0; imported goods 23.5). - Gross private fixed investment: 16.1. - Government expenditures: 16.0 (government consumption 12.9; government investment 3.1). - Exports: 35.3; Imports: 48.1 (final consumption goods*: 18.1; intermediate goods**: 30.0). - Total tax revenue (excl. other taxes): 19.7 (income tax 6.2; profit tax 3.2; consumption tax 10.3). - Total central government debt: 50.2 (domestic 10.3; external 39.9 — market 7.8; concessional 32.1). - Overall deficit: 3.5 (primary deficit 1.4; interest costs 2.1). ### Validation: fiscal multipliers (AFSM vs empirical SVAR estimates) - Multipliers defined as cumulative change in output over cumulative change in relevant fiscal variable. Headline AFSM numbers with bracketed empirical ranges from Fukač et al. (2021): - Tax revenues: - Consumption tax: Impact multiplier −0.7 [empirical: −1.8, −0.4]; 1-Year Multiplier −0.5 [empirical: −2.3, −0.2]; 3-Year Multiplier −0.4 [empirical: −1.7, 0.2]. - Personal income tax: Impact −0.4 [empirical: −4.4, 0.3]; 1-Year −0.7 [empirical: −2.7, 0.8]; 3-Year −1.3 [empirical: −1.7, 0.6]. - Corporate profit tax: Impact −0.2 [empirical: −4.4, −0.9]; 1-Year −0.2 [empirical: −2.3, 0.8]; 3-Year −0.4 [empirical: −1.7, 0.3]. - Expenditures: - Current expenditures: Impact 0.8 [empirical: −0.2, 1.4]; 1-Year 0.6 [empirical: −0.3, 1.0]; 3-Year 0.5 [empirical: −0.3, 0.4]. - Capital expenditures: Impact 1.5 [empirical: 0.1, 1.1]; 1-Year 0.9 [empirical: −0.7, 1.1]; 3-Year 0.9 [empirical: −0.4, 0.7]. - Conclusion: AFSM-implied multipliers fall within measured empirical ranges; expenditure multipliers are relatively potent in AFSM calibration. ### Capturing Armenia’s fiscal rules: satellite model mechanics and fiscal speed limit monitor - Satellite model implements the three rules via logistic switches r_{n,t} with thresholds b^n in {40, 50, 60}. - Rule-specific mechanics (notations preserved): - Rule 40 (capital expenditures constraint): P_t Ī_{t,g} = (1 − r_{1,t}) P_t I_{t,g} + r_{1,t} D_t. If b_t ≤ 40 then r_{1,t} = 0; if b_t > 40 then 0 < r_{1,t} ≤ 1 and capital expenditures constrained to overall deficit D_t. - Rule 50 (current expenditures growth cap): P_{c,t} C̄_{t} = (1 − r_{2,t}) P_{c,t} C_t + r_{2,t} (1 − ∆_{20NGDP,t}) P_{c,t} C̄_{t−1}. If b_t ≤ 50 then r_{2,t} = 0; if b_t > 50 then 0 < r_{2,t} ≤ 1 and nominal expenditures increase at average nominal GDP growth over past 20 quarters ∆_{20NGDP,t}. - Rule 60 (current expenditures capped by tax revenue net of deductibles): If b_t > 60 then 0 < r_{3,t} ≤ 1 and nominal expenditures equal R_t (total tax revenues less corporate tax deductibles) per the formula in source. - Implementation notes: - Overlaps among Rules 40, 50, and 60 are non-trivial; satellite extension useful for studying their interactions. - Fiscal speed limit monitor issues traffic-light indicators for each rule and requires corrective interventions when de jure limits are violated. ### Illustrative experiment: rule activation dynamics (2020 Q1–2021 Q4 shocks) - Shocks to primary deficits designed to raise public debt to breach 50 percent of GDP so Rule 50 binds. - Outcomes: - Current expenditures increase by 30 percent (debt financed), causing public debt to increase. - As debt exceeds 50 percent threshold, Rule 50 switch r_{2,t} transitions from 0 to 1 and prescribes faster expenditure reduction; debt levels off in 2022 and begins returning to target. - Rule 50 rolls back as debt falls below 50 percent and r_{2,t} returns to 0. - Rule 40 not activated in this illustration but monitor flags capital spending less than overall deficit; once Rule 50 binds it helps align capital expenditures with Rule 40 prescriptions. - Rule 60 remains satisfied throughout experiment. ### 6.1 Assessment of the First Wave of the Covid-19 Pandemic: calibration and shock design - Calibration sources: foreign shocks from IMF forecasts of October 2020; domestic shocks and judgments from MOF. - Key calibration judgments: - Export demand in 2020 contracts on average by 7.4 percent; assumed export-demand growth 2.2 percent in 2021. - Remittances decline by 9.5 percent in 2020 and recover by 4.1 percent in 2021. - Country-risk premium calibrated to match increases in Eurobond spreads of 0.8 and 0.4 percentage points in 2020 and 2021, respectively. - Absent other shocks consumption expected to decrease by 7.8 percent in 2020 and increase by 2.2 percent in 2021. - Total investment absent other shocks calibrated to contract by 21.7 percent in 2020 and increase by 2.2 percent in 2021, returning to trend by 2023. ### 6.1 AFSM simulation results: macro-fiscal impacts (key quantified impacts) - GDP decreases by 12.9 percent in 2020 and slowly recovers over the horizon. - Loss in GDP: 13 percent in 2020 and 38 percent over 2020-2023 relative to pre-crisis level (summary in Introduction). - Export demand contracts on average by 7.4 percent in 2020; remittances decline by 9.5 percent in 2020 and recover by 4.1 percent in 2021. - Country-risk premium increases calibrated at 0.8 and 0.4 percentage points in 2020 and 2021, respectively. - Inflation will be 2.5 percentage points below its target in 2020 and returns to it in the medium-term. - The Central Bank reduces the policy interest rate by 2.1 percentage points in response to lower inflation. - Nominal depreciation about 4 percent in 2020; in the medium-term the exchange rate appreciates. - In Q3 2020 GDP expected to fall by almost 15 percent below its long-term potential level. - By end-2023 public debt-to-GDP ratio can increase by about 19 percentage points and stay higher absent fiscal consolidation. - Two thirds of macro-fiscal effects due to external shocks; foreign demand shock explains more than 75 percent of the almost 15 percent GDP fall in 2020 and more than 75 percent of the almost 20 percentage points increase in public debt-to-GDP by end-2023. ### Fiscal deterioration and public debt dynamics during pandemic - Tax to GDP ratio projected to decrease by 3.7 and 2.7 percent below the long run levels in 2020 and 2021, respectively, and remain at 1 percent below steady-state over medium term. - Primary deficit widens by 6.5 and 4.7 percentage points of GDP in 2020 and 2021, respectively. - Nominal exchange rate depreciation exacerbates public debt burden due to important share of foreign-currency debt. - Primary deficit fiscal rule helps stabilize debt in the medium term, avoiding explosive dynamics. ### Fiscal response simulated and quantified effects (Armenia case) - Government announced socio-economic support package AMD 150 billion (2.5 percent of GDP) implemented in Q2 2020, plus corporate income tax moratorium. - AFSM-modeled main programs (total budgeted costs AMD 215 billion cushion about AMD 303 billion (4.6 percent of GDP) in contracting economy): - Economic support program: AMD 33.5 billion (0.5 percent of GDP) modeled as current spending shock: 0.26 percent of quarterly GDP Q2 2020; 0.19 percent Q3 2020; 0.08 percent Q4 2020. - Social support and wage subsidy program: AMD 36.5 billion (0.6 percent of GDP) modeled as transfers to liquidity-constrained households: 0.5 percent of GDP Q2 2020; 0.07 percent Q3 2020; 0.03 percent Q4 2020. - Corporate income tax deferral and reform: AMD 65 billion (1.1 percent of GDP) modeled as corporate profit tax shocks: −0.71 percent of GDP Q2 2020; −0.23 percent Q3 2020; −0.12 percent Q4 2020. - Private investment support program: AMD 80 billion (1.3 percent of GDP) modeled via shock to real return on capital in traded sector: 40 percent of program amount in Q3 2020 (0.51 percent of GDP) and rest in Q4 2020 (0.77 percent of GDP). - Macroeconomic and fiscal effects of the fiscal package (AFSM results): - Total budgeted costs AMD 215 billion cushion about AMD 303 billion (4.6 percent of GDP). - In 2020 the program can cushion GDP by 1.9 percentage points and employment by 3.7 percentage points (protecting 35,460 jobs). - AFSM projects the primary balance can deteriorate by 2.6 percentage points of GDP, relative to baseline, from −9.5 to −12.1 percent. - Public debt as share of GDP projected to increase by 3 percentage points relative to baseline and then stabilize in medium term. - AFSM indicates a larger fiscal package could push public debt to unsustainable levels given weak outlook and increase debt service burden. - Government activated escape clause of fiscal rule as public debt surpassed 60 percent of GDP; escape clause also in place in 2021; in 2022 Government committed to adhere to fiscal rules and implement a 5-year fiscal consolidation program. ### Quantitative net impacts on GDP and components (Table 4: percentage deviations from long-term trend) - 2020: - GDP: Baseline −12.9; Fiscal Policy −11.0 - Consumption: Baseline −11.7; Fiscal Policy −9.5 - Investment: Baseline −21.7; Fiscal Policy −19.4 - Exports: Baseline −23.4; Fiscal Policy −21.8 - Imports: Baseline −18.4; Fiscal Policy −15.3 - 2021: - GDP: Baseline −10.4; Fiscal Policy −9.7 - Consumption: Baseline −9.5; Fiscal Policy −8.8 - Investment: Baseline −19.9; Fiscal Policy −17.9 - Exports: Baseline −23.6; Fiscal Policy −22.2 - Imports: Baseline −19.1; Fiscal Policy −17.3 - 2022: - GDP: Baseline −6.7; Fiscal Policy −6.0 - Consumption: Baseline −7.2; Fiscal Policy −6.7 - Investment: Baseline −14.5; Fiscal Policy −13.2 - Exports: Baseline −16.2; Fiscal Policy −14.8 - Imports: Baseline −15.2; Fiscal Policy −14.1 - 2023: - GDP: Baseline −4.4; Fiscal Policy −3.7 - Consumption: Baseline −5.5; Fiscal Policy −4.8 - Investment: Baseline −9.8; Fiscal Policy −8.6 - Exports: Baseline −11.0; Fiscal Policy −9.6 - Imports: Baseline −11.3; Fiscal Policy −10.2 ### 6.2 Comparing alternative policies in the context of public investment (under-execution experiments) - Context: In 2018 and 2019 capital expenditures under-executed by 30 percent; AFSM used in fall 2019 to evaluate options assuming continued under-execution. - Baseline (replicating 2020-2022 MTEF): - Permanent increase of public investment by 1.9 percentage points of GDP in first year, then 0.3 percentage points annually. - Permanent decrease of current expenditures by 0.2 percentage points of GDP for entire period. - Increase of tax revenues equivalent to 0.3 percent of GDP starting 2021. - Financing: debt accumulation in first year and additional tax revenues thereafter; debt-to-GDP assumed to increase by 0.2 percentage points in 2020 then stay constant. - Outcomes: expansion of GDP, crowding-in of private investment and consumption; public debt accumulates about 1 percentage point of GDP. - Alternative 1: Debt repayment (reallocate saved funds to foreign public debt repayment) - Outcomes: lower productive capacity, lower output and employment relative to baseline; fiscal benefit: deleveraging and improved fiscal space medium-term. - Alternative 2: Increasing social benefits (reallocate saved funds to pensions and social transfers) - Outcomes: boosts private consumption among liquidity-constrained households but does not expand productive capacity; reduces investment and GDP relative to baseline; debt-to-GDP ratios breach 50 percent; current expenditures in simulation not adjusted in response to debt accumulation. - Alternative 3: Increasing public investment efficiency - Context: baseline public investment efficiency 60 percent (one Dram of public investment translates into 0.6 cents of public capital). - Required efficiency increase: about 16.5 percentage points to match baseline GDP with continued under-execution. - Outcomes: similar GDP effects as baseline and improves fiscal outlook; leads to lower public expenditures and declining medium-term public debt relative to baseline; implementation and timing caveats noted. - Comparative conclusions: - Increasing public investment efficiency dominates debt repayment and higher social spending in macroeconomic outcomes (GDP) and debt developments. - Trade-offs: debt repayment improves fiscal metrics but at cost of productive capacity; social transfers help vulnerable households but can worsen fiscal balances and reduce medium-term GDP. - Policy implication: structural reforms to improve public investment efficiency are central but achieving a 16.5 percentage point gain over simulation period is ambitious. ### Model applications, use cases, and validation summary - AFSM designed to: - Quantify macroeconomic effects of fundamental shocks and policies. - Serve as flexible policy scenario analysis tool to help formulate forward-looking fiscal strategies. - Accommodate design and assessment of fiscal interventions when fiscal rule monitors flag breaches. - Validation against empirical fiscal multipliers shows AFSM predictions fall within empirical ranges, with expenditure policies calibrated to be relatively potent reflecting prospective governance improvements. *Source: IMF working paper—Introduction section of the Ararat Fiscal Strategy Model documentation.* ### 1. Introduction ........................................................................................................ ### wpiea2022118-print-pdf - 1. Introduction ### Background and motivation - The Ministry of Finance (MOF) of Armenia developed forecasting tools over decades (elasticity-based calculations, linear trend models, univariate regressions, multivariate regressions, error correction models, SVARs) and adopted a financial programming framework in the last 10 years to organize real, external, fiscal, and monetary sectors with satellite models. - MOF and IMF staff developed the Ararat Fiscal Strategy Model (AFSM) under a three-year Technical Assistance project to provide an in-house, structural, dynamic DSGE framework tailored to Armenian needs and fiscal rules. ### Core AFSM features and structure - Model class: New-Keynesian DSGE, small open economy with nominal price and wage rigidities. - Household heterogeneity: - Two finite-life horizon household types: savers (smooth consumption via assets and debt) and liquidity-constrained consumers (consume net income each period). - Finite horizons and liquidity constraints break Ricardian equivalence and allow non-neutral fiscal effects. - Production and sectoral structure: - Two sectors: traded and non-traded goods. - Production uses labor, private capital (sector-specific), public infrastructure (public good), and imported inputs. - Labor is mobile across sectors. - Public capital accumulates via public investment and is subject to inefficiencies. - Fiscal block: - Spending instruments: consumption spending, transfers to households, public investment. - Revenue instruments: distortionary taxes on labor income, capital income, and consumption. - Three debt types: domestic, external commercial, and concessional debt, each with different interest rates; external commercial debt faces a country risk premium that depends on government indebtedness relative to GDP. - Policy and external environment: - Monetary policy: Taylor rule targeting price stability. - Exchange rate: fully flexible. - Rest-of-world variables: exogenous autoregressive processes, subject to shocks. - Calibration and validation: - Calibrated to Armenian macro ratios and fiscal multipliers from SVAR estimations. - Structure based on IMF STAMP/FP 2.0 canonical DSGE framework and draws on GIMF experience. ### Fiscal rules, fiscal speed limit monitor, and institutional use - AFSM embeds Armenia’s fiscal rules framework via a fiscal speed limit monitor that captures non-linearities and multiple debt thresholds in law. - The monitor functions as a traffic light for de jure constraints on current and capital expenditures and signals when expenditures violate statutory limits, requiring the user to design corrective fiscal interventions and rerun simulations. - AFSM intended role: quantitative policy scenario analysis to complement forecasting tools and inform budget planning and medium-term fiscal frameworks; not primarily a forecasting tool. ### Policy scenario applications and key quantitative findings - COVID-19 “first wave” application: - AFSM estimated real GDP and consumption declines of 12.9 and 11 percent, respectively, in 2020. - Loss in GDP: 13 percent in 2020 and 38 percent over 2020-2023 relative to pre-crisis level. - Debt dynamics: debt-to-GDP ratios projected to increase by 18.8 percentage points in less than 5 years and remain higher in the medium term. - Shock decomposition: external shocks (contraction in foreign demand, remittances, tourism, and higher country risk premium) accounted for the largest share of macro and fiscal effects, more than domestic shocks (decline in marginal propensity to consume and investment appetite). - Fiscal response evaluation: - An initial fiscal package of 3.6 percent of GDP was simulated as effective from a macro-fiscal cost-benefit perspective. - With the package, real GDP and consumption would decline by almost 2 percentage points less in 2020 relative to no fiscal action. - Cumulative cushioning of real GDP by fiscal measures: 4.6 percent over 2020-2023. - Jobs protected: 35460. - Public investment under-execution (2018) alternative policies comparison: - Baseline: official Fiscal framework for 2020-2022. - Alternatives assessed assuming continued capital under-execution: - (i) Reallocating saved budget funds to debt repayments. - (ii) Increasing social spending. - (iii) Increasing public investment efficiency. - Key quantitative insights: - Increasing public investment efficiency dominated debt repayment and higher social spending in macroeconomic outcomes (GDP) and debt developments. - To match baseline GDP, efficiency would need to increase by 16.6 percentage points from a baseline value of 60 percent (baseline assumption: one Dram of public investment translates into 0.6 cents of public capital). - Debt repayment and efficiency gains could both support declining medium-term debt burdens, whereas increasing social transfers could create debt sustainability challenges. ### Intended use and limitations - AFSM is primarily a simulation tool for quantifying macroeconomic effects and trade-offs of policy alternatives consistent with Armenian structural features and fiscal rules. - AFSM-based analysis is intended to complement other forecasting and policy tools used by MOF for budget planning and medium-term fiscal frameworks. *Source: IMF working paper—Introduction section of the Ararat Fiscal Strategy Model documentation.* ### Section 5 describes how the taxonomy of the fiscal rules in Armenia is captured in the AFSM ### Section 5 describes how the taxonomy of the fiscal rules in Armenia is captured in the AFSM ### Overview of the Armenian economy - Population: about 3 million people. - Income: US$4,500 per capita annual income (US$10,000 in PPP terms). - Trade and remittances: highly import dependent; inflows of remittances help finance large trade deficits. - Geography and infrastructure: landlocked; borders with Azerbaijan and Turkey are closed; Georgia and Iran are main ground trade routes; roads and railways past their lifecycle. ### Production structure and sector shares - Traded vs non-traded: - Traded sector corresponds to about 53 percent of GDP (period 2010-2019). - Non-traded sector represents about 47 percent of GDP. - Exports composition (of tradable production): - Mining: 61 percent of exports. - Prepared food: 20 percent of exports. - Agricultural and other: 19 percent of exports. - Employment and investment: - Traded sector employs 54 percent of the labor force and attracts 52 percent of total investment. - Non-traded sector employs 46 percent of the labor force and attracts 48 percent of total investment. - Market concentration: 69 percent of the market supplied by large firms. - Growth (2010-2019): - Average growth: 4.5 percent, mostly driven by total factor productivity. - Contributions: services average 2.7 percent, industry average 1.3 percent. - Construction sector: - At its boom increased GDP contribution to 25 percent in 2000s. - After Global Financial Crisis shrank by about 2.3 percent annually and reduced GDP contribution to 6 percent. ### Expenditure side and resource use - Labor share of gross value added (2010-2019): about 40 percent. - Imports: - Industrial supplies: about 14 percent of GDP. - Capital goods: about 7 percent of GDP. - Expenditure averages (2010-2019): - Private consumption: 82.7 percent of GDP. - Private investment: 18.9 percent of GDP. - Government consumption and investment: 16.3 percent of GDP. - Exports: 30.6 percent of GDP. - Imports: 47.7 percent of GDP. - Household consumption composition (2010-2019 averages): - Food products: 46 percent of total consumption. - Services: 32 percent. - Non-food products, including durables: 19 percent. - Housing and utilities: 13 percent. - Household income shares: - Wages: 50 percent of total income. - Public benefits and remittances: 26 percent of total income. - Household savings: - Average saving rate: 17 percent (2010-2019). - At least 40 percent of households are not able to save (captured in the AFSM). ### International balances (2010-2019) - Trade deficits: almost 20 percent of GDP. - Remittances: 12 percent of GDP on average. - Primary and secondary income inflows: about 10 percent of GDP. - Average current account deficits: near 7 percent of GDP. - During 2015-2019: - Trade deficits average: 13 percent of GDP. - Current account deficits average: 3 percent of GDP. ### Monetary and exchange rate policy - Inflation targeting: - Adopted in 2016 by the Central Bank of Armenia (CBA). - Target: stabilize 12-month consumer price inflation at 4 percent with 1.5 percentage points of tolerance. - Intermediate target: 12-quarter inflation forecast. - Instrument: refinancing rate. - Exchange rate: de jure and largely de facto flexible exchange rate with occasional interventions to smooth volatility. - Capital account: liberalized without major restrictions on capital flows. - Inflation and exchange rate outcomes (2010-2019): - Average CPI inflation: 3.5 percent (oscillating between -2.1 and 11.5 percent). - Nominal effective exchange rate appreciated by 40 percent. - Real effective exchange rate appreciated by 15 percent. - AMD/USD nominal exchange rate depreciated by 21 percent. ### Fiscal outcomes, composition, and debt - Fiscal averages (2010-2019): - Average overall fiscal deficit: 3.0 percent of GDP. - Primary deficit: 1.5 percent of GDP. - Revenues: 21.7 percent of GDP. - Personal income taxes: 5.9 percent of GDP. - Corporate income taxes: 3.2 percent of GDP. - Consumption taxes: 10.6 percent of GDP. - Primary expenditures: 24.5 percent of GDP on average. - Current spending (excluding debt service): 12.8 percent of GDP. - Capital spending: 3.6 percent of GDP. - Social and other transfers: 7.0 percent of GDP. - Interest payments related to debt service: about 1.5 percent of GDP. - Government debt dynamics: - Central government debt rose from 13.9 percent to 54 percent of GDP in 2017 after the Global Financial Crisis. - Debt declined to 50 percent of GDP in 2019 following fiscal consolidation. - Debt composition: - About 80 percent external. - Most external debt is concessional. - Remaining debt is domestic. - Since 2013 the Government has accessed international capital markets; last Eurobonds issuance in 2019. - Concerns: high share of external debt raises currency risk. ### Fiscal rules taxonomy and operational speed limits (as specified in Laws on the Budget System and Public Debt) - Government long-term objective: reduce debt level below 40 percent of GDP ("safe level"). - Three CG debt thresholds considered: 40, 50, and 60 percent of GDP. - General design: - Rules are operational expenditure rules with debt thresholds serving as medium-term fiscal anchors. - Numerical parameters calibrated to Armenian economy characteristics and business cycle stabilization. - Rules respect the "golden rule" of public finance and include escape clauses for large-scale natural disasters, wars, or economic shocks (e.g., Covid-19). - Required actions by threshold breach: - If CG debt exceeds 40 percent of GDP: - The overall deficit should not be greater than capital expenditures. - If CG debt is between 50 and 60 percent of GDP: - The previous rule applies. - The growth rate of primary current expenditures is capped by the average nominal GDP growth of the previous 7 years. - The Government must introduce a debt reduction program as part of its Medium-Term Expenditures Framework, which is a public document. - If CG debt exceeds 60 percent of GDP: - The previous two rules apply. - The growth rate of primary current expenditures is capped by the average nominal GDP growth of the previous 7 years minus 0.5 percentage points. - Current expenditures are capped by the anticipated volume of tax revenues. - The Government must submit a debt reduction program to the Parliament. - Legal and procedural notes: - Escape clause regulations set in The Law on the Budget System of the Republic of Armenia (Article 21, Paragraph 8.3). - Government Decree of August 23, 2018 (N942-Ն) sets the procedure for defining the exceptional situation that triggers the escape clause. ### Fiscal policy effects and multipliers - Ministry of Finance view: fiscal policy can have significant effects on GDP. - Empirical multipliers: estimates in Fukač et al. (2021) using SVARs for different types of taxes and expenditures used to validate AFSM calibration (summary referenced in Table 3 of source). ### AFSM model structure and assumptions (relevant to capturing fiscal rules) - Four high-level blocks: (i) household sector, (ii) production sector, (iii) government sector, and (iv) rest of the world; with different agents in each block. - Key modeling assumptions: - Private agents forward-looking, rational expectations, complete information, and solve intertemporal maximization problems subject to constraints. - Monetary and fiscal authorities follow policy rules targeting key macro variables (e.g., inflation and debt). - Some markets have monopolistic structures (e.g., non-traded goods), leading to Pareto sub-optimal allocations that may justify government intervention. - Markets clear and fluctuations are driven by fundamental shocks. - Technology grows exogenously at rate 푔1; for simplicity in AFSM description 푔1 = 0. *Source: wpiea2022118-print-pdf - Section 5 describes how the taxonomy of the fiscal rules in Armenia is captured in the AFSM* ### 3.1 Household Sector ### 3.1 Household Sector ### Model structure and motivation - Households live on a unit circle and are divided into two groups: - (1-f) share of liquidity constrained households (non-savers) who consume fully their current income in every period. - f share of liquidity unconstrained households (savers) who have access to financial assets and firms’ dividends and can smooth consumption over time. - Interpretation: savers ≈ middle/higher income households; non-savers ≈ lower-income households. - Importance of liquidity-constrained households in the AFSM: - Allow capture of fiscal policy Keynesian effects and influence the size of fiscal multipliers. - Generate more volatile booms and busts, motivating a larger role for fiscal policy in macro stabilization. - Help break Ricardian equivalence. ### Consumption aggregators and price index - Households consume a CES basket composed of domestic non-traded goods, domestic traded goods, and foreign traded goods with shares 휔, 휅, and 1ෙ휅ෙ휔 respectively, satisfying 휅൅휔൏1, and elasticity of substitution 휃஼. - Demand functions (expressed schematically): - 퐶ௗ,௧௜ = 휔 (푃ௗ,௧ / 푃௧)^(-휃஼) 퐶௧௜ - 퐶௛,௧௜ = 휅 (푃௛,௧ / 푃௧)^(-휃஼) 퐶௧௜ - 퐶௙,௧௜ = (1ෙ휔ෙ휅) (푃௙,௧ / 푃௧)^(-휃஼) 퐶௧௜ - Aggregate consumer price index: - 푃௧ = [휔 푃ௗ,௧^(1ෙ஼) + 휅 푃௛,௧^(1ெ஼) + (1ෙ휔ෙ휅) 푃௙,௧^(1ெ஼)]^(1/(1ெ஼)) ### Liquidity-unconstrained households (Savers) - Key institutional detail: savers face a probability (1-휉) of dying each period (Blanchard (1985), Yaari (1965)); 휉 calibrated to match average productive life in Armenia. Dying savers are immediately replaced; population size constant. - Savers maximize lifetime welfare choosing consumption, labor supply, and portfolio of domestic and foreign bonds: - Objective (schematic): max E_t ∑ (훽휉)^(s) [ (1/(1ෙ휎)) (퐶_o,a,t^(1ෙ휎)) - (1/(1ெ푁)) N_o,a,t^(1ெ푁) ] with habit parameter 휒 and consumption weight 훾 present in utility. - Budget constraint (nominal terms, schematic): - (1+휏^c_t) 푃_t 퐶_o,a,t + 퐵_o,a,t + S_t 퐵_o,a,t^*, ... = (1 - 휏^n_t) V_t N_o,a,t + (1/휉)[(1+ i_t-1) B_o,a,t-1 + (1+ i^*_t-1) S_t B_o,a,t-1^*] + D_o,a,t + P_t T_o,a,t - Key equilibrium conditions and implications: - Aggregation yields savers’ consumption linear in wealth ℧_t: 푃_t 퐶_t^o = MPC_t ℧_t, where MPC_t is the marginal propensity to consume (can be shocked). - Implied intra-temporal condition (labor supply): marginal rate of substitution between consumption and labor equals real wages (implicit savers’ labor supply), with 훹 the aggregate weighted time-endowment of savers and taxes creating distortion. - UIP condition: - (1+ i_t) = (1+ i_t^*) E_t [ (S_{t+1} / S_t) ] , equalizing net returns on domestic and foreign bonds adjusted for expected nominal exchange rate depreciation. ### Liquidity-constrained households (Non-Savers) - Non-savers do not save; consume all disposable income each period. - Non-savers maximize lifetime utility from consumption and leisure with the same parameters 휉, 훽, 훾, 휎 as savers. - Budget constraint (real disposable income limit, schematic): - (1+휏^c_t) P_t C_l,a,t ≤ (1 - 휏^n_t) V_t N_l,a,t + P_t T_l,a,t + S_t Rem_a,t^* - Disposable income components: labor income V_t N_l,a,t; government net transfers P_t T_l,a,t; foreign remittances S_t Rem_a,t^*. - Effective income tax and wage rates are identical for savers and non-savers. - Optimal aggregated labor supply (non-savers): - N_t^l = 1ෙ훹ෙ∑(...) [function negatively related to consumption and positively related to real wage], where willingness to work falls with consumption and rises with real wages. - Given labor supply and disposable income, budget constraint determines non-savers’ consumption. ### Remittances process - Aggregate remittances (foreign currency) are exogenous and follow an AR(1): - Rem_t^* = (1ෙ휌_Rem) Rem̅^* + 휌_Rem Rem_{t-1}^* + ε_t^Rem - Rem̅^* is steady-state remittances; 휌_Rem is autoregressive parameter; ε_t^Rem ~ N(0, σ_Rem^2). ### Aggregation - Aggregate variable X_t defined as sum of savers’ and non-savers’ components: - X_t = X_t^o + X_t^l - Example: aggregate consumption C_t = C_t^o + C_t^l *Source: 3.1 Household Sector, wpiea2022118-print-pdf* ### 3.3 Government Sector ### 3.3 Government Sector ### Monetary Authority - The Central Bank (CB) has a dual mandate: primary objective is price level stability and secondary objective is supporting economic growth. - Main policy instrument: short-term risk-free interest rate set following a Taylor-type rule: - i_t = ρ_i i_{t-1} + (1−ρ_i) \bar{i} + φ_π (π_t − π^*) + φ_y \tilde{y}_t + ε_{t}^i - Where \bar{i} is the natural interest rate; π^* is the CPI inflation target; \tilde{y}_t is the output gap; ε_{t}^i ~ N(0,σ_i^2) is a monetary policy shock. - Parameters: ρ_i is the interest smoothing coefficient; φ_π ≥ 1 is the response coefficient to inflation stabilization; φ_y ≥ 0 is the weight on output gap stabilization. ### Fiscal Authority: Structure, Instruments, and Budget Constraint - The Fiscal Authority (FA) conducts revenue, expenditure, and debt policies within a debt sustainability framework. - Expenditure instruments: - Current expenditures: C_{t,j} - Transfers: T_t (T_t = T_{t,l} + T_{t,h} across households) - Public investment: I_{t,j,s} for i ∈ {h,d} - Tax instruments and rates: - Consumption tax: τ_c,t - Labor income tax: τ_w,t - Capital/profit taxes sector-specific: τ_k,t^s and τ_k,t^n (notations preserved as in source) - Borrowing instruments and associated nominal interest rates: - Domestic debt: B_{t,j} with interest i_t - Foreign commercial debt: B_{t,j}^{fm} with interest i_t^{m} - Foreign concessional debt: B_{t,j}^{fgm} with interest i_t^{gm} - Total nominal domestic-currency debt: - B_t = B_{t,d} + S_t B_{t,dm} + S_t B_{t,dgm} - Primary deficit definition (D_t) and government budget constraint: - Primary deficit D_t equals expenditures (P_{n,t} C_{t,n} + P_{m,t}^I I_{t,m,s} + P_{n,t}^I I_{t,n,s} + P_t T_t + …) minus tax and non-resource revenue terms, plus changes in nominal debt components (ΔB_{t,j} terms). - Notation: ΔB_{t,j} = B_{t,j} − B_{t-1,j} for j ∈ {g,gf, gcf}. ### Fiscal Policy Processes and Rules - Net financial transfers to each household type (i ∈ {l,k}) and tax rates follow exogenous autoregressive processes: - T_{t,s} = ρ_T T_{t-1,s} + (1−ρ_T) \bar{T}_s + ε_{t}^{T_s} - τ_{c,t} = (1−ρ_{τc}) \bar{τ}_c + ρ_{τc} τ_{c,t-1} + ε_{t}^{τc} - τ_{w,t}, τ_{k,t}^{s}, τ_{k,t}^{n} analogous with their respective ρ and ε terms. - Shocks ε are normally distributed with variances σ^2 as specified for each process. - Public capital accumulation (sector-specific G_{t,s} for s ∈ {d,h}): - G_{t,s} = (1−δ_g) G_{t-1,s} + ε_g I_{t,g,s} - Government investment efficiency parameter ε_g (can be shocked) captures inefficiencies in converting investment into productive public capital. - Sector-specific public investment I_{t,g,s} follows an exogenous AR process: - I_{t,g,s} = ρ_{I,g,s} I_{t-1,g,s} + (1−ρ_{I,g,s}) \bar{I}_{g,s} + ε_{t}^{I,g,s}, with ε_{t}^{I,g,s} ~ N(0,σ_{I,g,s}^2). ### Fiscal Rule: Counter-cyclical, Debt-stabilizing Primary Deficit - Primary deficit-to-GDP rule: - pd_t = (1−ρ_pd) pd* + ρ_pd pd_{t-1} − ρ_y \tilde{y}_t − ρ_b (b_t − \bar{b}) + ε_t^{pd} - Where pd_t and b_t are primary deficit-to-GDP and total debt-to-GDP ratios; pd* and \bar{b} are government-set targets and long-term anchors. - ε_t^{pd} ~ N(0,σ_{pd}^2). - Parameters: ρ_y ≥ 0 is weight on countercyclical policy; ρ_b ≥ 0 is weight on speed of fiscal consolidation. - Financing shares: - Currency structure of debt is exogenous. - Given exogenous concessional borrowing, a constant share of the primary deficit and debt servicing is financed by domestic debt issuance. - Financing rule parameter φ_f determines the share financed domestically; foreign market debt is the residual. ### Policy Implementation and Main Fiscal Instrument - In the AFSM calibration for Armenia, consumption spending C_{t,j} is used as the main instrument to align fiscal strategy with the budget rule (i.e., C_{t,j} is endogenously determined by the fiscal rule, budget constraint, and other fiscal processes). - The AFSM is flexible: taxes and public investment can alternatively act as the main fiscal instrument. ### Key Calibration and Steady-State Parameters (quarterly unless stated) - General assumptions: - Exogenous annual growth rate: 4 percent (stated assumption). - Select steady-state structural parameters and values (preserved exactly): - Discount factor (β): 0.9951 - Probability of survival (ζ): 0.9932 - Relative risk aversion parameter (σ): 1 - Share of consumption in households’ utility (γ): 0.7 - Depreciation of physical capital in tradable sector (δ^T): 0.01 - Depreciation of physical capital in non-tradable sector (δ^N): 0.01 - Steady state growth rate, quarterly (g): 0.0125 - Steady-state domestic nominal risk-free interest rate (i): 9.4 - Steady-state foreign nominal risk-free interest rate (i^*): 9.4 - Share of labor in tradable sector (α_l^T): 0.55 - Share of physical capital in tradable sector (α_k^T): 0.33 - Share of labor in non-tradable sector (α_l^N): 0.45 - Share of physical capital in non-tradable sector (α_k^N): 0.35 - Share of public capital in non-tradable and tradable sectors (ψ): 0.1 - Price elasticity of demand for final consumption goods (θ_c): 6 - Price elasticity of labor demand (θ_n): 6 - Intra-temporal elasticity of substitution of consumption goods (θ_s): 0.75 - Intra-temporal elasticity of substitution of investment goods (θ_i): 0.75 - Price elasticity of foreign demand for Armenian exports (θ_x): 1.5 - Share of domestic goods in consumption (ω): 0.35 - Share of domestic traded goods in consumption (κ): 0.3 - Share of non-traded investment goods in non-traded sector investment (μ_N): 0.7 - Share of non-traded investment goods in traded sector investment (μ_T): 0.7 - Share of liquidity constrained households (f): 0.4 - Risk premium elasticity with respect to foreign liabilities (ς): 0.05 - Private net foreign liabilities (b^p): 4*0.45 - Annual target for foreign average headline inflation (π^*_): 2 - Remittances (share of GDP) (Rem^*): 0.119 - Technology level in non-tradable sector (A̅_N): 1.5 - Technology level in tradable sector (A̅_T): 1 - Transitory parameters (selected exact values): - Habit persistence in liquidity constrained households’ consumption (χ): 0.65 - Wage adjustment costs (ζ_w): 180 - Price adjustment costs for both traded and non-traded sectors (ζ_p^N, ζ_p^T): 96 - Price inflation indexation in traded and non-traded sectors (π_index): 0 - Private investment adjustment costs (ζ_I): 1 - (Non-)tradable-sector-specific labour adjustment costs (ζ_l^N, ζ_l^T): 1 - (Non-)tradable-sector-specific imported inputs adjustment costs (ζ_m^N, ζ_m^T): 1 - Persistence of foreign demand for Armenian exports (ρ_Y*): 0.8 - Persistence of remittances inflows (ρ_Rem): 0.5 - Persistence of terms-of-trade (ρ_tot): 0.5 - Persistence of foreign nominal interest rate (ρ_i*): 0.7 - Persistence of foreign headline inflation (ρ_π*): 0.4 - Persistence of foreign inflation (ρ_πf): 0.4 - Persistence of labor-augmenting technology in (non-)traded sector (ρ_A^N, ρ_A^T): 0.95 - Persistence of real export prices (ρ_p^x): 0.5 - Fiscal policy exact values (selected): - Depreciation of public capital (δ_g): 0.02 - Government investment efficiency (ε_g): 0.6 - Domestic public debt target (share of GDP) (b̄_dom): 0.103 - Foreign public debt target (share of GDP) (b̄_for): 0.078 - Foreign public concessional debt target (share of GDP) (b̄_fcon): 0.321 - Annual nominal interest rate on concessional loans (i_g): 1.2 - Weight on business cycle stabilization in fiscal rule (ρ_y): 0.06 - Weight on debt stabilization in fiscal rule (ρ_b): 0.06 - Share of fiscal deficits financed domestically (φ_f): 0.1789 - Long-run level of consumption effective tax (τ̄_c): 0.13 - Long-run level of labor income effective tax (τ̄_w): 0.15 - Long-run level of effective profit tax rate in tradable sector (τ̄_k^T): 0.12 - Long-run level of effective profit tax rate in non-tradable sector (τ̄_k^N): 0.19 ### External Financing and Risk Premia - Foreign lenders charge a premium over the risk-free rate that increases with the total net external debt-to-GDP ratio (public + private foreign debt): - 1 + i_t^m = (1 + i_t^*) prem_t - prem_t = ξ [ (B_{t-1}^{gf} − B_{t-1}^{gcf} + B_{t-1}^{p,f}) / GDP_{t-1} − NFD^* ] - ξ measures sensitivity of premium to debt (degree of capital account openness); premium can be shocked. - NFD^* is a target for total net external debt-to-GDP ratio consistent with public debt targets. ### Equilibrium and Resource Constraints - Equilibrium defined by prices and quantities satisfying agents’ optimality conditions, monetary and fiscal policy rules, government budget constraint, and market-clearing conditions. - Domestic resource constraints: - Non-traded goods: Y_{N,t} = C_{N,t} + C_{t,j} + I_{t,j,N} + I_{N,t,N} + A C P_{N,t} + A C N_{t} + A C M_{t} + A C W_t (notations preserved). - Traded goods: Y_{T,t} = C_{T,t} + I_{t,j,T} + I_{T,t,T} + X_t + A C P_{T,t} + A C N_{T,t} + A C M_{T,t} + A C W_t. - Nominal GDP definition preserved exactly: - NGDP_t = P_t^N C_t + P_{N,t} C_{t,j} + P_{T,t}^I I_{t,j,T} + P_{N,t}^I I_{t,j,N} + P_{T,t}^I \tilde{I}_{t,T} + P_{N,t}^I \tilde{I}_{t,N} + P_{T,t} X_t − S_t P_t M_t. - External resource constraint expressed as law of motion for net foreign assets with debt stock, interest terms, trade flows, and Remittances terms (notation preserved). *Source: wpiea2022118-print-pdf — Section 3.3 Government Sector (IMF).* ### 0.02 Long-run level of (non)tradable-sector-specific public capital ### 0.02 Long-run level of (non)tradable-sector-specific public capital investment ### Calibration and Key Parameter Values - Long-run levels: - Long-run level of (non)tradable-sector-specific public capital investment: 0.02 - Long-run level of net transfers to liquidity constrained households (푇ത௟): 0.13 - Long-run level of net transfers to liquidity unconstrained households (푇ത௢): -0.06 - Fiscal rule and tax persistence: - Fiscal deficit persistence in fiscal policy rule (휌௣௕, 휌ఛ௞ௗ): 0.84 - Persistence of (non-)traded-sector-specific profit tax (휌ఛ௡): 0.8 - Persistence of labor income tax (휌ఛ௖): 0.9 - Persistence of consumption tax (휌ூ௚௛, 휌ూ௚ௗ): 0.8 - Persistence of public capital investment in (non)tradable sectors (휌்௢, 휌்௟): 0.8 - Persistence of net-transfers to liquidity (un)constrained households (휌௜௖): 0.5 - Persistence of concessional interest rates: 0 - Monetary policy: - Domestic headline annual inflation target (휋ത): 4 - Persistence of monetary policy rate (휌௜): 0.75 - Weight on inflation stabilization in monetary policy rule (휙గ): 1.6 - Weight on business cycle stabilization in monetary policy rule (휙௬): 0.8 - Persistence of monetary policy shocks (휌௠௣௖): 0.6 ### Transitory Parameters and Adjustment Costs - Habit persistence (휒): 0.65 (following Stepanyan et al. (2009)). - Tradable and non-tradable price inflation indexation (휍ௗ, 휍௛): No price inflation indexation assumed in the baseline. - Rotemberg-style quadratic adjustment costs: - Price stickiness parameters: 휁௣,ௗ = 휁௣,௛ = 96 (implies 18 months nominal price stickiness). - Nominal wage stickiness: 휁௪ = 180 (judgmental, higher than source estimate near 80). - Remaining adjustment cost parameters for private investment, labor, and imported inputs (휁ூ, 휁ே,ௗ, 휁ே,௛, 휁ெ,ௗ, 휁ெ,௛) set to 1. ### Fiscal Policy Parameters and Targets - Public investment efficiency (휖௚): 0.6 (normalized from Dabla-Norris et al. (2011) public investment management index for Armenia at 2.39). - Depreciation rate of public capital (훿௚): 8 percent annually. - Implied public capital-to-output share: 64 percent. - Public debt targets and financing (steady-state shares and targets): - Total government debt target: 50.2 percent of GDP. - Domestic debt target: 10.3 percent of GDP. - Foreign debt target: 7.8 percent of GDP. - Concessional debt steady state: 32.1 percent of GDP. - Share of domestic debt financing (휙ௗ): 0.179 (calibrated for balanced-growth path given permanent annual growth rate of 4 percent). - Baseline fiscal rule parameters: - 휌௕ = 0.06 (implies about 2.8-year half-life of a percentage point deviation from the debt target). - 휌௬ = 0.06 (counter-cyclical; 1 percent output gap decreases annual primary balance by 0.25 percentage points of GDP). - 휌௣ௗ = 0.84 (implies relatively short half-life of primary deficits of about 1 year). - Long-run effective tax rates (휏̅஼, 휏̅ே, 휏̅௞ௗ, 휏̅௞௛): Calibrated to match value-added, labor income tax, and profit tax ratios to GDP. - Public capital investment and net transfers: - Total public investment: 3.8 percent of GDP (split equally between non-traded and traded sector-specific investment). - Total net transfers: about 10 percent of GDP. - Transfers per household group (combined with assumption that non-savers form one third of labor force): - 푇ത௟ / GDP = 0.13 - 푇ത௢ / GDP = -0.06 ### Monetary Policy Parameters (additional) - Headline inflation target (휋ത): 4 percent (target of the CBA). - Monetary policy rule estimates (Barseghyan (2013)): - Weight on expected inflation (휙గ): 1.6 - Weight on output gap (휙௬): 0.8 - Monetary policy rate smoothing (휌௜): 0.7 ### Great Ratios and Selected Fiscal Statistics (percent of GDP) — model vs data - Expenditures on GDP (Model steady state in last column): - Private consumption expenditures: 80.3 - ... domestic goods: 57.0 - ... imported goods: 23.5 - Gross private fixed investment: 16.1 - Government expenditures: 16.0 - ... government consumption: 12.9 - ... government investment: 3.1 - Exports of goods and services: 35.3 - Imports of goods and services: 48.1 - ... final consumption goods*: 18.1 - ... intermediate goods**: 30.0 - Fiscal policy ratios (Model steady state in last column where available): - Total revenues (incl. grants and non-tax revenues): - (not available in model column) - ... grants: - - Total tax revenue (excl. other taxes): 19.7 - ... income tax: 6.2 - ... profit tax: 3.2 - ... consumption tax: 10.3 - ... other: - - Total expenditures: - (not available in model column) - Total expenditures excl. transfers: 18.1 - ... current (incl. debt-servicing costs): 14.9 - ... capital: 3.2 - ... net transfers: 5.1 - Total central government debt: 50.2 - ... domestic: 10.3 - ... external: 39.9 - ... market: 7.8 - ... concessional: 32.1 - Overall deficit: 3.5 - ... primary deficit: 1.4 - ... interest costs: 2.1 Note: table entries compare data averages for 2000-2019, 2010-2019, 2015-2019 and the model steady state; the values above capture the model steady state column where provided. ### Shock Variances and Calibration Summary - Standard deviations of structural shocks: all shock variances set at a default unitary value (unit variance) because the model is not used to match actual data volatilities; scaling of volatilities applied by judgment in specific simulations. - Calibration objective: match basic dimensions of the Armenian economy and government fiscal policy; Table 2 compares model key characteristics with actual GDP uses and fiscal accounts. ### Validation: Fiscal Multipliers (AFSM vs Empirical SVAR estimates) - Definition: Multipliers defined as cumulative change in output over cumulative change in relevant fiscal variable. Headline numbers are AFSM predictions; bracketed numbers are empirical estimates (Fukač et al. (2021)). - Tax revenues — AFSM headline numbers and empirical ranges: - Consumption tax: - Impact multiplier: -0.7 [empirical: -1.8, -0.4] - 1-Year Multiplier: -0.5 [empirical: -2.3, -0.2] - 3-Year Multiplier: -0.4 [empirical: -1.7, 0.2] - Personal income tax: - Impact multiplier: -0.4 [empirical: -4.4, 0.3] - 1-Year Multiplier: -0.7 [empirical: -2.7, 0.8] - 3-Year Multiplier: -1.3 [empirical: -1.7, 0.6] - Corporate profit tax: - Impact multiplier: -0.2 [empirical: -4.4, -0.9] - 1-Year Multiplier: -0.2 [empirical: -2.3, 0.8] - 3-Year Multiplier: -0.4 [empirical: -1.7, 0.3] - Expenditures — AFSM headline numbers and empirical ranges: - Current expenditures: - Impact multiplier: 0.8 [empirical: -0.2, 1.4] - 1-Year Multiplier: 0.6 [empirical: -0.3, 1.0] - 3-Year Multiplier: 0.5 [empirical: -0.3, 0.4] - Capital expenditures: - Impact multiplier: 1.5 [empirical: 0.1, 1.1] - 1-Year Multiplier: 0.9 [empirical: -0.7, 1.1] - 3-Year Multiplier: 0.9 [empirical: -0.4, 0.7] - Conclusion from validation: - AFSM-implied multipliers fall within measured empirical ranges. - Tax policy multipliers closer to moderate values; expenditure policy multipliers closer to values associated with potent expenditure policies. - AFSM calibrated to accommodate increased potency of expenditure policies, reflecting prospective improvements in fiscal accountability, transparency, and public investment governance. ### Capturing Armenia’s Fiscal Rules: Satellite Model and Fiscal Speed Limit Monitor - Rules implemented via three logistic switches (푟௡,௧ for n = 1,2,3): - 푟௡,௧ = 1 / (1 + exp(—A (푏௧ — 푏ത௡))) with debt-to-GDP thresholds 푏ത௡ in {40, 50, 60}. - Parameter A calibrated for rapid switching; 푟 ranges between 0 and 1 and activates when actual debt breaches thresholds. - Effects on expenditures when switches activate: - Rule 40 (capital expenditures constraint): - Modified nominal capital expenditures: 푃௧ 퐼̄௧௚ = (1 — 푟ଵ,௧) 푃௧ 퐼௧௚ + 푟ଵ,௧ 퐷௧. - If 푏௧ ≤ 푏തଵ then 푟ଵ,௧ = 0 and capital expenditures follow baseline AFSM. - If 푏௧ > 푏തଵ then 0 < 푟ଵ,௧ ≤ 1 and capital expenditures constrained to overall deficit 퐷௧ (potential increase in capital spending comes at expense of current expenditures). - Rule 50 (current expenditures growth cap): - Modified nominal current expenditures: 푃ௗ,௧ 퐶̄௧௚ = (1 — 푟ଶ,௧) 푃ௗ,௧ 퐶௧௚ + 푟ଶ,௧ (1 — ∆ଶ଴푁퐺퐷푃௧) 푃ௗ,௧ 퐶̄௧ିଵ௚. - If 푏௧ ≤ 푏തଶ then 푟ଶ,௧ = 0 and current expenditures follow baseline AFSM. - If 푏௧ > 푏തଶ then 0 < 푟ଶ,௧ ≤ 1 and nominal expenditures increase at average nominal GDP growth over past 20 quarters (∆ଶ଴푁퐺퐷푃௧). - Rule 60 (current expenditures capped by tax revenue net of deductibles): - Modified nominal current expenditures when debt > 60 percent of GDP equate nominal expenditures to total tax revenues less corporate tax deductibles (expression provided in source). - If 푏௧ ≤ 푏തଷ then 푟ଷ,௧ = 0 and current expenditures follow baseline AFSM. - If 푏௧ > 푏தଷ then 0 < 푟ଷ,௧ ≤ 1 and nominal expenditures equal R௧ (total tax revenues less corporate tax deductibles) per the formula given. - Implementation notes: - The de jure rules are captured in a satellite model due to occasional binding, exclusions, and instrument activations (non-linearities). - The AFSM monitors adherence to de jure rules via a fiscal speed limit monitor (traffic-light indicator for Rules 40, 50, 60). - Default AFSM adjusts current spending as primary fiscal variable; in satellite model fiscal intervention is endogenous. ### Illustrative Experiment: Rule Activation and Monitor Responses - Experiment setup: - Series of shocks to primary deficits during 2020: Q1 — 2021: Q4 with 휀௧௣ௗ = 1, designed to raise public debt to breach 50 percent of GDP so Rule 50 binds. - Key dynamics and outcomes: - Current expenditures increase by 30 percent as a result of shocks; primarily debt financed, causing public debt to increase. - As debt exceeds 50 percent threshold, Rule 50 switch (푟ଶ,௧) transitions from 0 to 1; Rule 50 prescribes faster reduction of expenditures than baseline AFSM, leading debt to level off in 2022 and begin returning to target. - Rule 50 rolls back as debt falls below 50 percent threshold and switch returns to 0. - Fiscal speed limit monitor behavior (Figure 7 description summarized): - Rule 50 monitor (middle right panel) calls for intervention when current expenditures grow faster than nominal GDP; if debt still below 50 percent, expenditures follow baseline rule until debt breaches threshold. - Later in 2024 monitor flags current expenditures growing faster than nominal GDP again, but no intervention required if debt on descending path. - Rule 40 not activated in this illustration; monitor flags its violation in red when capital spending is less than overall deficit. Once Rule 50 binds, it helps align capital expenditures with Rule 40 prescriptions. - Rule 60 remains satisfied throughout the experiment; no intervention needed. - Insight: - Overlaps among Rules 40, 50, and 60 are non-trivial; satellite extension to AFSM is useful for studying complexities and interactions of fiscal rules and interventions. ### Model Applications and Use Cases - AFSM is validated against empirical fiscal multipliers and designed to: - Quantify macroeconomic effects of fundamental shocks and policies. - Serve as a flexible policy scenario analysis tool to help formulate forward-looking fiscal strategies. - Accommodate design and assessment of fiscal interventions when fiscal rule monitors flag breaches. *Source: Authors’ expert judgement and empirical estimates (from the AFSM chapter content).* ### 6.1 An Assessment of the First Wave of the Covid-19 Pandemic ### 6.1 An Assessment of the First Wave of the Covid-19 Pandemic ### Overview and context - The Armenian economy was hit hard by the Covid-19 pandemic. In March 2020, tackling the health risks became the Government’s top priority. - The MOF estimated that the GDP could contract by 3.7 percent in 2020 due only to lockdowns and other public-health-related measures. - After a long month of lockdown, in April 2020 a substantial disruption of both the demand and supply sides of the economy materialized. - The service sector, which is about one half of the economy, suffered the most due to constraints on international travel, social events, and physical distancing. Construction was the second most affected sector. This translated into major job losses. - The Macroeconomic Policy Department of the MOF was asked to use the AFSM to assess possible scenarios associated with the pandemic. The AFSM-based analysis was expected to provide a quantification of the macroeconomic effects of the pandemic as well as the mitigating effects of policy responses, including the potential policy trade-offs. - The assessment described here took place during April–October of 2020 and is presented as it was done in that period before knowing the actual macroeconomic and fiscal outcomes for 2020. ### Shock classification and calibration approach - Shocks are classified as domestic and foreign: - Domestic shocks originate from the health situation in the country, lockdowns, and economic uncertainties that affect consumption and investment. - Foreign shocks are mainly related to the collapse in foreign demand, a decline in remittances, and an increase in country-risk premia. - Calibration sources: - For foreign shocks: IMF forecasts of October 2020. - For domestic shocks: information and forecasts about the domestic economy from the MOF. - Calibration and key judgement calls used in the AFSM: - The pandemic is expected to worsen in the second quarter of 2020, and the recovery starts in the third quarter. - A sharp decline in trading partners’ output immediately reduces the demand for Armenian exports (tourism, transportation, and manufacturing). In 2020, this demand is expected to contract on average by 7.4 percent, with the largest decline in the second quarter and the recovery starting in the third quarter. - In 2021, the average growth of export demand is assumed at 2.2 percent, with countries closing the output gap only at the end of the year. - Remittances are expected to decline by 9.5 percent in 2020—mainly because of the contraction in Russia—and slowly recover by 4.1 percent in 2021. The shock is used to calibrate their annual growth rates. - The country-risk premium shock is calibrated to match increases in the interest rate spread for Eurobonds (relative to U.S. government bonds) of 0.8 and 0.4 percentage points in 2020 and 2021, respectively, and dissipate afterwards. - Consumption is expected to be negatively impacted by the pandemic; a shock to the marginal propensity to consume is imposed to capture effects of lockdowns, increased economic uncertainty, and precautionary savings motives. Absent other shocks, consumption is expected to decrease by 7.8 percent in 2020, and increase by 2.2 percent in 2021. - To reflect the pandemic negative effect on investment, sector-specific shocks to the real returns on capital are used to capture protracted uncertainty that reduces firms’ willingness to invest. The shocks profile is calibrated to obtain, absent other shocks, an average contraction of total investment of 21.7 percent in 2020, and an increase of 2.2 percent in 2021. After that, investment growth is assumed to return gradually to its trend level by 2023. - The foreign demand is calculated using export-weighted average growth rates of the Russian Federation, Euro Area, China, and the U.S. - The remittances shock is calibrated using econometric estimates of elasticities and the IMF’s WEO forecast for the GDP growth rates of Russia, while accounting for the travel restrictions of 2020. ### AFSM simulation results and key macroeconomic impacts - Using the calibrated shocks, the AFSM is simulated to quantify macroeconomic effects. The simulations are summarized in Figures 8–12 and Table 4 in the source. - Key quantified impacts (absent other interventions unless specified): - GDP decreases by 12.9 percent in 2020, and slowly recovers over the horizon period. - Export demand contracts on average by 7.4 percent in 2020; assumed export-demand growth is 2.2 percent in 2021. - Remittances decline by 9.5 percent in 2020 and recover by 4.1 percent in 2021. - Country-risk premium increases are calibrated at 0.8 and 0.4 percentage points in 2020 and 2021, respectively. - Consumption is expected to decrease by 7.8 percent in 2020 and increase by 2.2 percent in 2021 (absent other shocks). - Total investment is calibrated to contract by 21.7 percent in 2020 and increase by 2.2 percent in 2021, returning gradually to trend by 2023. - The results underscore the severe impact that the pandemic can have on the Armenian economy. *Source: wpiea2022118-print-pdf - 6.1 An Assessment of the First Wave of the Covid-19 Pandemic* ### 11.7 percent, while investment and exports are hit the hardest, falling by more than 20 percent ### 11.7 percent, while investment and exports are hit the hardest, falling by more than 20 percent ### Macroeconomic impact of the pandemic (AFSM simulations) - Estimated direct and cumulative macroeconomic GDP losses can get as high as 38 percent of GDP over the period 2020-2023, relative to the pre-crisis level. - Inflation will be 2.5 percentage points below its target in 2020 and returns to it in the medium-term. - The CBA reduces the policy interest rate by 2.1 percentage points in response to lower inflation. - The policy response and external shocks induce a nominal depreciation of about 4 percent in 2020; in the medium-term the exchange rate appreciates. - In the third quarter of 2020, GDP is expected to fall by almost 15 percent below its long-term potential level. - By the end of 2023, public debt-to-GDP ratio can increase by about 19 percentage points and stay at a higher level in the medium term absent fiscal consolidation. - Two thirds of the macro-fiscal effects are due to the external shocks alone; the foreign demand shock explains more than 75 percent of the almost 15 percent GDP fall in 2020:3 and more than 75 percent of the almost 20 percentage points increase in the public debt-to-GDP ratio by end-2023. ### Shock decomposition and transmission mechanisms - Shock landscape is dominated by: - foreign demand shock - country-risk premium shock - remittances shock - Pandemic-related shocks affect both demand and supply, leading to contractions in: - output - private consumption - private investment - exports and imports - employment - Shocks also cause declines in: - wage- and price-inflation - nominal interest rates - and induce nominal (and real) exchange rate depreciation - Transmission channels highlighted by the AFSM: - Declines in foreign demand, consumption (via marginal propensity to consume), and investment (via real returns on capital) directly reduce exports, consumption, and investment. - Decline in remittances reduces liquidity-constrained households’ consumption. - Increase in country-risk premium can lead to export expansion via exchange rate depreciation. - Lower demand reduces firms’ output and prices, lowering inflation below target and reducing labor, capital, and imported input demands. - Falling consumption and taxes shrink the tax base, widening primary deficits and increasing public debt absent policy measures. - Monetary accommodation (lower interest rates) with an open capital account leads to capital outflows and exchange rate depreciation, which raises import prices and can exacerbate foreign-currency public debt burden. ### Fiscal deterioration and public debt dynamics - Tax to GDP ratio is projected to decrease by 3.7 and 2.7 percent below the long run levels, in 2020 and 2021, respectively, and remain at 1 percent below its steady-state value over the medium term. - Primary deficit widens by 6.5 and 4.7 percentage points of GDP, in 2020 and 2021, respectively. - To finance increased deficits the government requires additional borrowing, pushing public debt-to-GDP ratios up. - Nominal exchange rate depreciation exacerbates the public debt burden because an important part of debt is in foreign currency. - The primary deficit fiscal rule helps stabilize debt in the medium term, avoiding debt explosive dynamics, even if debt reaches a much higher level. ### Fiscal policy responses (Armenia case, AFSM representation and calibration) - Government announced a socio-economic support package of AMD 150 billion (2.5 percent of GDP) to be implemented in Q2 2020, plus a corporate income tax moratorium. - AFSM matched policy actions to model instruments; simulations done April–October 2020 (ex-ante). - Main programs calibrated and modeled: - Economic support program: AMD 33.5 billion (0.5 percent of GDP). - Modeled as government current spending shock, calibrated to 0.26 percent of quarterly GDP in Q2 2020, 0.19 percent in Q3 2020, and 0.08 percent in Q4 2020. - Pre-allocated amount AMD 23.5 billion, AMD 10 billion reserves. - Social support and wage subsidy program: AMD 36.5 billion (0.6 percent of GDP). - Modeled as transfers to liquidity-constrained households, calibrated to 0.5 percent of GDP in Q2 2020, 0.07 percent in Q3 2020, and 0.03 percent in Q4 2020. - Limited corporate income tax payments deferral and corporate tax reform: AMD 65 billion (1.1 percent of GDP). - Modeled through corporate profit tax shocks in traded and non-traded sectors with values -0.71 percent of GDP in Q2 2020, -0.23 percent in Q3 2020, and -0.12 percent in Q4 2020. - Private investment support program: AMD 80 billion (1.3 percent of GDP). - Modeled via shock to the real return on capital in the traded sector: 40 percent of program amount in Q3 2020 (0.51 percent of GDP) and the rest in Q4 2020 (0.77 percent of GDP). ### Macroeconomic and fiscal effects of the fiscal package (AFSM results) - Total budgeted costs AMD 215 billion cushion about AMD 303 billion (4.6 percent of GDP) in contracting economy. - In 2020 the model estimates the program can help cushion: - GDP by 1.9 percentage points. - Employment by 3.7 percentage points (equivalent to protecting 35,460 jobs). - The AFSM projects the primary balance can deteriorate by 2.6 percentage points of GDP, relative to the baseline scenario, from -9.5 to -12.1 percent because of increased public spending and declining tax revenues. - Public debt as share of GDP is projected to increase by 3 percentage points relative to the baseline and then stabilize in the medium term. - AFSM indicates that a larger fiscal package could push public debt to unsustainable levels given the weak outlook and increase debt service burden. - Government activated the escape clause of the fiscal rule; public debt surpassed the threshold of 60 percent of GDP. Escape clause was also in place in 2021. In 2022, Government committed to adhere to fiscal rules and implement a 5-year fiscal consolidation program. ### Quantitative net impacts on GDP and components (Table 4: percentage deviations from long-term trend) - 2020: - GDP: Baseline -12.9; Fiscal Policy -11.0 - Consumption: Baseline -11.7; Fiscal Policy -9.5 - Investment: Baseline -21.7; Fiscal Policy -19.4 - Exports: Baseline -23.4; Fiscal Policy -21.8 - Imports: Baseline -18.4; Fiscal Policy -15.3 - 2021: - GDP: Baseline -10.4; Fiscal Policy -9.7 - Consumption: Baseline -9.5; Fiscal Policy -8.8 - Investment: Baseline -19.9; Fiscal Policy -17.9 - Exports: Baseline -23.6; Fiscal Policy -22.2 - Imports: Baseline -19.1; Fiscal Policy -17.3 - 2022: - GDP: Baseline -6.7; Fiscal Policy -6.0 - Consumption: Baseline -7.2; Fiscal Policy -6.7 - Investment: Baseline -14.5; Fiscal Policy -13.2 - Exports: Baseline -16.2; Fiscal Policy -14.8 - Imports: Baseline -15.2; Fiscal Policy -14.1 - 2023: - GDP: Baseline -4.4; Fiscal Policy -3.7 - Consumption: Baseline -5.5; Fiscal Policy -4.8 - Investment: Baseline -9.8; Fiscal Policy -8.6 - Exports: Baseline -11.0; Fiscal Policy -9.6 - Imports: Baseline -11.3; Fiscal Policy -10.2 ### Program-specific mechanisms and expected outcomes - Economic support program: - Increases current government spending with both domestic and imported components. - Positively affects production and import demand; mitigates employment decline. - Social support program: - Transfers target vulnerable, liquidity-constrained households to stabilize disposable income and stimulate consumption. - Expected to mitigate negative impacts on production via higher consumption. - Corporate tax reform: - Lowers effective corporate income tax to incentivize investment and employment. - Stimulates demand in short term and production via capital accumulation in medium term. - Private investment program: - Promotes productive capital formation, bolstering medium-term production. - Targets traded sector to offset export contraction and mitigate trade-balance pressure from other support measures. - General equilibrium effects: - Increased demand prompts higher production and labor demand relative to no-policy case, partially restoring employment. - Smaller decline in marginal production costs induces a smaller CPI inflation drop; CBA reduces policy rate timidly, leading to a smaller exchange rate depreciation. *Source: Authors’ estimates and calculations from AFSM simulations contained in the cited chapter.* ### 6.2 Comparing Alternative Policies in the Context of Public Investment ### 6.2 Comparing Alternative Policies in the Context of Public Investment ### Under-Execution - After the 2018 Velvet revolution, a full-scale review revealed inefficiencies in public infrastructure programs across all stages from project selection to implementation. - In both 2018 and 2019, capital expenditures were under-executed by 30 percent. - The government took drastic actions (change managers and contractors, prosecute in some cases), raising uncertainty about medium-term implementation speed and what to do with the associated “savings.” - In fall 2019 the AFSM was used to investigate policy options assuming public investment under-execution would continue in the medium term. ### Baseline Scenario (replicating the 2020-2022 MTEF) - Simulation horizon: 2020-2023. - Policy measures included: - Permanent increase of public investment by 1.9 percentage points of GDP in the first year, and then by 0.3 percentage points annually for the following years. - Permanent decrease of current expenditures by 0.2 percentage points of GDP for the entire period. - Increase of tax revenues (by raising tax rates) equivalent to 0.3 percent of GDP, starting from 2021. - Financing assumption: - Expenditure increase financed through debt accumulation in the first year, and through additional tax revenues for the rest of the simulation period. - Debt-to-GDP ratio assumed to increase by 0.2 percentage points in 2020, and stay constant at its initial value for the rest of the period. - Any fiscal discrepancies balanced by endogenously adjusting lump-sum taxes on savers. - Calibrations: - Public investment to GDP shares set to increase by 1.9 percentage points in 2020, 2.2 percentage points in 2021, 2.5 percentage points in 2022, and 2.8 percentage points in 2023. - Current expenditures calibrated as an annual decline of 0.2 percentage points of GDP. - Tax rate shocks judged to generate an increase in total tax revenues equivalent to 0.3 percent of GDP, starting from 2021. - Key simulated outcomes: - MTEF policies induce an expansion of GDP and crowd in private investment and consumption. - Permanent increase in public capital expenditures raises productivity of labor and private capital, stimulating investment and employment. - Consumption increases due to liquidity-constrained consumers raising expenditures from higher disposable income. - Higher aggregate demand pushes wages and consumer prices up; the central bank responds by increasing interest rates. - Higher interest rates attract capital inflows, inducing nominal exchange rate appreciation and real appreciation. - Although debt-financed, primary deficits are modest and public debt is estimated to accumulate by about 1 percentage point of GDP. ### Alternative Policy Scenario 1: Debt Repayment - Design: - Models capital expenditure under-execution by lowering public investment levels to match the under-executed amount. - Savings from under-execution are reallocated to foreign public debt repayments (negative shock to debt-to-GDP). - Key outcomes: - Productive capacity is lower than in the baseline, leading to lower output, weaker demand, and less employment. - Smaller crowding-in effects on consumption and investment relative to baseline. - Fiscal benefit: public debt is put on a deleveraging path and fiscal space is created in the medium-term. - Note: debt repayments may lower the country risk premium and increase fiscal credibility, but these effects may not fully offset negative effects from the decline in productive capacity. ### Alternative Policy Scenario 2: Increasing Social Benefits - Design: - Reallocates saved budget funds to an increase of pensions and other social benefits. - Implemented via a calibrated positive shock to transfers to liquidity-constrained households, matching the saved funds from under-execution. - Key outcomes: - Social transfers boost private consumption (direct beneficiaries: liquidity-constrained households). - Relative to baseline, the under-execution plus higher social benefits stimulate consumption but reduce GDP and investment. - The alternative does not expand the economy’s productive capacity and can crowd out private investment. - Capital and employment increase by less, and output is lower in the medium term than in the baseline. - Fiscal effects: - Raising transfers negatively impacts the primary fiscal balance, leading to debt accumulation and higher public debt-to-GDP ratios. - Under this alternative the debt-to-GDP ratios breach the threshold of 50 percent. - Note: in this simulation current expenditures are not adjusted in response to debt accumulation. ### Alternative Policy Scenario 3: Increasing Public Investment Efficiency - Context: - Public investment efficiency in Armenia is relatively low: out of a Dram of investment, only about 60 percent turns into productive capital. - Design: - Compensate for under-execution by increasing the marginal efficiency of new public investment projects so that, with a lower level of public investment, the same increases in productive capacity are achieved. - Calibrated to fully offset the expected fall in public capital associated with under-execution. - Required efficiency increase: - Estimated required increase in efficiency is about 16.5 percentage points over the simulation period. - Key outcomes: - Delivers similar effects on GDP compared with the baseline and improves the fiscal outlook in the medium term. - Leads to lower public expenditures and puts public debt on a decreasing path over the medium term, relative to the baseline. - Ambition caveat: implementation and effects of structural reforms take time to materialize. ### Comparative Findings and Policy Implications - A cross-scenario comparison shows: - Increasing public investment efficiency dominates debt repayment and increasing social spending in terms of GDP outcomes. - Increasing public investment efficiency can also ensure a declining medium-term debt path. - Policy implications: - Structural reforms to improve public investment efficiency play a key role and are called for. - Speed and feasibility are central concerns: achieving a 16.5 percentage point efficiency gain over the simulation period is ambitious. - Trade-offs exist: debt repayment improves immediate fiscal metrics but at the cost of productive capacity; larger social transfers help vulnerable households but can worsen fiscal balances and reduce medium-term GDP. - Modeling caveat: - The AFSM comparison may underestimate positive productivity effects from some social spending; this assumption could make increasing public investment efficiency relatively more attractive in the AFSM framework. *Source: IMF authors’ calculations as described in the chapter "6.2 Comparing Alternative Policies in the Context of Public Investment."* ### References ### References ### Purpose - Compiles the bibliographic sources cited in the chapter. ### Key themes and topics represented in the cited literature - "A Medium-Scale DSGE Model for the Integrated Policy Framework" - "A Quantitative Model for the Integrated Policy Framework" - Economic Growth - Dynamic Stochastic General Equilibrium (DSGE) models - Public Investment Efficiency and Growth - Public Investment, Growth, and Debt Sustainability - A Conceptual Model for the Integrated Policy Framework - Nominal Rigidities and the Dynamic Effects of a Shock to Monetary Policy - Estimates of Fiscal Multipliers for Armenia - Monetary Policy and Exchange Rate Volatility in a Small Open Economy - Efficiency-Adjusted Public Capital and Growth - Fiscal Rules in Armenia - The Global Integrated Monetary and Fiscal Model (GIMF) – Theoretical Structure - Time to Build and Aggregate Fluctuations - Debt Sustainability, Public Investment, and Natural Resources - STAMP: A DSGE Model for Structural Analysis of Macroeconomic Policies - Government Spending Effects in Low-Income Countries - SIGMA: A New Open Economy Model for Policy Analysis - Interest and Prices: Foundations of a Theory of Monetary Policy *The Ararat Fiscal Strategy Model: A Structural Framework for Fiscal Policy Analysis in Armenia Working Paper No. WP/2022/118* --- _Source: https://www.imf.org/-/media/files/publications/wp/2022/english/wpiea2022118-print-pdf.pdf_