## FISCAL INTEGRATION FOR THE ECCU (content unit 1eccea2020002) ## Source details **Canonical URL:** [FISCAL INTEGRATION FOR THE ECCU (content unit 1eccea2020002)](https://www.imf.org/-/media/files/publications/cr/2020/english/1eccea2020002.pdf) ## Other formats - [Markdown version](/-/media/files/publications/cr/2020/english/1eccea2020002.pdf.md) - [Structured JSON version](/-/media/files/publications/cr/2020/english/1eccea2020002.pdf.json) --- ### Overview and context - Revised Treaty of Basseterre (signed June 2010) called for a single financial and economic space with free movement of goods, people and capital, and harmonization of monetary and fiscal policies. - ECCB Monetary Council instituted a public debt to GDP target of 60 percent by 2030. - Natural-disaster imputed fiscal cost ranges from ½ to 1½ percent of GDP (IMF 2018). ### Fiscal integration: rationale, constraints, and sequencing - Rationale for greater integration: - Pooling capacity to mitigate small-country disadvantages. - Correcting sub-optimal interactions and excessive competition (e.g., tax incentives, CBI programs). - Leveraging asymmetries for improved risk-sharing. - Constraints and implementation considerations: - Need resilient country-specific fiscal responsibility frameworks and strong fiscal institutions. - Ex-ante disaster mechanisms: national saving funds; contingent financing. - Improvements require fiscal space, capacity, and concessional financing; careful sequencing necessary. - Risks: cross-border spillovers from weaker members, political and democratic-accountability issues when centralizing fiscal mechanisms. ### Tax incentives: measurement, impacts, and magnitudes - Trend: expanded tax holidays, longer loss carry-forward years, higher investment tax allowances, lower statutory CIT rates, customs concessions — often not rules-based or coordinated. - Estimation approaches for foregone revenue: - Based on CIT returns: apply standard CIT rate to taxable profits of companies benefiting from tax holidays minus actual CIT collections. - Based on VAT returns: potential CIT base estimated as one-fourth of the difference between total sales and total purchases. - Based on sector-specific value-added from national accounts: potential CIT base = one-fourth of value added for beneficiary sectors. - Key quantified findings: - Total foregone revenue averaged 5.8 percent of GDP in 2010–18 in the ECCU region, or 21 percent of total tax revenue. - Cost of tax incentives per job in the formal sector is as high as $2,500 per job. - Sectoral and distributional findings: - Some positive impact on tourism-sector metrics (e.g., number of hotel rooms). - Pooled OLS: increase in foregone revenue by 1 million EC dollars associated with an increase in the number of hotel rooms by 0.2 percent. - No robust impact found on overall FDI, GDP growth, or employment (Appendix I Table 1-4). - FDI to the Caribbean positively correlated with the stock of public infrastructure during natural disaster shocks, implying a tradeoff between tax incentives and revenue for infrastructure. ### Empirical specification and selected coefficient excerpts (FDI and tourism regressions) - Dependent variables: FDI; number of hotel rooms (tourism proxy). - Estimators: Fixed Effect Panel (FE) and GMM (IV-GMM) with country fixed effects; FE emphasized due to small sample size. - Selected coefficient values as presented (preserve original precision and significance markers where shown): - CIT rate columns include values: 0.006 0.001 -0.001 0.006 0.028 0.016*. - Tax holiday coefficient examples: -0.042* -0.091***. - Investment allowance rate examples: -0.002 -0.006**. - Foregone revenue reported as -0.001* 0.000. - Public consumption examples: -0.055 -0.088** -0.063 -0.055 -0.087** -0.200***. - REER examples: 0.002 0.002 0.007 0.002 0.009 0.020***. - US PPPGDP examples: 0.591** 0.547** 0.725** 0.590** 0.884*** 0.976***. - Number of flights (lag): 0.028** 0.029** 0.045 0.027** 0.038 0.006. - Infrastructure stock (lag) interactions with natural disaster dummy: 0.054* 0.054* 0.086* 0.054* 0.043 0.073***. - Sample sizes and fit examples: - Observations: 140 140 102 140 96 81. - Number of countries: 10 across specifications. - R-squared examples: 0.565 0.576 0.504 0.565 0.445. - Appendix I selected tables (excerpted numerical highlights): - Appendix I. Table 1 (number of hotel rooms; FE and GMM): CIT rate (lag): -0.006, -0.006, -0.013**, -0.006, -0.009, 0.006; Tax holiday (lag): -0.004, -0.074***; Investment allowance rate (lag): -0.003*, -0.002; Foregone revenue (lag): 0.000, 0.001**; Infra. stock (lag 2): 0.113***, 0.116***, 0.110**, 0.110***, 0.073, -0.214*; Observations: 135, 135, 101, 135, 93, 80; Number of countries: 10; R-squared: 0.54, 0.31, 0.319, 0.468, 0.327. - Appendix I. Table 2 (OLS; number of hotel rooms): CIT rate (lag): -0.017***, 0.000, 0.027***, 0.001, 0.031**; Foregone revenue (lag): 0.002***; Public consumption: -0.064***, -0.124***, -0.173***, -0.124***, -0.134***; Number of flights (lag): 0.033***, 0.041***, 0.050***, 0.041***, 0.085***; Observations: 135, 135, 101, 135, 93; R-squared: 0.882, 0.894, 0.843, 0.899, 0.639. - Appendix I. Table 3 (Overall GDP; FE): CIT rate: 0.005***, 0.004***, 0.003**, 0.004**; Public consumption: -0.014***, -0.015***, -0.010***, -0.009**; Observations: 160, 157, 115, 96; Number of countries: 11, 11, 11, 6; R-squared: 0.935, 0.935, 0.921, 0.911. - Appendix I. Table 4 (Total Employment; FE): CIT rate: 0.004***, 0.005***, 0.003*, 0.005**; Public consumption: -0.007**, -0.006*, -0.004, -0.005; Observations: 150, 147, 105, 86; Number of countries: 11, 11, 11, 6; R-squared: 0.82, 0.83, 0.76, 0.79. ### Citizenship-by-Investment (CBI) programs: competition, coordination, and theory - Recent dynamics: - Several countries reduced prices and eased requirements to attract applications (examples: Dominica end-2016; St. Lucia; Antigua and Barbuda late 2017; St. Kitts and Nevis early 2018). - ECCU initiative in March 2019 to harmonize CBI due diligence processes. - Rationale for regional coordination: - Short-term: protect financial integrity and transparency; stabilize revenue flows; support financial stability. - Medium-term: coordinate on pricing and use of quotas to sustain revenues. - Theoretical framing: - Bertrand price competition where required CBI investment is the perceived price. - With identical countries, a minimum price or cap on new citizens can improve joint revenues. - With asymmetric demands (L-type low-demand and H-type high-demand), outcomes: - H-type countries charge higher prices and grant more new citizenships in Non-Cooperative Equilibrium. - Minimum prices bind more on L-type countries; maximum quotas restrict H-type countries more. - Combination of minimum price and maximum quota can generate a Pareto efficient outcome (illustrated with linear demand and substitutability parameter 훿>0). - Graphical/analytical conclusions: - Pareto efficiency achieved when H-type marginal revenue gain from raising minimum price equals L-type marginal revenue loss. - A combination of price and quantity instruments can ensure a “fair” distribution and an efficient outcome more profitable for both types relative to non-cooperation. - Empirical precedent: Party to the Nauru Agreement (PNA) Vessel Day Scheme used quantity (days) and price (minimum fee) instruments to increase revenues and sustainability. ### Recommendations on tax incentives and CBI - Reduce harmful tax competition and the “race to the bottom.” - Coordinate CBI programs on due diligence, pricing, and quotas to protect financial integrity and revenues. - Prioritize raising revenue to finance public infrastructure to attract FDI more sustainably than tax concessions. --- ### A regional stabilization fund (RSF): objectives, quantitative findings, and design - Objectives: - Quantify benefits and costs of an RSF; demonstrate scope for risk-sharing given imperfectly correlated macro-fiscal shocks. - Main quantitative findings: - An RSF would need to be about ½ as large as the sum of individual-country saving funds (SFs) while providing a similar degree of insurance. - SFs/RSF designed to incentivize less procyclical fiscal policies would increase average public investment over the cycle at the expense of current spending. - Reallocating spending toward investment yields substantial increases in output, employment, and wages once GE responses are accounted for. - After opportunity costs, benefits from cross-country risk-sharing likely outweigh opportunity costs. - Procyclicality and financing constraints: - Government spending in ECCU is procyclical: consumption rises in booms and crowds out public investment in recessions. - ECCU financing constraints: not integrated into global financial markets; external financing mostly official project-based loans with lags; domestic markets shallow; CCRIF under-insurance due to high reinsurance costs. ### Monte-Carlo VAR simulations and RSF sizing - Method: - VARs per country with five endogenous variables: cyclical output, tax revenue, non-tax revenue, recurrent and capital expenditure. - Two Monte-Carlo experiments of 1,000 simulations each (off-sample projections). - Saving-disbursement rule: - Governments save a share σ of revenue gap above trend; disburse share φ of predicted decline of capital expenditure below trend. - SF stock evolution: SFS_{i,t+1} = (1 + i_{i,t}) SFS_{i,t} + (S_i − F_i). - Size implications (numerical example preserved): - For probability of SF depletion at 1 percent and share of “excess consumption” saving of 1, estimated RSF ≈ 5 percent of regional GDP, while sum of individual countries’ SF ≈ near 10 percent. - RSF size about ½ of sum of individual SFs for same saving and withdrawal flows. - Distributional observation: - All countries benefit from pooling: estimated individual SFs are larger than allotted shares under regional pooling, implying implicit cross-country transfers. ### General equilibrium impacts from RSF and public investment reallocation - Investment increases from RSF: - With RSF depletion probability 0.01, public investment would increase in range 0.5–1.5 percent of GDP per year on average. - GE model features: - Small-open-economy islands; public capital enters productivity as externality Y_{i,t} = Z_{i,t-1}^{ψ} K_{i,t-1}^{α_i} L_{i,t}^{1−α_i}. - Output elasticities from permanent 1 percent of GDP reallocation (increase in capital, decrease in current spending): - Elasticities range between 0.9 and 2.1 when government reduces transfers or the wage bill (elasticities reflect permanent increases in output levels). - Largest output elasticity when cutting public sector employment. - Quantified level impacts using country-specific increases in investment (from RSF simulations): - Level increases in GDP range between 0.9 percent for St. Vincent to 2.4 percent for Antigua. - Level increases in consumption are even larger due to freed domestic resources from reduced purchases of government services. - Spending-adjustment calibration in GE simulations: - Reduction in current spending weights: purchases of goods and services 0.5; transfers 0.3; public employment 0.2. ### RSF simulation illustration: natural disaster in Antigua and Barbuda - Scenario assumptions: - Total damages = 10 percent of GDP with two-thirds borne by public sector. - Fund size calibrated to 5 percent of ECCU GDP. - Stochastic simulation: countries hit independently; average annual loss of 1 percent of GDP; annual probability of a shock hitting a country = 0.1; resulting damages = 10 percent of GDP. - Key simulated outcomes (RSF vs No RSF): - Output contraction reduced via quicker reconstruction and crowding-in of private investment. - Private investment in affected country higher post-disaster under RSF. - Fall of consumption in affected country much reduced on impact and long run. - Consumption in unaffected countries falls slightly as they recapitalize the fund. - Fund maintained ample resources and recapitalized within five years. ### Opportunity costs, financing, and long-run assessment - Opportunity-cost setup: - RSF capitalized by domestic borrowing; debt-servicing financed by increasing non-distortionary consumption taxes and distortionary capital taxes in equal measure. - GE model captures distortionary taxation effects on capital accumulation. - Long-run results: - Stabilization benefits likely outweigh opportunity costs; consumption higher over long run with RSF. - Opportunity costs could be lower if RSF capitalized more by non-distortionary taxes (e.g., carbon tax) or grants. - Model likely underestimates utility gains because it omits lower consumption volatility benefits. - Cost of funds: - RSF assets assumed invested at world risk-free rate; effective cost = difference between domestic and foreign interest rates. ### RSF governance, modalities, sequencing, and risks - Guiding principles: - Objective: improve macro stabilization through transfers to finance public expenditure including capital/reconstruction investment on negative shocks. - Scope: address temporary shocks uncorrelated over time; avoid permanent transfer scheme. - Governance and safeguards: - Financing via national fiscal accounts; no ECCB involvement to protect reserve position and quasi currency board arrangement. - RSF operations transparently recorded in fiscal accounts; clear saving and investment rules. - Country-level fiscal responsibility frameworks and peer review; compliance criteria for sanction and access to fund. - Triggers: objective, transparent, automatic; significant natural disasters and/or large GDP growth declines as alternatives to unobservable output gap. - Supranational agency management with authority to assess damage, enforce rules, and build asset-liability management capacity. - Decision-making mechanisms: consensus, qualified majority, or reverse qualified majority voting could be considered. - Modality: - Caps on transfers and contributions: access capped by accumulated savings for a set period (example: 10 years); contributions capped to avoid chronic net contributors. - Actuarial inflows as alternative (e.g., CCRIF probabilistic models). - Rainy-day fund as contribution-transfer arrangement with prudent initial limits; gradual expansion as institutions strengthen. - Voluntary lending beyond transfer caps at mutually agreeable rates and maturities for very large shocks. - Sequencing: - Stage 1: Reallocate funds to RSF with prudent transfer limits; upgrade national fiscal frameworks and transparency. - Stage 2: Gradually relax transfer limits as institutional credibility grows. - Stage 3: Consider topping up transfers through voluntary lending. - Realistic timeline: long term given institutional complexity. - Costs and challenges: - Political and institutional hurdles; slow consensus due to asymmetries. - Enforcement/legal risks: non-observance of commitments, inappropriate use of funds, disagreements on shock sizing, legal challenges. - Moral hazard risk substantial; requires enforcement and sanctions mechanisms. ### Conclusions on RSF - Short-term priorities: - Create fiscal space to reduce public debt and build resilient infrastructure and insurance buffers. - Substantial scope for raising revenues immediately. - Long-term role: - Regional pooling more cost-effective than exclusive national buffers; RSF requires only ½ resources of sum of individual funds for same stabilization. - RSF complements, not substitutes, country-level resilience buffers and requires strong supranational governance, transparent accounting, and careful sequencing. - Preconditions: - Strong national and regional institutions; transparent methodology for revenue and spending assessment; accounting mechanism for net-credit appropriation; rules on access. - Political will and time required; further analysis and technical assistance may be needed to refine institutional details. --- ### Appendix II: Calculation steps for country and regional stabilization funds (SF / RSF) - Step 1: Estimate unrestricted VAR per country (sample 1990–2018) with endogenous vector X_i of cyclical GDP, tax revenue, non-tax revenue, current primary expenditures, capital expenditures; residuals ε_i ~ N(0, Ω). - Step 2: Monte-Carlo experiment generating 1,000 forecasts X̂_τ per country using simulated shocks ε̂_τ with Choleski factorization Ω = W′W and v_τ ~ N(0,1); produces PDFs for 2019–2030. - Step 3: Transform deviations from trend into percent of GDP with deterministic trend projection starting at constant prices of 2018. - Step 4: Specify saving S_i and public investment F_i: - S_i = σ [G_i − G_i^{trend}] if G_i − G_i^{trend} > 0, else 0. - F_i = φ [X_i − X_i^{trend}] if X_i − X_i^{trend} < 0, else 0. - SF stock evolution: SFS_{i,t+1} = (1 + i_{i,t}) SFS_{i,t} + (S_i − F_i). - Step 5: Public debt dynamics: F_{i,t+l+1} = (1 + i_{i,t+l+1}) F_{i,t+l} − PPP_{i,t+l} + (S_i − F_i); implicit interest rate i_{i,t} = interest expenditures_t / public debt_{t-1}. - Step 6: Solve for φ ∈ [0,1] meeting transversality: E_i[ SFS_{T+N} / Y_{T+N} ] = SFS_T / Y_T, with SF depletion probability π at horizon T+N and government choices (σ, π, N). - Step 7: Aggregate to RSF; due to imperfect correlation cycles smaller and RSF < sum of individual SFs. - Step 8: Structural fiscal consolidation can be incorporated by shifting projected trends to calibrate public debt and participation incentives. --- ### Appendix III: DGE model overview and calibration highlights - Model: real DGE with N countries, limited factor mobility, constant returns firms, public capital externality Y_{i,t} = Z_{i,t-1}^{ψ} K_{i,t-1}^{α_i} L_{i,t}^{1−α_i}. - Households: supply labor domestically; invest and save; preferences separable or non-separable; η denotes labor supply elasticity. - Frictions: portfolio adjustment costs; private investment adjustment costs. - Calibration (annual frequency): - World risk-free rate R^* = 2 percent. - Depreciation: private capital 0.081; public capital 0.035. - Elasticity of output w.r.t. public capital ψ = 0.05. - Steady state targets as sample averages 2014–2018. - RSF calibration: RSF size = 5 percent of ECCU GDP; recapitalization parameter ρ_{SRSF} = 0.5; replacement parameter κ = 0.5. - Appendix III empirical targets (selected, as labeled): - Output (ECCU average): 100.0. - Domestic government interest rate (ECCU average): 5.3. - External government interest rate (ECCU average): 3.0. - Domestic government debt (ECCU average): 28.4. - External government debt (ECCU average): 37.1. - Public investment (ECCU average): 5.3. - Goods and services (ECCU average): 5.0. - Transfers (ECCU average): 6.3. - Wage bill (ECCU average): 10.1. - Exports (ECCU average): 42.0. - Imports (ECCU average): 54.0. - Private investment (ECCU average): 19.6. --- ### Government fiscal parameters and payment systems context - Tax calibration: - Tax on capital returns set to 20 percent. - Consumption tax adjusted to balance government budget constraint; implicit tax rates range between 9 and 21 percent, averaging 14 percent across ECCU countries. - RSF flows: - Disaster transfer to rebuild share κ of destroyed public capital; net recapitalization payments governed by parameter ρ_{SRSF} and country weights w_{i,i}. - Idle RSF funds invested at international risk-free rate. - RSF law of motion: S S T_v_i = R_{i−1} * S S T_{v,i−1} + Σ_{N_i=1}[SST_vF_{n,i,i} − SST_vSSt_{i,i}]. ### Payment systems, cash costs, and ECCB CBDC motivations - Retail payment system indicators and costs: - Average cash holding per capita ~ US$897 (ECCU average). - Cross-country dispersion: ~US$400 in St. Vincent and the Grenadines (lowest); ~US$1,800 in Montserrat (highest). - Value of checks in ECCU > 160 percent of GDP (World Bank Global Payment Systems Survey). - Number of credit and debit cards per person < 0.8. - ECCB spent EC$8.4 million in 2018 on cash services = US$ 4.9 per capita ≈ 12 percent of its net interest income. - Merchant card fees typically ~5 percent (or more) to merchants in ECCU; cardholder annual fees US$40 to US$170. - ECCB CBDC (DXCD) pilot: - Announced pilot on March 12, 2019 using DLT; partnership with Bitt Inc.; two phases: - Development and testing: March 2019–February 2020. - Rollout and implementation in Antigua and Barbuda, Grenada, Saint Lucia, and St. Kitts and Nevis: April–September 2020. - ECCB expects digital currency adoption would reduce cash demand by half over time and improve retail payment efficiency. - Broader CBDC motivations: - Advanced economies: falling cash use, convenience, negative interest rate implementation, contestability. - Emerging/developing economies: operational and cost efficiency, financial inclusion, digitalization, enhanced financial integrity (non-anonymous CBDC aids monitoring). --- ### Technology adoption, CBDC pilot design, risks, and safeguards - Technology and mobile adoption: - Fixed-line phones in early 2000s ~35 per 100 persons in ECCU; mobile subscriptions reached 120 per 100 persons. - Non-bank private e-payment services limited (e.g., CaribePay, JAD Cash). - Structural constraints: small populations per member (< 0.1 million on average); absence of legal/regulatory framework for mobile wallets. - ECCB DXCD pilot objectives: - Reduce use of cash and checks by half. - Improve retail payment system efficiency. - Stimulate competition in the payments industry. - Preserve cash for convenience. - Key risks identified: - Financial disintermediation if DXCD substitutes for bank deposits: - ECCU banks rely heavily on deposit funding (over 90 percent for some banks; system-wide deposit funding ratio ~69 percent). - Demand deposits pay negligible interest (~0–0.1 percent). - Share of demand deposits ~16 percent of total liabilities; about half held by households and private businesses, equivalent to 5–20 percent of total liabilities. - Deposit-demand regressions (2006–18 for six ECCU economies) show negative and significant coefficients on opportunity cost of holding demand deposits (except one fixed effects household model). - Financial integrity risk if DXCD allows anonymous or large-value transactions; central bank role in customer due diligence unclear. - Cybersecurity and operational risks: malware, fraud; larger impact without modern safeguards. - Pilot design and safeguards: - ECCB sole issuer; two-tier distribution with financial institutions handling customer interaction and AML/CFT. - Pilot participants: ~15 financial institutions in four ECCU members. - Use cases: peer-to-peer, business-to-business, business-to-customer. - Limits: holding and transaction values per wallet limited; DXCD bears no interest; no foreign-exchange use. - ECCB can observe transaction data anonymously and outstanding DXCD per wallet; financial institutions observe identity and purpose for their customers. - Measures to mitigate intermediation risk: small-value instrument, holding/transaction limits, no interest, financial institutions manage deposit exchange and liquidity. - Measures to mitigate integrity risk: DLT-based identity authentication, two-tier AML/CFT responsibilities, transaction size limits. - Measures to mitigate cybersecurity risk: pilot scope limited (no linkage to RTGS, banks' systems, or Automated Clearing House during pilot); post-pilot testing before integration. - Pilot role and recommended follow-up: - Examine intermediation, integrity, and cybersecurity risks; assess policy gaps; gather DLT/CBDC knowledge. - Recommended actions: coordinate ECCB and supervisors on liquidity and funding stress testing; review AML/CFT guidelines; strengthen cybersecurity operations and testing; establish data and privacy governance frameworks. - Limitations: DXCD is new money; DLT relatively untested in central-bank context; connection to wider systems requires further testing. ### Payment-system modernization targets (ECCB) - By 2025 targets: - Increase use of credit cards by 50 percent. - Increase use of debit cards by 40 percent. - Increase use of electronic transfers by 60 percent. - Legal/regulatory context: - Payment System Acts passed between 2008 and 2010 do not reflect rapid technological developments; no specific legislation for e-wallet providers. - E-government modernization can raise electronic payments for government services and reduce unit costs. --- ### Digital payments and productivity: Appendix I empirical summary - Dataset: 96 countries, 2000–2017. - DID-style specification: dependent = log labor productivity; key interaction = time dummy (t>2014) × digital-payment-penetration dummy (increase > thresholds between 2014 and 2017). - Findings: - Rapid growth in digital-payment penetration associated with higher labor productivity. - Interaction coefficient β3: positive in Models 1 and 2 (not significant); positive and significant in Model 3 (High increase). - Selected coefficient estimates (Model 1, Model 2, Model 3): - Log human capital index: 2.591*** (0.137); 2.636*** (0.137); 2.622*** (0.136). - Log R&D expenditure: 0.0519*** (0.0168); 0.0537*** (0.0169); 0.0530*** (0.0168). - Time dummy (t>2014): 0.0266 (0.0190); 0.0406*** (0.0149); 0.0371*** (0.0129). - Digital payment penetration dummy: -0.307** (0.127); -0.0158 (0.116); -0.221 (0.146). - Interaction (time × digital-penetration): 0.0351 (0.0220); 0.0163 (0.0205); 0.0609** (0.0258). - Observations: 1,012; Number of countries: 96; Sample period: 2000–2017. --- Final source attribution: *Source: IMF — Eastern Caribbean Currency Union chapter (content unit 1eccea2020002).* ### References ___________________________________________________________________________________23 ### 1eccea2020002 - References ### Overview and Context - The Revised Treaty of Basseterre (signed June 2010) called for deeper economic integration across OECS members, including a single financial and economic space with free movement of goods, people and capital, and harmonization of monetary and fiscal policies. - The ECCB Monetary Council instituted a public debt to GDP target of 60 percent by 2030. - The ECCU faces large adverse shocks from natural disasters, with an imputed fiscal cost ranging from ½ to 1½ percent of GDP (IMF 2018). ### Fiscal Integration: Rationale and Challenges - Key rationales for greater fiscal policy integration: - Pooling capacity and resources to mitigate small-country disadvantages. - Correcting sub-optimal interactions and excessive competition (e.g., tax incentives, CBI programs). - Leveraging asymmetries across countries for improved risk-sharing. - Constraints and implementation considerations: - Need for resilient and credible country-specific fiscal responsibility frameworks supported by strong fiscal institutions. - Ex-ante mechanisms to deal with natural disasters (self-insurance via national saving funds; contingent financing). - Careful sequencing required; improvements need fiscal space, capacity, and concessional financing. - Risks: cross-border spillovers from weaker members could trigger region-wide crises; political and democratic-accountability issues when centralizing fiscal mechanisms. ### The Impact of Tax Incentives - Trend: ECCU countries have expanded tax incentives (longer tax holidays, longer loss carry-forward years, higher investment tax allowances, lower statutory CIT rates, customs concessions), often not rules-based or coordinated. - Estimation approaches used to measure foregone revenue: - Based on CIT returns: apply standard CIT rate to taxable profits of companies benefiting from tax holidays minus actual CIT collections (may understate foregone revenue). - Based on VAT returns: potential CIT base estimated as one-fourth of the difference between total sales and total purchases in VAT returns. - Based on sector-specific value-added from national accounts: potential CIT base calculated as one-fourth of value added for beneficiary sectors. - Key quantified findings: - Total foregone revenue averaged 5.8 percent of GDP in 2010–18 in the ECCU region, or 21 percent of total tax revenue. - Measured against employment, cost of tax incentives is equivalent to as high as $2,500 per job in the formal sector. - Sectoral and distributional findings: - Tax incentives show some positive impact on tourism sector metrics (e.g., number of hotel rooms). - Empirical results: a decline in the CIT rate associated with an increase in the number of hotel rooms; an increase in foregone revenue by 1 million EC dollars associated with an increase in the number of hotel rooms by 0.2 percent (pooled OLS). - No robust impact found on overall FDI, GDP growth, or employment (Table 1, Appendix I Table 1-4). - FDI to the Caribbean is positively correlated with the stock of public infrastructure during natural disaster shocks, highlighting a tradeoff between granting tax incentives and raising revenue for infrastructure. ### Empirical Specification (selected points from Table 1) - Dependent variable: FDI. - Estimation methods reported include Fixed Effect Panel and GMM. - Reported coefficient highlights (as presented): - CIT rate columns include values: 0.006 0.001 -0.001 0.006 0.028 0.016*. - Tax holiday coefficient examples: -0.042* -0.091***. - Investment allowance rate examples: -0.002 -0.006**. - Foregone revenue reported as -0.001* 0.000. - Public consumption examples: -0.055 -0.088** -0.063 -0.055 -0.087** -0.200***. - REER examples: 0.002 0.002 0.007 0.002 0.009 0.020***. - US PPPGDP examples: 0.591** 0.547** 0.725** 0.590** 0.884*** 0.976***. - Number of flights (lag): 0.028** 0.029** 0.045 0.027** 0.038 0.006. - Infrastructure stock (lag) interactions with natural disaster dummy: 0.054* 0.054* 0.086* 0.054* 0.043 0.073***. - Sample sizes and fit: - Observations reported: 140 140 102 140 96 81. - Number of countries: 10 across specifications. - R-squared examples: 0.565 0.576 0.504 0.565 0.445. ### Citizenship-by-Investment (CBI) Programs: Competition and Coordination - Recent dynamics: - Several countries reduced prices and eased requirements to attract more applications (examples provided in text: Dominica end-2016; St. Lucia; Antigua and Barbuda late 2017; St. Kitts and Nevis early 2018). - ECCU launched an initiative in March 2019 to harmonize CBI programs’ due diligence processes. - Rationale for regional coordination: - Short-term: protect financial integrity and transparency, dispel concerns about abuse, stabilize revenue flows and support financial stability. - Medium-term: coordinate on pricing and use of quotas to sustain revenues. - Theoretical framing: - Modeled as Bertrand price competition where required CBI investment is the price perceived by applicants. - With identical countries, a minimum price or cap on new citizens can improve joint revenues. - With asymmetric demands (two types: low-demand L and high-demand H countries), outcomes differ: - H-type countries charge higher prices and grant more new citizenships than L-type in the Non-Cooperative Equilibrium. - Minimum prices bind more on L-type countries; maximum quotas restrict H-type countries more. - A combination of price and quantity instruments can generate a Pareto efficient outcome for member countries (illustrated with linear demand specification and substitutability parameter 훿>0). ### Recommendations and Policy Implications (explicit and implicit in text) - Strengthen country-specific fiscal responsibility frameworks supported by strong institutions. - Develop ex-ante disaster mechanisms: national saving funds, contingent financing, investment in resilient infrastructure, and comprehensive disaster resilience strategies (noted pilots in Dominica and Grenada). - Coordinate regional fiscal policies to: - Reduce harmful tax competition and the “race to the bottom.” - Consider harmonized approaches to CBI programs (due diligence, pricing, quotas). - Explore pooling and regional stabilization mechanisms to provide buffers against large shocks. - Prioritize raising revenue to finance public infrastructure to attract FDI more sustainably than through tax concessions. *Source: Content extracted from "FISCAL INTEGRATION FOR THE ECCU" (References section) in the supplied PDF content.* ### 15. Pareto improvement can be achieved from a Non-Cooperative Equilibrium to a ### 15. Pareto improvement can be achieved from a Non-Cooperative Equilibrium to a cooperative one. ### Cooperative vs Non-Cooperative Equilibrium: Pareto improvements - Cooperative Equilibrium prices are determined by the intersection of the best response functions of 퐿퐿-types and 퐻퐻-types under cooperation. - In the analysis: - The dashed blue (green) line represents the iso-revenue curve for 퐿퐿-types (퐻퐻-types) corresponding to their Non-Cooperative Equilibrium revenue level. - Any point located above the dashed blue line (to right of the dashed green line) represents prices that generate higher CBI revenue for the 퐿퐿-types (퐻퐻-types) than the Non-Cooperative Equilibrium. - The shaded area in the figure represents sets of prices that are revenue improvements for both sets of countries relative to the non-cooperative scenario. - In Figure 3, equilibrium prices under a minimum price and a maximum quota are higher for Ls and Hs than their respective non-cooperative prices, 푝푝퐿퐿푁푁 and 푝푝퐻퐻푁푁. ### Combination of price and quantity instruments to reach Pareto efficiency - Key intuition: - The minimum price is always binding for the type-L countries (but not always for the type-H countries). The marginal revenue from increasing the minimum price is always higher for H-type countries: an additional increase in the minimum price generates more revenue for H-type countries than for L-type countries. - When the minimum price becomes high enough, its marginal benefit turns negative for L-type countries but remains positive for H-type countries. - Pareto efficiency is achieved when the revenue gain for H-type countries from an additional unit to the minimum price equals the resulting loss of revenue for L-type countries. This Pareto-efficient outcome may not dominate the non-cooperation equilibrium if only one instrument (minimum price or maximum quota) is used. - Only a combination of both the minimum price and the maximum quota can ensure a “fair” distribution of the additional revenue from cooperation and, if chosen appropriately, always guarantee an efficient outcome that is more profitable to both types compared to the non-cooperative equilibrium. - Graphical implications (Figure 4 description): - Any point that lay between the blue line and the green line in Figure 4 can be achieved by setting a combination of a minimum price and a maximum quota. - The green line always passes above (푝푝̂퐿퐿,푝푝̂퐻퐻); otherwise, a maximum quota could make type-H earn more revenue while type-L earn less compared to non-cooperation, which is impossible. - Analogously, the blue line always lay below (푝푝̂퐿퐿,푝푝̂퐻퐻). - The set of efficient price distributions (black line) always passes through the shaded area; therefore a right combination of minimum price and maximum quota exists that is efficient and profitable to all types relative to non-cooperation. ### Empirical precedent - The combination of price and quantity instruments has been effective in cross-country policy coordination, e.g., Party to the Nauru Agreement (PNA) and the Vessel Day Scheme: - The Scheme sets limits on the number of fishing days auctioned by each member per calendar year (quantity instrument) and a minimum benchmark fee for a fishing day (price instrument). - Outcomes included increased revenue through licensing fees, decreased catch to ensure sustainable fishing, and stronger international negotiation leverage for the PNA parties than individual action. *Source: IMF Staff calculations.* ### D. A Regional Stabilization Fund for the ECCU in the Longer Term ### Objectives and motivation for an RSF - Purpose: - Quantify economic benefits and costs from establishing a regional stabilization fund (RSF). - Demonstrate scope for risk-sharing because macro-fiscal shocks across ECCU countries are imperfectly correlated. - Main findings: - An RSF would need to be about half as large as the sum of individual-country saving funds (SFs) while providing a similar degree of insurance. - SFs/RSF designed to incentivize less procyclical fiscal policies would increase average public investment over the cycle at the expense of current spending. - Reallocating spending toward investment yields substantial increases in output, employment, and wages once general equilibrium responses are accounted for. - After incorporating opportunity costs for RSF setup and maintenance, benefits from cross-country risk-sharing are likely to outweigh opportunity costs. ### Procyclicality, financing constraints, and need for self-insurance - Government spending in ECCU countries has been procyclical: increases in revenue during booms lead to higher government consumption; during recessions, low revenue and rigid current spending (wages and transfers) crowd out public investment. - Financing constraints: - ECCU countries are not integrated into global financial markets; bulk financing is external, mostly official multilateral and bilateral loans that are project-based and subject to lags. - Domestic financial markets are shallow; regional securities have limited depth and bonds are seldom traded. - Vulnerability to natural disasters requires immediate financing for rehabilitation and reconstruction; available insurance instruments can be insufficient and/or expensive (e.g., CCRIF under-insurance due to high reinsurance costs). ### Quantitative approach: Monte-Carlo analysis and model setup - Methodology: - Estimate VAR models for each country including five endogenous variables: cyclical fluctuations of output, tax and non-tax revenue, and recurrent and capital expenditure. - VARs capture cyclical depth, persistence, and co-movement and identify the joint distribution of macroeconomic shocks. - Two Monte-Carlo experiments of 1000 simulations each were run, including off-sample projections of the five endogenous variables. - Empirical results: - Impulse-response analysis shows government consumption (current primary spending) and public investment increase when output and revenue are high (procyclical behavior). - The regional aggregation maintains procyclical patterns with less pronounced cycles, indicating offsetting cyclical asymmetries across countries due to timing, exposure to shocks (including natural disasters), economic structure, and CBI program revenue size and volatility. ### Design and size of RSF; risk pooling gains - Saving-disbursement rule in simulations: - Governments save a share of the revenue gap above trend into a SF during booms; savings are disbursed during recessions to finance a share of the predicted decline of capital expenditure below trend. - The investment gap share is calibrated to ensure SF stock stability over time for a given probability of depletion set by governments. - Size implications: - Simulations indicate the RSF size is about ½ of the sum of individual countries’ SFs for the same saving and withdrawal flows. - Example: For a probability of SF depletion at 1 percent and share of “excess consumption” saving of 1, the estimated RSF is about 5 percent of regional GDP, while the sum of individual countries’ SF adds to near 10 percent. - Distributional observation: - All countries benefit from risk pooling under the simulated saving-investment rules: the size of estimated individual saving funds is in all cases larger than the allotted shares under regional pooling, implying implicit cross-country transfers due to cyclical asymmetries. ### General equilibrium (GE) impact of the RSF and public investment reallocation - Investment increases from RSF: - Under an RSF with depletion probability of 0.01, ECCU countries would increase public investment in the range of 0.5–1.5 percent of GDP per year on average. - GE model features: - A collection of small-open economies (islands) with limited factor mobility. - Firms produce with constant returns; country-specific public capital enters as an externality to productivity. - Crowding-in of private investment and employment depends on production function parameters, household labor supply response, and how government reduces current spending. - Output elasticities from a permanent 1 percent of GDP reallocation (increase in capital, decrease in current spending): - Output elasticities are generally higher when the government reduces transfers or reduces the wage bill. - Ranging between 0.9 and 2.1 for reductions in transfers and goods and services (elasticities reflect permanent increases in output levels from spending composition change, not changes in total spending). - Output elasticity is largest when cutting public sector employment because workers reallocate to more productive private activity. - Quantified level impacts: - Using country-specific increases in investment (from RSF simulations), assuming a weighted adjustment of current spending and fixed labor supply, level increases in GDP range between 0.9 percent for St. Vincent to 2.4 percent for Antigua. - Level increases in consumption are even larger, partly because reduction in purchases of government services frees up domestic resources for consumption. ### Calibration details for spending adjustment used in GE simulations - The reduction in current spending is implemented via lower: - purchases of goods and services (weight of 0.5), - transfers (weight of 0.3), - public employment (weight of 0.2). *Source: IMF Staff calculations.* ### 32. Disbursement from the RSF help stabilize the consumption path and speed up the ### 32. Disbursement from the RSF help stabilize the consumption path and speed up the recovery significantly ### Simulation illustration: natural disaster shock in Antigua and Barbuda - Scenario assumptions: - Total damages of 10 percent of GDP with two-thirds thereof borne by the public sector. - Fund size calibrated to 5 percent of ECCU GDP. - Stochastic simulation assumes countries are hit independently by natural disaster shocks with an average annual loss of 1 percent of GDP; the annual probability of a shock hitting a country is 0.1 and resulting damages are 10 percent of GDP. - Key simulated outcomes (comparative, RSF versus No RSF): - Output contraction is reduced directly via quicker reconstruction of public capital and indirectly through crowding-in of private investment. - Private investment in the affected country is higher post-disaster under the RSF. - The fall of consumption in the affected country is much reduced by RSF support, especially on impact but also in the long run. - Consumption in unaffected countries falls slightly under the RSF as these countries are assumed to help recapitalize the fund. - The fund maintains ample resources following the disbursement and is recapitalized within five years. ### Opportunity costs and model framework - Setup for assessing opportunity costs: - Countries set up the stabilization fund by borrowing domestically. - Debt-servicing cost is defrayed by increasing non-distortionary consumption taxes and distortionary capital taxes in equal measure. - The general equilibrium model incorporates the effects of distortionary taxation on capital accumulation. - Long-run results from simulations: - Stabilization benefits are likely to outweigh opportunity costs: with RSF support, consumption in each country would be higher over the long-run and the drag from higher distortionary taxes does not offset these benefits. - Opportunity costs could be lower if the RSF were capitalized more by non-distortionary taxes (example: a carbon tax) or if countries could access grants. - The model likely underestimates utility gains because it captures only changes in consumption levels and not lower consumption volatility brought about by the RSF. - Cost of funds note: - Since RSF assets are assumed to be invested at the world risk-free rate, the effective cost of funds is the difference between domestic and foreign interest rates. ### Guiding principles of the RSF: objective and scope - Main objective: - Improving macroeconomic stabilization through effective risk-sharing by providing transfers to finance public expenditure including capital/reconstruction investment when countries experience negative shocks such as natural disasters. - Scope constraints: - Intended only to address temporary shocks that are unrelated over time and are not consistently positive or negative, to avoid the fund becoming a permanent transfer scheme. - Design should include a framework to monitor shocks and periodic transparent re-assessments of operating rules to correct perceptions of sustained subsidization. ### Governance framework (elements to mitigate moral hazard) - Financing and institutional safeguards: - Financing with fiscal resources through national fiscal accounts performed by all governments. - No involvement of the ECCB to protect its reserve position and the integrity of its quasi currency board arrangement. - Transparency and accounting: - RSF operations should be transparently recorded in fiscal accounts with clear saving and investment rules to address misappropriation risk and clarify sovereign ownership and each sovereign’s net asset position vis-à-vis the RSF. - Fiscal responsibility and conditionality: - Country-level fiscal responsibility frameworks underpinned by the regional debt target are essential complements. - Each country’s medium-term fiscal frameworks should be subject to peer review; compliance should be a criterion for sanctions and for determining access to the fund. - Benefits of adopting a regional fund: (i) limits increase in recurrent spending in good times; (ii) protects fiscal space for public investment which supports growth; (iii) insures a more cost-effective response to shocks; and (iv) reduces public debt dispersion outcomes. - Triggers and management: - Triggers for disbursement should be objective, transparent and automatic; output gap in theory is best candidate but is un-observable in real time—significant natural disasters and/or large declines in GDP growth could be alternative triggers. - The fund should be managed by a supranational agency with authority to: (i) objectively assess damage from shocks and reporting standards; (ii) impose and enforce rules on resource allocations and compliance and penalties for noncompliance; and (iii) build capacity in asset-liability management. - Fiscal and financial transparency is a key guiding principle; decision-making mechanisms (consensus voting versus qualified majority or reverse qualified majority voting) could be considered. ### Modality - Caps on transfers and contributions: - Member countries’ access may be capped by its accumulated savings for a set period of time (example given: 10 years); contributions can be capped to avoid better-performing countries becoming net contributors for an extended period. - Alternatively, inflows could be governed by actuarial mechanisms to ensure no sustained subsidization (example reference: CCRIF’s probabilistic models). - Rainy-day fund with transfers: - Primary function can be a contribution-transfer arrangement; initial prudent limits on transfers beyond each country’s own contributions, expanding gradually as institutions and oversight strengthen. - Voluntary lending options: - Countries could lend voluntarily beyond transfer caps at mutually agreeable interest rates and maturities to help scale up re-allocation in very large shocks and incentivize use of windfall resources such as CBI inflows. ### Sequencing and timing - Recommended multi-stage implementation: - Stage 1: Countries reallocate funds to the RSF but retain very prudent limits on transfers to test framework elements and assess quantitative and institutional parameters without imperiling own resources; concurrently upgrade national fiscal frameworks and transparency. - Stage 2: Gradually relax transfer limits as institutional and sanctions mechanisms gain credibility. - Stage 3: Consider topping up transfers through voluntary lending. - Realistic timeline: - Given complexity and time needed for building regional and upgrading national institutions, the RSF could only be realistically implemented over the long term. ### Costs and challenges - Political and institutional hurdles: - Early-stage operations require a strong institutional framework underpinned by an ECCU-wide consensus, which may be slow to materialize due to asymmetries in member states’ positions. - Benefits could be slow to materialize while administrative expenses and uncertainty could be substantial. - Enforcement and legal risks: - Difficulties in enforcing rules among sovereign states could include: - Non-observance of payment or re-payment commitments. - Inappropriate use of funds received (e.g., for current spending instead of public investment). - Disagreements about the size and nature of shocks. - Potential legal challenges from interpretation differences or rule gaps. - Potential tensions over RSF operation could spill over to broader ECCU collaboration. - Risk of moral hazard is substantial and requires carefully designed enforcement and sanctions mechanisms. ### Conclusions and policy implications - Short-term priorities: - Create fiscal space to both reduce public debt and build resilient infrastructure and insurance buffers for natural disasters. - There is scope for substantial potential of raising revenues in the immediate future. - Long-term role of regional pooling: - Maintaining large resilience buffers exclusively at national levels becomes increasingly costly; regional pooling arrangements could improve cost and effectiveness of policy interventions. - A regional saving-investment fund can be more cost-effective than self-insurance funds, requiring only ½ of the resources compared to the sum of individual countries’ funds for the same stabilization effect. - The RSF saves resources during booms, contains pro-cyclical government consumption, invests proceeds during recessions, leads to higher public investment and increases growth, and can pool scarce institutional capacity. - The RSF requires a strong supranational institutional framework and should complement, not substitute, country-level resilience buffers. - Preconditions for success: - Strong preconditions at national and regional levels, a solid governance framework, and careful sequencing. - Transparent accounting and judicious use of higher revenues primarily for building resilience for tax incentives and CBI revenues coordination. - For the RSF, strong governance pre-requisites would realistically take significant time before full operationalization; countries’ willingness to participate depends on fiscal sustainability prospects of other countries to ensure availability of funds in bad times. - A transparent methodology for assessing revenue and spending positions, possibly with independent institutions’ support, is necessary, together with an accounting mechanism for net-credit appropriation of pooled funds and decisions on access rules (e.g., whether members can become net-debtors of the RSF). - Regional initiatives establishing minimum common standards for Medium-term Fiscal Frameworks and fiscal responsibility frameworks would support national institutions. - Political realities: - Implementation would entail substantial political will; processes may face debates, delays, or failures to reach consensus, and further analysis or tailored technical assistance may be needed to refine institutional details. *Source: IMF staff calculations and analysis as presented in the chapter text.* ### 2. Our control variables include a set of domestic and external indicators. Public ### 2. Our control variables include a set of domestic and external indicators. Public ### Control variables - Public consumption: defined as the sum of compensation to employees and purchases of goods and services; serves as a proxy for the underlying fiscal position. Public consumption is expressed as percent of GDP in regressions. - Real Effective Exchange Rate (REER): indicates competitiveness (real appreciation implies loss of competitiveness). - Real US PPP GDP: indicates push factors from the U.S. economy. - Number of airlifts (number of flights): proxy for air access / attractiveness for foreign investors. - Stock of infrastructure: potential factor indicating attractiveness for foreign investors. - Natural disaster dummy: captures occurrence of natural disasters. - Data transformations: FDI, number of hotel rooms, infrastructure, and U.S. PPP GDP underwent natural logarithmic transformation; government consumption is expressed as percent of GDP. ### Methodology - Dynamic panel specification (following Klemm and Van Parys (2009)): - Follows model: - 퐹퐹퐹퐹퐹퐹 푖푖푖푖 = 훿훿퐹퐹퐹퐹퐹퐹 푖푖푖푖 −1 + 푇푇푇푇푇푇 푖푖푖푖 훾훾 + 푋푋 푖푖푖푖 훽훽 + 휇휇 푖푖 + 휀휀 푖푖푖푖 - Dependent variable: 퐹퐹퐹퐹퐹퐹 푖푖푖푖 - Tax variables vector: 푇푇푇푇푇푇 푖푖푖푖 - Control variables vector: 푋푋 푖푖푖푖 - Country fixed effects: 휇휇 푖푖 - Error term: 휀휀 푖푖푖푖 - Estimators: - One-way fixed effect (FE) estimator to limit selection bias; country fixed effects account for heterogeneity. - System Generalized Methods of Moments IV (IV-GMM) used as robustness to address potential endogeneity concerns. - Due to small sample size, FE estimations are preferred and emphasized; GMM results reported indicatively. - Same specification applied with number of hotel rooms as dependent variable to proxy tourism-sector FDI. ### Results — overall FDI regressions - Tax policy variables: - Neither the CIT rate nor tax incentives are significant either statistically or economically in baseline FE regressions (Columns 1–5) and IV-GMM (Column 5), suggesting no material impact on attracting overall FDI. - Control variables (consistent findings across specifications): - Public consumption: statistically significant and negatively associated with FDI across all specifications. - REER: not a significant driver of FDI. - Number of flights: coefficient positive and statistically significant. - U.S. PPP GDP: coefficient positive and statistically significant. - Infrastructure: significant only when interacted with the natural disaster dummy; positive coefficient on the interaction indicates that raising revenue to finance infrastructure investment might be a more efficient way to attract FDI than granting tax incentives. - Interpretation: - Increase in public consumption interpreted by foreign investors as deterioration in fiscal outlook, deterring FDI. - Better air access and higher U.S. PPP GDP stimulate foreign investment in the region. ### Results — tourism sector (number of hotel rooms) and other outcomes - When using number of hotel rooms as proxy (tourism-focused investment): - Mixed results on tax incentives. - CIT rate: negatively associated with number of hotel rooms; statistically significant when controlling for investment allowance rate — this suggests lower taxes may have some positive impact on tourism-sector investment. - In a pooled OLS regression: an increase in the foregone revenue by 1 million EC dollars is associated with an increase in the number of hotel rooms by 0.2 percent for the ECCU countries (Appendix Table 2). - Tax incentive measures: - Do not seem to have an impact on overall FDI, GDP growth, or employment for the tourism-dependent Caribbean economies (Appendix Table 3 and 4). ### Key numerical findings and significance (selected from Appendix Tables) - Pooled OLS result: foregone revenue increase of 1 million EC dollars → number of hotel rooms increases by 0.2 percent (mentioned in text; Appendix Table 2 reports related coefficients). - Appendix I. Table 1 (Fixed Effect Panel and GMM; dependent variable: number of hotel rooms): reported coefficients include (excerpted layout as in source) - CIT rate (lag): -0.006, -0.006, -0.013**, -0.006, -0.009, 0.006 (various specifications) - Tax holiday (lag): -0.004, -0.074*** - Investment allowance rate (lag): -0.003*, -0.002 - Tax competition index (lag): 0.027 - Foregone revenue (lag): 0.000, 0.001** - Public consumption: 0.006, 0.004, 0.018*, 0.006, 0.010, -0.072*** - US PPPGDP: 0.129**, 0.129**, 0.197***, 0.131**, 0.258***, 0.387** - Infra. stock (lag 2): 0.113***, 0.116***, 0.110**, 0.110***, 0.073, -0.214* - Infra. stock (lag 2) interacted w/ natural disaster dummy: -0.012, -0.012, -0.027*, -0.013, -0.017, 0.000 - Observations: 135, 135, 101, 135, 93, 80 - Number of countries: 10, 10, 10, 10, 10, 10 - R-squared: 0.54, 0.31, 0.319, 0.468, 0.327 - Appendix I. Table 2 (OLS; dependent variable: number of hotel rooms): selected coefficients - CIT rate (lag): -0.017***, 0.000, 0.027***, 0.001, 0.031** - Tax competition index (lag): 0.170** - Foregone revenue (lag): 0.002*** - Public consumption: -0.064***, -0.124***, -0.173***, -0.124***, -0.134*** - Number of flights (lag): 0.033***, 0.041***, 0.050***, 0.041***, 0.085*** - Infra. stock (lag 2): 0.275***, 0.189***, 0.115***, 0.181***, -0.16 - Observations: 135, 135, 101, 135, 93 - Number of countries: 10, 10, 10, 10, 10 - R-squared: 0.882, 0.894, 0.843, 0.899, 0.639 - Appendix I. Table 3 (Fixed Effect Panel; dependent variable: Overall GDP — selected coefficients) - CIT rate: 0.005***, 0.004***, 0.003**, 0.004** - Public consumption: -0.014***, -0.015***, -0.010***, -0.009** - US PPPGDP: 0.067***, 0.067***, 0.116***, 0.108*** - Infrastructure (lag): 0.022*, 0.019, 0.009, 0.026 - Observations: 160, 157, 115, 96 - Number of countries: 11, 11, 11, 6 - R-squared: 0.935, 0.935, 0.921, 0.911 - Appendix I. Table 4 (Fixed Effect Panel; dependent variable: Total Employment — selected coefficients) - CIT rate: 0.004***, 0.005***, 0.003*, 0.005** - Public consumption: -0.007**, -0.006*, -0.004, -0.005 - US PPPGDP: 0.074***, 0.080***, 0.095***, 0.104** - Infra. stock (lag): 0.026*, 0.023, 0.040*, 0.018 - Observations: 150, 147, 105, 86 - Number of countries: 11, 11, 11, 6 - R-squared: 0.82, 0.83, 0.76, 0.79 ### Appendix II — Calculation of Country and Regional Stabilization Funds (SF / RSF) - Step 1: Estimate an unrestricted VAR for each country over sample spanning 1990–2018: - Endogenous vector X_i includes cyclical components of: GDP; tax revenue; non-tax revenue; current primary expenditures; capital expenditures. - Endogenous variables expressed as a share of each indicator’s trend: 푇̂ 푖푖푖푖 = 푇푇 푖푖푖푖 / 푇푇 푖푖푖푖 - Residuals: 휖휖 푖푖 ~ N(0, Ω). - Step 2: Monte-Carlo experiment: - Generate 1,000 forecasts 푋푋 휏휏 for each country using simulated shocks 휖휖̂ 휏휏, τ = t+1,...,T. - Simulations use Choleski factorization Ω = W′W and v_τ ~ N(0,1). - Simulations produce probability density functions for each endogenous variable for years 2019–2030. - Step 3: Transform simulations in deviations from trend into percent of GDP: - Deterministic trend projection: T_{i,t+l} = T̂_{i,t+l} * T_{i,t+l}^{trend} with trend growth rate g; starting points at constant prices of last sample year, 2018. - Step 4: Specify saving S_i and public investment F_i flows vis-à-vis the SF: - Saving: S_i = σ [G_i − G_i^{trend}] if G_i − G_i^{trend} > 0, else S_i = 0. - Public investment: F_i = φ [X_i − X_i^{trend}] if X_i − X_i^{trend} < 0, else F_i = 0. - SF stock evolution: SFS_{i,t+1} = (1 + i_{i,t}) SFS_{i,t} + (S_i − F_i), where i_{i,t} is interest rate on savings. - Step 5: Public debt dynamics in each simulation: - Debt identity: F_{i,t+l+1} = (1 + i_{i,t+l+1}) F_{i,t+l} − PPP_{i,t+l} + (S_i − F_i); F_{i,t+l} > 0. - Implicit interest rate i_{i,t} = interest expenditures_t / public debt_{t-1}. - Step 6: Solve for φ ∈ [0,1] (share of government investment financed from SF) that meets transversality condition: - E_i[ SFS_{T+N} / Y_{T+N} ] = SFS_T / Y_T, with SF probability of depletion π at horizon T+N; (σ, π, N) are parametric government choices. - Step 7: Regional Stabilization Fund (RSF): - Aggregation across countries yields smaller cycles than sum of individual SFs; RSF will be necessarily smaller than sum of individual SFs due to imperfect correlation of cycles. - RSF allows cross-country financing flows implicitly. - Step 8: Structural fiscal consolidation can be incorporated by shifting projected trends in simulations to calibrate public debt and participation incentives. ### Appendix III — Dynamic General Equilibrium (DGE) model (overview) - Model type: real, dynamic, general equilibrium with N countries and limited factor mobility; each country is a small open economy with world interest rate R_i^*. - Households: - Supply labor domestically. - Decide consumption C_{i,i}, investment into domestic private capital F_{i,i}, savings into domestic bonds P_{i,i} and international bonds P_{i,i}^h*. - Interest rates R_i and R_i^{h*} correspond to domestic and international bonds. - Preferences may be separable or non-separable over consumption and labor; η denotes labor supply elasticity. - Firms and production: - Production function includes country-specific public capital Z_{i,t-1} as an externality: - Y_{i,t} = Z_{i,t-1}^{ψ} K_{i,t-1}^{α_i} L_{i,t}^{1−α_i} - Trade: final goods traded with world markets (exports X_{i,t}); no intra-regional trade. - Domestic consumption and investment are Leontief composites with a share φ_i of imported world output. - Frictions: - Portfolio adjustment costs govern asset allocation outside steady state. - Adjustment costs to private investment allow smooth capital dynamics. *Source: IMF staff calculations (excerpts from chapter text and appendices).* ### 2. Government. The government in each country collects taxes on consumption and capital ### 2. Government. The government in each country collects taxes on consumption and capital returns. ### Government fiscal accounts and financing - Total spending components: - Interest expenses. - Public investment (푋푋푖푖,푖푖). - Current primary spending: purchases of goods and services (government consumption) 퐺퐺푖푖,푖푖, transfers (푇푇푖푖,푖푖), and the government wage bill. - Overall fiscal deficit financing: - Issuance of domestic debt (푃푃푖푖,푖푖) or foreign debt (푃푃푖푖,푖푖푔푔∗). - Interest rate on foreign debt denoted 푅푅푖푖푔푔∗. - Tax parameters used in calibration: - Tax on capital returns set to 20 percent. - Consumption tax adjusted to balance the government’s budget constraint, resulting in implicit tax rates ranging between 9 and 21 percent (averaging 14 percent across ECCU countries). ### Regional Stabilization Fund (RSF): design and flows - RSF is the sole inter-country link when regional markets are closed; used for consumption risk-sharing. - Disaster response and recapitalization: - If hit by a natural disaster, affected country receives transfer 푆푆푇푇푣푣 푆푆푆푆푡푡푖푖,푖푖 to rebuild share 휅휅 of destroyed public capital. - Net recapitalization payments 푆푆푇푇푣푣퐹퐹 푛푛푖푖,푖푖 reflect payments to replenish the fund net of interest revenue: - 푆푆푇푇푣푣퐹퐹 푛푛푖푖,푖푖 = (1−휌휌푆푆푆푆푆푆) 푤푤푖푖,푖푖 [푆푆푇푇푣푣 − 푅푅푖푖−1∗푆푆푇푇푣푣푖푖−1] + 휌휌푆푆푆푆푆푆 푆푆푇푇푣푣퐹퐹푛푛푖푖. - 휌휌푆푆푆푆푆푆 regulates speed of recapitalization; 푤푤푖푖,푖푖 are country weights. - Idle RSF funds assumed invested at the international risk-free rate. - If the fund was fully capitalized in previous period (푆푆푇푇푣푣푖푖−1 = 푆푆푇푇푣푣), governments would receive a payment from the RSF (푆푆푇푇푣푣퐹퐹 푛푛푖푖,푖푖 < 0). - Law of motion for RSF balance: - 푆푆푇푇 푣푣푖푖 = 푅푅푖푖−1∗푆푆푇푇 푣푣푖푖−1 + Σ_{푁푁푖푖=1}[푆푆푇푇푣푣퐹퐹 푛푛푖푖,푖푖 − 푆푆푇푇푣푣푆푆푆푆𝑡𝑡 푖푖,푖푖]. ### Resource constraint and market equilibrium - Country i resource constraint (combining household and government constraints): - 푌푌푖푖,푖푖 + 푃푃푖푖,푖푖 ℳ푖푖,푖푖 + Σ_{푚푚=푔푔,ℎ} 푃푃푖푖,푖푖 푚푚∗ + 푆푆푇푇푣푣푆푆푆푆 𝑡𝑡푖푖,푖푖 = 퐶퐶푖푖,푖푖 + 퐹퐹푖푖,푖푖 + 푋푋푖푖,푖푖 + 퐺퐺푖푖,푖푖 + 푃푃푖푖,푖푖 풳풳풳푖푖,푖푖 + Σ_{푚푚=푔푔,ℎ} 푅푅푖푖−1푚푚∗ 푃푃푖푖,푖푖−1푚푚∗ + 푆푆푇푇푣푣퐹퐹 푛푛푖푖,푖푖. - Total imports ℳ푖푖,푖푖 defined as sum of imported inputs for consumption and investment (private and public): - ℳ푖푖,푖푖 = 휙휙푖푖[퐶퐶푖푖,푖푖 + 퐹퐹푖푖,푖푖 + 퐺퐺푖푖,푖푖 + 푋푋푖푖,푖푖]. - Prices: 푃푃푖푖,푖푖 ℳ and 푃푃푖푖,푖푖 풳풳 denote import and export prices respectively. - Equilibrium defined as sequences of prices and quantities satisfying household and firm optimality and clearing in output, capital, bonds, and labor markets. ### Calibration and steady-state targets - Frequency and key parameters: - Calibrated at annual frequency. - World risk-free rate 푅푅∗ set to 2 percent. - Depreciation rates: private capital 0.081; public capital 0.035. - Elasticity of output w.r.t. public capital 휓휓 set to 0.05. - Data and assumptions: - Steady state targets computed as sample averages over 2014 to 2018. - Interest rates on domestic and foreign debt are implicit interest rates inferred from fiscal accounts. - Public sector employment inferred from fiscal data assuming public sector wages set 15 percent above private sector. - External household debt (푃푃푖푖ℎ∗) assumed zero in steady state. - No-arbitrage implies steady-state interest rate on external household debt (푅푅푖푖ℎ∗) equals interest rate on domestic bonds (푅푅푖푖). - RSF calibration: - Size of RSF set to 5 percent of ECCU GDP. - Recapitalization parameter 휌휌푆푆푆푆푆푆 set to 0.5. - Replacement parameter 휅휅 set to 0.5. - Appendix III empirical targets (selected entries, in percent or percent of ECCU GDP as labeled): - Output (ECCU average): 100.0. - Domestic government interest rate (ECCU average): 5.3. - External government interest rate (ECCU average): 3.0. - Domestic government debt (ECCU average): 28.4. - External government debt (ECCU average): 37.1. - Public investment (ECCU average): 5.3. - Goods and services (ECCU average): 5.0. - Transfers (ECCU average): 6.3. - Wage bill (ECCU average): 10.1. - Exports (ECCU average): 42.0. - Imports (ECCU average): 54.0. - Private investment (ECCU average): 19.6. ### Impact of greater public investment: elasticities and mechanisms - Method: - Steady state equilibrium equations differentiated to derive elasticities of output with respect to public capital, 푡푡푑푑/푑푑푡푡푑푑/푑푑 (elasticity notation preserved as in source). - Elasticities derived for separable period utility and non-separable utility (no wealth effects on labor supply). - Utility specifications: - Separable: 푈(퐶퐶,퐿퐿) = log퐶퐶 − 휒휒 퐿퐿^{1+휂휂 −1}(1 + 휂휂 −1)^{−1}. - Non-separable (no wealth effects on labor): 푈(퐶퐶,퐿퐿) = log(퐶퐶 − 휒휒 퐿퐿^{1+휂휂 −1}(1 + 휂휂 −1)^{−1}). - Key determinants of elasticities: - Degree of crowding-in of private capital (shift of capital demand via capital share 훼훼푖푖 and elasticity parameter 휓휓). - Impact of labor demand shifts on equilibrium employment (slope of labor supply via parameter 휂휂). - Type of current spending adjustment affects labor supply via wealth and reallocation effects: - Reductions in purchases of goods and services free resources for domestic final use and shift labor supply leftward (under separable preferences). - Reducing government wage bill via fewer public employees shifts effective labor supply to private sector rightward (reallocation effect). ### Opportunity costs of RSF capitalization - Approach: - Steady state equilibrium equations differentiated to express opportunity costs in terms of steady state consumption. - Assumed fund of 5 percent of GDP capitalized through issuance of domestic government debt; interest costs financed by raising one tax instrument or combination. - Effects: - Raising tax on capital returns shifts capital demand left, reducing private investment, output, and consumption. - Opportunity costs derived assuming fixed labor supply (휂휂 = 0), so wealth effects on labor supply are absent. - Change in consumption tax (under fixed labor supply) does not impact equilibrium output. - Main text exercise assumes half of debt-servicing cost for the fund is raised through the distortionary tax on capital returns. ### Stabilization benefits: simulation approach - Method: - Full multi-country model simulated with natural disaster shocks to quantify benefits from macroeconomic stabilization. - Natural disaster shocks assumed uncorrelated across countries. - With RSF: simulate long horizon and record average consumption for each country. - Without RSF disbursements following disasters (휅휅 = 0): rerun model. - Stabilization benefits for each country defined as percentage difference in consumption levels between simulations. ### Payment systems, cash costs, and central bank digital currency (CBDC) motivations (ECCU context) - Retail payment system characteristics: - Extensive use of cash and checks; underutilized electronic payments. - Average cash holding per capita about US$900 in the ECCU (figure notes ECCU average US$897). - Wide cross-country dispersion: e.g., about US$400 in St. Vincent and the Grenadines (lowest) and US$1,800 in Montserrat (highest). - Value of checks in ECCU totaled over 160 percent of GDP (World Bank Global Payment Systems Survey). - Number of credit and debit cards per person in ECCU less than 0.8; growth stagnant during 2010–15. - Bank credit transfers usage is quite low in ECCU (relative to comparators). - Cash-related costs: - ECCB spent EC$8.4 million in 2018 on cash services, equivalent to US$ 4.9 per capita or about 12 percent of its net interest income. - Cash costs include importation, transportation, distribution, insurance, and security. - Costs and limitations of electronic payments: - Merchant card fees typically charged at a rate of 5 percent (or more) to merchants in ECCU, compared to 1–3 percent (in the U.S. and the U.K.). - Cardholders often pay annual membership fees ranging from US$40 to US$170. - Fees for wire transfers in ECCU generally less than U.S. banks but comparable to U.S. credit unions (outgoing) or higher (incoming); fees in ECCU are significantly higher than in Europe where faster payments exist. - Limited range of mobile banking services in ECCU, except for a few banks. - ECCB CBDC pilot: - ECCB exploring digital version of EC dollar (DXCD); announced pilot on March 12, 2019 using distributed ledger technology (DLT). - Pilot partnership with Bitt Inc.; two phases: development and testing (March 2019–February 2020); rollout and implementation in Antigua and Barbuda, Grenada, Saint Lucia, and St. Kitts and Nevis (April–September 2020). - ECCB expects digital currency adoption would reduce demand for cash (by half over time) and improve retail payment system efficiency. - Broader motivations for CBDC (contextual summary cited from IMF policy literature): - Advanced economies: falling cash use, robustness, convenience, potential negative interest rate implementation, contestability of payment markets. - Emerging/developing economies: improving operational and cost efficiency, financial inclusion, support digitalization, enhancing financial integrity (non-anonymous CBDC could enhance monitoring). *Source: 1eccea2020002 - 2. Government. The government in each country collects taxes on consumption and capital returns.* ### 8. Another puzzle is that technology is ### 8. Another puzzle is that technology is available but underutilized ### Technology adoption and payment services in the ECCU - Mobile phones enabled technological “leapfrogging”: fixed line phone spread in the early 2000s was about 35 per 100 persons in the ECCU (higher than Latin America, below 20 per 100 persons; lower than North America, nearly 70 per 100 persons). - Mobile subscriptions have reached 120 per 100 persons today. - Despite high mobile penetration, non-bank private e-payment services are limited: only a handful (e.g., CaribePay and JAD Cash in St. Kitts and Nevis) have emerged with limited scope and scale. - Structural constraints: - Very small population per member economy, averaging less than 0.1 million. - Comparators: about 1 million in other Caribbean economies and 40 million in East African economies (noted for M-Pesa’s success). - Absence of a legal and regulatory framework specifically for mobile wallet services. ### ECCB objectives for issuing a central bank digital currency (DXCD) - Main stated objectives of the DXCD pilot: - Reduce the use of cash and check by half. - Improve efficiency of the retail payment system. - Stimulate competition in the payment industry to support economic activity and growth. - Design intention: modernize the payment system while preserving cash for its convenience. ### Key risks and analytical findings regarding CBDC issuance - Monetary policy implications expected to be limited under the ECCB's fixed exchange rate regime. - Main risks to consider: - Financial disintermediation risk if DXCD is a close substitute to bank deposits. - ECCU banks rely heavily on deposit funding (over 90 percent of total liabilities for some banks; system-wide deposit funding ratio about 69 percent). - Demand deposits pay only negligible interest (about 0-0.1 percent). - Share of demand deposits is about 16 percent of total liabilities, about a half of which are held by households and private businesses, equivalent to 5-20 percent of total liabilities. - Empirical evidence: negative correlation between the opportunity cost of holding demand deposits and the share of demand deposits in total deposits; deposit demand regressions for six ECCU economies (2006–18) show opportunity cost coefficients consistently negative and significant (except one fixed effects household model), implying a decline in opportunity cost could shift funds to a default risk-free CBDC away from bank demand deposits. - Financial integrity risk if DXCD allows anonymous and/or large-value transactions; questions on central bank role in customer due diligence and monitoring illicit activity. - Cybersecurity and operational risks: malware and fraud affect payments, clearing, and settlement systems; potential for larger fraud impact without modern safeguards, detection, and prevention controls. ### ECCB approach and DXCD pilot design features (safeguards) - Overall design principles: - ECCB sole authority to issue and redeem DXCD; full control of supply. - Preserve a “two-tier system”: - Private sector (financial institutions) to interact with customers and carry out AML/CFT and customer due diligence. - Central bank to provide trust and manage the DXCD scheme in line with payment system policies. - Pilot participation: about 15 financial institutions in four ECCU members (Antigua and Barbuda, Grenada, St. Kitts and Nevis, and St. Lucia). - Use cases: peer-to-peer, business-to-business, business-to-customer transactions. - Limits: holding and transaction values per wallet are limited; no interest accrued; no use for foreign exchange transactions; ECCB does not intend to eliminate physical cash. - Process of issuance: - Financial institutions request DXCD from the ECCB; ECCB sends DXCD using a DLT system; financial institutions distribute DXCD to clients in exchange for deposits or cash. (Clients do not have direct access to central bank accounts.) - Access to transaction data: - ECCB can observe each transaction data (anonymously) and outstanding DXCD stock per wallet, but not detailed payer/payee identity or transaction purpose. - Financial institutions fully observe identity and purpose for transactions involving their customers and maintain client databases; they are responsible for AML/CFT compliance for their customers. - Measures to mitigate financial intermediation risk: - DXCD designed as a small-value retail payment instrument with limits on holdings and transaction values, informed by existing cash and electronic transaction limits and AML/CFT rules. - DXCD will not bear interest to discourage saving in DXCD. - Financial institutions allowed to control amounts of DXCD depositors can exchange for deposit accounts to assist liquidity management. - Measures to mitigate financial integrity risk: - Transaction size limits. - Use of DLT for identity authentication and payment tracking. - Retention of two-tier distribution to leverage private sector AML/CFT capacity; ECCB to supervise compliance rather than directly monitor individual transactions. - Measures to mitigate cybersecurity risk: - Pilot scope limited in system integration: DXCD system not planned to be linked to ECCB core payment systems (such as RTGS), banks' operating systems, or the Automated Clearing House during pilot. - Post-pilot testing required before integration to assess vulnerabilities after connection to other systems. ### Role of the pilot and recommended follow-up actions - Purpose of the pilot: - Examine financial intermediation, financial integrity, and cybersecurity risks; assess policy gaps; gather knowledge about DLT and CBDC impacts. - Specific policy and supervisory considerations highlighted: - Financial intermediation: coordinate ECCB and national supervisors to analyze liquidity and funding of each bank and credit union, including liquidity stress testing. - Financial integrity: review and potentially upgrade AML/CFT operational guidelines and regulations to cover digital currency use. - Cybersecurity: consider developing proactive cybersecurity IT operations (e.g., “real-time 24/7” monitoring), strengthen IT operational resiliency (asset, control, change and vulnerability management), enhance security testing before deployment, improve third party and cloud risk management, and allocate more supervisory resources to ensure robust cyber practices at financial institutions. - Data and privacy governance: establish frameworks to ensure protection of sensitive financial or personal data. - Limitations and learning needs: - DXCD is new money; DLT technology is relatively untested in this central bank context; knowledge and experience are limited. - Even with pilot safeguards, connection to wider payment and operating systems will require additional rounds of testing to assess risks and vulnerabilities. *Source: 1eccea2020002 - 8. Another puzzle is that technology is available but underutilized (PDF chapter/section).* ### 18. The ECCB’s move to pilot digital currency reflects its dissatisfaction over the slow ### 18. The ECCB’s move to pilot digital currency reflects its dissatisfaction over the slow progress in payment system modernization ### Rationale for the ECCB digital-currency pilot - The ECCB’s pilot reflects dissatisfaction over the slow progress in payment system modernization. - Due to limited economies of scale, the private sector may not be able to fully exert its potential on its own, either by providing e-money or mobile payment services at a low cost. - The ECCU case contrasts with larger economies where a central bank digital currency could be redundant because the private sector already plays an important role in retail payments. ### Pilot design, risks, and recommended next steps - The digital currency pilot project should proceed cautiously, as planned. - Safeguard measures embedded in the design would help mitigate risks, but risks pertaining to central bank digital currency and DLT have yet to be fully understood. - After the pilot, the ECCB plans to thoroughly review its results. - Further work may be warranted, especially to: - further test the digital currency system; - strengthen cybersecurity and AML/CFT operations; - update legal and regulatory frameworks. - The authorities are encouraged to conduct a comprehensive cost and benefit analysis of the new digital currency ecosystem. ### Limits of a central bank digital currency (CBDC) and complementary actions - Central bank digital currency alone would not solve the problem. - To minimize the risk of financial disintermediation, CBDC should be designed as a means of payment for small value transactions. - A small-scale CBDC would be unlikely to serve the business needs of small and medium-sized enterprises, which need fast and cheap payment methods for larger value transactions. - Accordingly, in tandem with piloting DXCD, the authorities must continue efforts to modernize payment systems based on bank deposits. ### Payment-system modernization targets and legal/regulatory context - The pilot project is part of the ECCB’s broader initiative to promote a digital economy and modernize payment systems. - By 2025, the ECCB aims to: - increase the use of credit cards by 50 percent; - increase the use of debit cards by 40 percent; - increase the use of electronic transfers by 60 percent. - The lack of updated legislation has deterred the development of FinTech and non-bank payment services; authorities plan to review their legal and regulatory frameworks. - Accelerating e-government reform would increase electronic payments for government services (e.g., transfer programs and pension), helping increase electronic transaction volumes and reduce unit costs for banks and payment service fees. - Note: The Payment System Act was passed in various member economies between 2008 and 2010 and does not reflect rapid technological developments over the past decade; there is no specific legislation for new payment system providers using e-wallets. - In most ECCU economies, Ministries of Finance use a centralized, fully automated government financial management system, but weaknesses in payments infrastructure still require printing and hand delivery of paper checks to execute spending. ### Appendix I — Estimating relationships between digital payments and productivity growth: setup - Dataset covers 96 countries for the sample period from 2000 to 2017 (annual). - Model specification (difference-in-difference style): - Dependent variable: log labor productivity (LLP_g_LP_i_t). - Key explanatory variables: - F_t>2014: time dummy = 1 if year t is greater than 2014. - F_DDP: digital payment penetration dummy = 1 if increase in digital payment penetration is greater than certain thresholds between 2014 and 2017 (thresholds: low = below 25 percentile, median = 50 percentile, high = above 75 percentile). - Interaction term F_t>2014 * F_DDP; main interest is coefficient β3. - Controls X_i_t include human capital index and R&D expenditure. - Fixed effects μ_i capture time-invariable heterogeneity across countries. - Error term ε_i_t ~ IID(0, σ_ε^2). - Data limitation: digital payments data available only for 2014 and 2017; the increase from 2014 to 2017 is treated as a “natural experiment.” ### Appendix I — Empirical findings - Results suggest countries experiencing rapid growth in digital-payment penetration are associated with higher labor productivity. - The estimated coefficients on the interaction term β3 are: - positive in Model 1 and Model 2 but not significant; - positive and significant in Model 3, implying that countries with large increases in digital-payment penetration are associated with higher labor productivity in 2014–17. ### Appendix I Table 1 — Difference in Difference Regressions (selected coefficients) - Dependent variable: log labor productivity - Observations: 1,012 - Number of country: 96 - Sample period: 2000-2017 - Change in digital payment per population b/w 2013 and 2017: - Model 1: <25 percentile (Low) - Model 2: = 50 percentile (Median) - Model 3: > 75 percentile (High) Selected coefficient estimates (standard errors in parentheses; significance: *** 1 percent, ** 5 percent, * 10 percent) - Log human capital index: - Model 1: 2.591*** (0.137) - Model 2: 2.636*** (0.137) - Model 3: 2.622*** (0.136) - Log R&D expenditure: - Model 1: 0.0519*** (0.0168) - Model 2: 0.0537*** (0.0169) - Model 3: 0.0530*** (0.0168) - Time dummy (t>2014 = 1): - Model 1: 0.0266 (0.0190) - Model 2: 0.0406*** (0.0149) - Model 3: 0.0371*** (0.0129) - Digital payment penetration dummy: - Model 1 (Low): -0.307** (0.127) - Model 2 (Median): -0.0158 (0.116) - Model 3 (High): -0.221 (0.146) - Time dummy * country dummy (interaction term of interest): - Model 1: 0.0351 (0.0220) - Model 2: 0.0163 (0.0205) - Model 3: 0.0609** (0.0258) ### Appendix I — Data description (sources and definitions) - Labor productivity: - Calculated as real GDP at chained PPPs (in 2011 US dollars) divided by total employment. - Data source: University of Groningen Penn World Table. - Digital payment penetration: - Calculated as the proportion of population having made or received digital payments over the past year. - Data available for 119 economies; outliers removed by truncating the top and bottom 5 percent of the dataset based on labor productivity. - Data source: World Bank’s Global Findex database. - Human capital index: - Calculated as the contributions of health and education to the productivity of the next generation of workers. - Ranges between 0 and 1; takes the value 1 only if a child born today can expect to achieve full health and complete education potential. - Data source: World Bank Development Indicators. - R&D expenditure: - Gross domestic expenditures on research and development (R&D) as a share of GDP; includes capital and current expenditures in business enterprise, government, higher education and private non-profit; covers basic research, applied research, and experimental development. - Data source: World Bank Development Indicators. *International Monetary Fund — Eastern Caribbean Currency Union chapter (content unit 1eccea2020002).* --- _Source: https://www.imf.org/-/media/files/publications/cr/2020/english/1eccea2020002.pdf_