## _sdn1210

## Source details

**Canonical URL:** [_sdn1210](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/sdn/2012/_sdn1210.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/sdn/2012/_sdn1210.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/sdn/2012/_sdn1210.pdf.json)

---

### Executive summary: context, framework, and high-level findings
- Context and purpose:
  - Financial crisis highlighted cross-border linkages and volatile capital flows, prompting interest in multilateral principles for capital-source and capital-recipient policies.
  - Note discusses analytical underpinnings, extends prior Fund analysis, and draws implications for desirability of multilateral coordination.
  - Not intended to prejudge IMF’s institutional view on capital flows or to guide IMF surveillance.
- Core analytical framework and assumptions:
  - Starts from a stylized neoclassical free world and introduces departures to identify spillovers, externalities, and welfare implications.
  - Countries may be large (able to influence world interest rates) or price-takers.
  - Interventions considered: capital controls (discriminate by residency), prudential regulations (do not discriminate by residency), and monetary/fiscal policies insofar as they affect cross-border capital flows.
  - Measures may be temporary (cyclical) or structural (e.g., closure to certain flows).
- Key theoretical findings:
  - Terms-of-trade manipulation is theoretically possible: large creditors or debtors could restrict supply/demand of capital to affect world interest rates.
  - Domestic distortions (e.g., excessive foreign borrowing, learning-by-doing externalities, financial-stability risks) can justify national interventions that reduce risky inflows.
  - A national tax on foreign borrowing or sterilized FX intervention combined with inflow controls can be welfare-improving relative to no intervention under specific domestic distortions.
  - A capital control by a country with a domestic distortion reduces that country’s inflows (or raises its current account surplus), lowers world demand for capital, and depresses world interest rates — a pecuniary externality that can be globally efficient under some conditions.
- Multilateral implications and spillovers:
  - Three potential multilateral concerns:
    1. Capital controls may vitiate external adjustment (e.g., inflow controls sustaining an undervalued currency).
    2. Recipient-country controls can deflect inflows onto other recipients, exacerbating their inflow problems.
    3. Source-country policies (monetary policy, prudential regulation) that increase the volume or riskiness of outflows may worsen recipient-country problems.
  - Spillovers do not automatically preclude justified national policies; but if policies exacerbate distortions in other countries and those countries face high costs to respond, multilateral coordination is likely beneficial.
  - Coordination may require borrowers to reduce inflow controls or lenders to partially internalize risks from large or risky outflows (lender participation is politically and practically thornier).
- Empirical evidence (surveyed):
  - Capital-deflection evidence between recipient countries is scant: most studies find no, quantitatively small, or ambiguous deflection effects.
  - Advanced-economy interest rates are a key determinant of capital flows to EMEs and particularly of inflow surges.
  - Pervasive capital account regulations can help sustain undervalued exchange rates over long periods, raising multilateral issues.
- Policy implications and recommendations (overview):
  - Red flags for multilateral attention:
    - Capital account regulations unjustified on macroeconomic or financial-stability grounds but aimed at competitive advantage.
    - Policies that deflect flows excessively across countries.
    - Policies that transmit risk from source to recipient countries.
  - Practical coordination directions (broad, judgment-based):
    - Use inflow controls less intensively once spillovers are internalized, particularly during generalized surges to multiple countries.
    - Source-country monetary and prudential policies should weigh spillovers on global flows and on macroeconomic and financial-stability risks facing recipient countries.
    - Supervisory policies in source countries should internalize risks imposed on foreign countries by domestic financial institutions.
    - Multilateral coordination should focus on internalizing spillovers rather than rigid capital-account rules; any rules are likely to be “soft” and judgment-based.
  - Where controls underpin beggar-thy-neighbor strategies or terms-of-trade manipulation, multilateral coordination and red flags are warranted.
- Organizational note:
  - Sections: II (pure theory), III (empirical literature), IV (implications), V (conclusion).
  - Boxes include Box 1 (Learning-By-Doing Externalities), Box 2 (Equivalence of Capital Controls and Reserve Accumulation), Box 3 (Global Efficiency of Correcting Domestic Externalities).

### Box 1 — Learning-By-Doing Externalities in Exportable Production: model, insights, and policy trade-offs
- Model setup and production technology:
  - Two-period traded/non-traded good economy with international borrowing/lending at rate (1+r).
  - Learning-by-doing externality: higher production of traded goods raises future productivity in tradable sector.
  - Traded goods: yT1 = f(l1); yT2 = A(·) f(l2); f'(·) > 0, f''(·) < 0; A'(·) ≥ 0; A''(·) < 0.
  - Non-traded goods: yN1 = g(l1); yN2 = g(1−l); g'(·) > 0, g''(·) < 0.
- Decentralized equilibrium vs national planner:
  - Competitive equilibrium sets marginal rate of transformation equal to real exchange rate p (equation (1) as presented).
  - National social planner internalizes externality and allocates more labor to traded goods (equation (2) as presented).
  - Optimal production subsidy to reproduce planner’s allocation: s* = '() + ( ) / (1) Ay f l r > 0 (equation (3) as presented).
  - Effect: producers face more depreciated real exchange rate (given by 1/ps) while consumers face p.
- Capital controls and sterilized intervention as second-best:
  - Production subsidy is first-best but may be budgetarily or politically infeasible.
  - Alternative: controls on capital inflows plus sterilized intervention that absorbs an amount R of inflows; cost of sterilization ≡ (i−r)R (as presented).
  - Necessity of limits on inflows because sterilization cost otherwise infinite.
  - Welfare impact near R = 0: representative citizen’s indirect utility increasing in R at R=0 (equation (4) as presented); at R=0, i = r so first term zero; second term positive as intervention depreciates real exchange rate and reallocates labor to traded sector.
  - Shortcoming: controls-cum-intervention distort consumption, induce excessive saving, and generate an excessive current account surplus — suboptimal relative to production subsidy.
- International spillovers and pecuniary externalities:
  - One country imposing capital controls lowers world interest rates and can deflect capital, exacerbating inflow problems elsewhere — a pecuniary externality.
  - If all excessive borrowers act, lower world interest rates benefit borrowers and hurt creditors but need not harm global efficiency.
  - Caveat: when learning-by-doing is tied to exportables production and export expansion cannot be replicated by all, unilateral controls-cum-undervaluation can be unilaterally and multilaterally inefficient.
- Key findings and policy implications (bulleted):
  - Finding: Learning-by-doing in tradable production creates a positive production externality that private firms do not internalize, leading to underproduction of tradables in decentralized equilibrium.
  - Policy implication: First-best intervention is a production subsidy targeted to traded goods; optimal subsidy s* is strictly positive (equation (3)).
  - Alternative policy: Capital controls plus sterilized FX intervention can raise tradables output and be welfare-enhancing versus no intervention, but are second-best because they distort consumption (excess saving) and generate excessive current account surplus.
  - International implication: Unilateral undervaluation via controls and sterilized intervention is worse for other countries than a production subsidy because it undermines others’ tradables promotion and shifts intertemporal terms of trade against creditors.
  - Practical constraint: Subsidies require budgetary resources and taxing nontraded sectors may be infeasible where those sectors are informal.
  - Risk: If externality does not exist or infant industries fail to mature, subsidies or undervaluation can reduce social welfare.

### Box 5 — Gains from Coordination under Costly Capital Controls: calibration, scenarios, and quantified outcomes
- Calibration and baseline:
  - Two-region model: borrowing region and lending region; relative size three-to-one.
  - Laissez faire equilibrium:
    - Borrowing region imports capital in the amount of 3.1 percent of GDP.
    - World interest rate: 4.1 percent.
  - Externality in borrowing region: i ξ = 6 percent.
  - Resource cost of imposing capital controls: functional form 2().= (as reported).
  - Parameter α set to 0.05.
- Uncoordinated (Nash) outcome among national planners:
  - Borrowing-region planners correct two-thirds of the externality and impose a 4 percent inflow tax.
  - Lenders impose no controls.
  - Resulting equilibrium:
    - Capital flows in borrowing region decline to 2.3 percent of GDP.
    - World interest rate falls to 3.1 percent per year.
    - Welfare: borrowers gain equivalent of a 0.048 percent increase in consumption; lenders lose .009 percent due to lower interest rate.
    - If borrowers compensated lenders for their loss, borrowers would still have a welfare gain of 0.020 percent.
- Key quantitative changes (selected figures reported):
  - Laissez faire:
    - Borrower capital control: 0%
    - Borrower CA/GDP: -3.09%
    - Interest rate: 4.12%
    - Δwelfare LF: 0
  - Nash equilibrium (uncoordinated national planners):
    - Borrower capital control: 4.04%
    - Borrower CA/GDP: -2.32%
    - Interest rate: 3.09%
    - Δwelfare LF: borrower 0.0479%, lender -.0088%
    - Δwelfare Nash: 0.0017% (net global welfare gain reported under coordination; see context below)
  - Full coordination:
    - Borrower inflow tax reduced from 4 percent to 3.1 percent.
    - Lending region imposes a control of 0.9 percent on outflows.
    - Sum of controls almost unchanged from Nash, but aggregate cost () falls due to convexity.
    - Net global welfare gain from coordination: equivalent to a .0017 percent increase in consumption of the borrowing country.
- Welfare and efficiency observations:
  - Moving from laissez faire to uncoordinated Nash yields gains for borrowers much larger than additional gains from Nash to full coordination.
  - With convex cost of controls, it is more efficient to split regulation between borrowers (inflow taxes) and lenders (outflow controls).
  - Coordination with restricted lender (global planner cannot impose controls in lending region) yields smaller welfare gains than full coordination; compensatory transfers from lenders to borrowers are needed for a Pareto improvement.
- Coordination among borrowers:
  - Borrower-country coordination is necessary because uncoordinated actions can produce excessively high controls and flows too low relative to coordinated optimum.
- Source-country rationale for lender participation:
  - Convex cost structure implies global efficiency requires splitting burden between inflow and outflow controls.
  - Obstacles: capital-exporting countries do not directly benefit from coordinated equilibrium unless they internalize externalities.
  - Reasons lenders may restrict outflows: public-good motive (global financial stability), prudential regulation to limit losses to domestic financial institutions, incidental terms-of-trade/monopoly effects on lending.
  - Efficiency implication: when lenders have incentives, optimal split equalizes marginal costs of inflow and outflow controls.

### Empirical findings on capital account policies and spillovers (surveyed evidence)
- General empirical patterns:
  - Evidence mixed; stronger findings for effects on composition of flows (e.g., short-term debt) than on aggregate volume.
  - Identification challenges: countries facing large inflows are more likely to impose controls, complicating causal inference.
  - Most evidence suggests capital controls have little effect on overall flows for countries with relatively open capital accounts; effectiveness depends on extensiveness of controls, enforcement capacity, and incentives to circumvent.
- Deflection and cross-border spillovers:
  - Direct evidence of cross-border deflection is mixed: some countries experience increases in inflows when others impose controls, others experience decreases (Forbes et al. (2011); IMF (2011) reach mixed conclusions).
  - Deflection-away result implies a capital control war between capital-receiving countries is unlikely; deflection may be efficient where countries facing excessive inflows are more likely to impose controls.
- Role of source-country policies:
  - Advanced-economy interest rates are among the most important determinants of capital flows to EMEs.
  - Spillovers from monetary, financial, fiscal, regulatory, and structural policies operate in part through cross-border financial flows and can worsen pre-existing distortions.
  - Spillovers from macroprudential policies occur where domestic prudential or regulatory deficiencies amplify risky cross-border capital flows.

### Implications summarized (four implications drawn)
- Implication 1: Capital controls should not substitute for warranted external adjustment.
  - Discretion on use circumscribed, particularly when currency is undervalued from a multilateral perspective.
  - Presumption that undervaluation combined with inflow controls should raise red flags.
  - Exceptions: safeguarding financial stability when inflow surges exacerbate risks; genuine learning-by-doing externalities (high bar).
- Implication 2: Countries should not seek to exploit market power.
  - Policies that produce strategic spillovers (terms-of-trade manipulation) merit multilateral scrutiny.
  - Examples: outflow restrictions by large creditor countries raising world interest rates; quantitative easing by large debtor countries lowering world interest rates.
- Implication 3: Capital flows should be managed in both source and recipient countries.
  - Coordination desirable because spillovers between source and recipient countries are larger than among recipients.
  - Objectives: ensure source countries internalize spillovers from monetary, prudential-regulatory, and other policies; require monetary policy to consider international ramifications; extend regulatory mandates to cover cross-border activities and foster cooperation between regulators.
  - Practical challenge: flows too small to be systemic in source country may still threaten recipient country stability.
- Implication 4: Coordination may be needed to avoid capital control wars across recipient countries.
  - Multilateral efforts could help calibrate controls, accounting for possible deflection of flows to other countries.
  - In practice, implies less intensive controls than if flows concentrated in fewer countries.
  - Distinguish between control war (escalation) and multiple appropriate simultaneous controls in response to common shocks.
- Conclusion and broader perspective:
  - Cross-border capital flows pose macroeconomic and financial-stability risks but also provide benefits (finance for projects, efficient resource allocation, employment).
  - Four implications offer an analytical basis for multilateral institutions like the IMF to identify spillovers and suggest remedies to internalize external effects.
  - Historical perspective: international cooperation on capital flows was advocated by Keynes and White and remains central to the IMF’s purposes.

*Source: _sdn1210.*

### Executive Summary ......................................................................................................

### EXECUTIVE SUMMARY

### Context and purpose
- The financial crisis highlighted profound cross-border financial linkages and highly volatile capital flows, prompting interest in multilateral principles to guide policies in both capital-source and capital-recipient countries.
- This note discusses analytical underpinnings for such principles, extends prior Fund analysis, and draws implications for the desirability of multilateral coordination of country policies.
- The note is not intended to prejudge the IMF’s ongoing work to articulate a potential institutional view on capital flows, nor to guide bilateral or multilateral IMF surveillance.

### Core analytical framework and assumptions
- Begins from a highly stylized, neoclassical-free world, introducing departures one at a time to identify spillovers, externalities, and welfare implications of interventions.
- Countries may be large (able to influence world interest rates) or price-takers.
- Interventions considered include capital controls, prudential regulations, and monetary/fiscal policies insofar as they affect the volume or price of cross-border capital flows.
- Distinctions:
  - Capital controls discriminate by residency.
  - Prudential regulations do not discriminate by residency.
  - Measures may be temporary (cyclical) or structural (e.g., closure to certain flows).

### Key findings (theory)
- Terms-of-trade manipulation is theoretically possible: large creditors or debtors could restrict supply/demand of capital to affect world interest rates and gain terms-of-trade advantages.
- Domestic distortions (e.g., excessive foreign borrowing, learning-by-doing externalities, financial-stability risks) can justify national-level interventions that reduce risky inflows:
  - Example: domestic borrowers may not internalize collateral-constraint externalities, leading to excessive and excessively risky foreign borrowing; a national tax on foreign borrowing can be welfare-improving.
  - Example: learning-by-doing externalities in exportable production can motivate sterilized FX intervention to sustain an undervalued currency; inflow controls may be used to maintain that exchange rate (Box 1; Box 2 shows reserve-accumulation equivalence).
- A capital control by a country with a domestic distortion reduces that country’s inflows (or raises its current account surplus), lowering world demand for capital and depressing world interest rates; this is a pecuniary externality and, under some conditions, globally efficient (Box 3).

### Multilateral implications and spillovers
- Three potential multilateral concerns:
  1. Capital controls may vitiate external adjustment (e.g., inflow controls sustaining an undervalued currency).
  2. Recipient-country controls can deflect inflows onto other recipients, exacerbating their inflow problems.
  3. Source-country policies (including monetary policy and prudential regulation) that increase the volume or riskiness of outflows may worsen recipient-country problems.
- Spillovers do not automatically imply policy should not be pursued: if policies are justified from a national standpoint (reducing domestic distortions), they may be appropriate even if they generate cross-border spillovers.
- However, if policies in one country exacerbate distortions in other countries and it is costly for those countries to respond, multilateral coordination of unilateral policies is likely to be beneficial.
- Coordination may require:
  - Borrowers to reduce inflow controls, or
  - Lenders to partially internalize risks from excessively large or risky outflows — the latter is politically and practically thornier.

### Empirical evidence (surveyed)
- Capital-deflection evidence between recipient countries is scant: most studies find no, quantitatively small, or ambiguous deflection effects.
- More compelling evidence exists on source-country policies: advanced-economy interest rates are a key determinant of capital flows to EMEs and particularly of inflow surges.
- There is evidence that pervasive capital account regulations can help sustain undervalued exchange rates over long periods, raising multilateral issues.

### Policy implications and recommendations
- Red flags and areas for multilateral attention:
  - Capital account regulations that appear unjustified from a macroeconomic or financial-stability standpoint but aim to gain unfair competitive advantage.
  - Policies that deflect flows excessively across countries.
  - Policies that transmit risk from source to recipient countries.
- Practical coordination directions (broad, judgment-based rather than prescriptive rules):
  - Use inflow controls less intensively once spillovers are internalized, particularly during generalized surges to multiple countries.
  - Source-country monetary and prudential policies should give appropriate weight to spillovers on global flows and on macroeconomic and financial-stability risks facing recipient countries.
  - Supervisory policies in source countries should internalize risks imposed on foreign countries by domestic financial institutions.
  - Multilateral coordination might focus on internalizing spillovers rather than on rigid capital-account rules; any rules are likely to be “soft” and involve application of judgment.
- Where controls are used to underpin beggar-thy-neighbor external sector strategies or terms-of-trade manipulation, multilateral coordination and red flags are plainly warranted.
- Recognition that coordination between recipient and source countries may be required for globally efficient outcomes, but it is less obviously in the interest of source countries and therefore more difficult to achieve.

### Organizational note
- The note’s sections:
  - Section II: pure theory of capital controls (with further analytic results in Korinek, 2012).
  - Section III: survey of empirical literature.
  - Section IV: discussion of implications.
  - Section V: conclusion.
- Boxes in the note include specific analytical illustrations, e.g., Box 1 (Learning-By-Doing Externalities in Exportable Production), Box 2 (Equivalence of Capital Controls and Reserve Accumulation), and Box 3 (Global Efficiency of Correcting Domestic Externalities).

*Source: _sdn1210 - Executive Summary.*

### Box 1. Learning-By-Doing Externalities in Exportable Production

### Box 1. Learning-By-Doing Externalities in Exportable Production

### Model setup and production technology
- Two-period, traded/non-traded good economy with international borrowing/lending at rate (1+r).
- Learning-by-doing externality: higher production of traded goods raises future productivity in tradable sector.
- Traded goods sector (labor only):  
  - Period 1: yT1 = f(l1)  
  - Period 2: yT2 = A(·) f(l2)  
  - f'(·) > 0, f''(·) < 0 (standard neoclassical production function).  
  - A'(·) ≥ 0; A''(·) < 0.
- Non-traded goods sector (no externality):  
  - yN1 = g(l1); yN2 = g(1−l)  
  - g'(·) > 0, g''(·) < 0.

### Decentralized competitive equilibrium versus national planner
- Competitive equilibrium (takes externality as given) sets marginal rate of transformation equal to the real exchange rate p:  
  - '(f(l1)) / '(1) = p    (equation (1) as presented)
- National social planner (internalizes positive externality) allocates more labor to traded goods:  
  - [1'() + ( ) / (1)]Ayf l r f g l p = ... (equation (2) as presented)
  - (Retained exact functional relationships and notation as in source.)
- Optimal government subsidy to reproduce planner’s allocation:  
  - s* = '() + ( ) / (1) Ay f l r > 0    (equation (3) as presented)
- Effect: producers face more depreciated real exchange rate (given by 1/ps) while consumers face p.

### Capital controls and sterilized intervention (alternative to subsidy)
- Production subsidy is first-best but requires budgetary resources and may be hard to implement.
- Alternative: impose controls on capital inflows and support an “undervalued” exchange rate via sterilized intervention that absorbs an amount R of inflows.
- Cost of sterilized intervention of amount R when domestic interest rate is i and world rate is r: (i−r)R (presented as ().irR− in source notation).
- Necessity of limits on inflows (capital controls) because sterilization cost otherwise infinite.
- Welfare impact near R = 0: representative citizen’s indirect utility is increasing in R at R=0:  
  - U_R'|0 = β[... expression ...] > 0 (equation (4) as presented).  
  - At R=0, i = r so first term is zero; second term positive because intervention depreciates real exchange rate and reallocates labor to traded sector.
- Shortcoming: controls-cum-intervention distort consumption decision, induce excessive saving, and produce an excessive current account surplus — thus not optimal (compared with production subsidy).

### International spillovers and pecuniary externalities
- If other borrowers face excessive foreign borrowing, one country's imposition of capital controls lowers world interest rates and deflects capital, exacerbating inflow problems elsewhere.
- Such spillovers are pecuniary externalities. From an efficiency standpoint, if all excessive borrowers act, lower world interest rates benefit borrowers and hurt creditors without harming global efficiency.
- Caveat: when exportables are differentiated or learning-by-doing is tied to exportables production, all countries can expand exports, but not all can run larger external surpluses; hence unilateral controls-cum-undervaluation can be unilaterally and multilaterally inefficient.

### Key findings and policy implications (bulleted)
- Finding: Learning-by-doing in tradable production creates a positive production externality that private firms do not internalize, leading to underproduction of tradables in the decentralized equilibrium.
- Policy implication: The first-best intervention is a production subsidy targeted to traded goods production; the optimal subsidy s* is strictly positive (equation (3)).
- Alternative policy: Capital controls plus sterilized foreign exchange intervention can raise tradables output and be welfare-enhancing versus no intervention, but are second-best because they distort consumption (excess saving) and generate excessive current account surplus.
- International implication: Unilateral undervaluation (via controls and sterilized intervention) is worse than a production subsidy from the perspective of other countries because it undermines others’ attempts to promote tradables and shifts intertemporal terms of trade against creditor countries.
- Practical constraint: Subsidies require budgetary resources and taxes on nontraded sectors may be infeasible when those sectors are largely informal.
- Risk: If the externality does not exist, or infant industries fail to mature, politically motivated interventions (subsidies or undervaluation) can reduce social welfare.

*Source: Box 1, _sdn1210 - Box 1. Learning-By-Doing Externalities in Exportable Production*

### Box 5. Gains from Coordination under Costly Capital Controls

### Box 5. Gains from Coordination under Costly Capital Controls

### Calibration and equilibrium scenarios
- Two-region model: a borrowing region and a lending region; relative size assumed to be three-to-one.
- Laissez faire equilibrium:
  - Borrowing region imports capital in the amount of 3.1 percent of GDP.
  - World interest rate: 4.1 percent.
- Assumptions on externalities and costs:
  - Externality in the borrowing region: i ξ = 6 percent.
  - Resource cost of imposing capital controls takes the functional form 2().= (as reported in source).
  - Parameter α set to 0.05.
- Uncoordinated (Nash) outcome among national planners:
  - Planners in the borrowing region optimally correct only two-thirds of the externality, imposing a 4 percent inflow tax.
  - Lenders impose no controls (no externalities among lenders).
  - Resulting equilibrium:
    - Capital flows in the borrowing region decline to 2.3 percent of GDP.
    - World interest rate falls to 3.1 percent per year.
    - Welfare effects: borrowers gain the equivalent of a 0.048 percent increase in consumption; lenders lose .009 percent due to the lower interest rate.
    - If borrowers compensated lenders for their loss, borrowers would still enjoy a welfare gain of 0.020 percent.

### Key quantitative changes reported (selected figures from Table 1 and text)
- Laissez faire:
  - Borrower capital control: 0%
  - Borrower CA/GDP: -3.09%
  - Interest rate: 4.12%
  - Δwelfare LF: 0
- Nash equilibrium (uncoordinated national planners):
  - Borrower capital control: 4.04%
  - Borrower CA/GDP: -2.32%
  - Interest rate: 3.09%
  - Δwelfare LF: borrower 0.0479%, lender -.0088%
  - Δwelfare Nash: 0.0017% (net global welfare gain reported under coordination; see below for context)
- Full coordination:
  - Borrower inflow tax reduced from 4 percent to 3.1 percent.
  - Lending region imposes a control of 0.9 percent on outflows.
  - Sum of controls is almost unchanged from Nash, but aggregate cost () falls due to convexity.
  - Net global welfare gain from coordination: equivalent to a .0017 percent increase in consumption of the borrowing country.
- Coordination with restricted lender (global planner cannot impose controls in lending region):
  - Borrowers impose controls smaller than the unilaterally optimal Nash controls but larger than under full coordination.
  - World interest rate and welfare marginally increase compared to the Nash equilibrium, but gains are smaller than with lender participation.
  - The global planner lowers borrowers’ controls because doing so raises the world interest rate and discourages borrowing without imposing the deadweight loss ().
  - For Pareto improvement in this scenario, compensatory transfers from lenders to borrowers are needed.

### Welfare and efficiency observations
- Going from laissez faire to the uncoordinated Nash equilibrium yields gains for borrowers that are an order of magnitude larger than the additional gains from moving from Nash to fully coordinated policies.
- Under full coordination, the convex cost of controls implies it is more efficient to split regulation between borrowers (inflow taxes) and lenders (outflow controls) because sharing the burden reduces aggregate distortionary costs.
- Coordination with restricted lender yields smaller welfare gains than full coordination; transfers may be required for Pareto improvements.

### Coordination among borrowers
- Borrower-country coordination is necessary because:
  - When countries fail to coordinate, each may impose controls that are too high from its own perspective.
  - Ignoring repercussions on other borrowers and their reactions leads countries to overestimate the benefit of unilateral controls.
  - At the end of a capital-control war, borrowing countries would find controls too high and flows too low relative to the coordinated optimum: each borrower would be better off jointly agreeing to lower controls from the uncoordinated equilibrium levels.

### Source-country (creditor) policies and rationale for lender participation
- If the cost of capital controls is convex, global efficiency requires splitting the burden between borrowing countries (inflow controls) and lending countries (outflow controls).
- Obstacles to lender participation:
  - Capital-exporting countries do not directly benefit from moving to the coordinated equilibrium because in the uncoordinated equilibrium they incur no costs.
- Reasons lenders may still restrict outflows:
  - Desire to contribute to global financial stability (public-good motive).
  - Crises in recipient countries cause losses for financial institutions in source countries; reducing outflow risk can be a form of prudential regulation.
  - Source countries may gain terms-of-trade improvements on their lending by reducing capital supply; incidental monopoly power can help debtor countries reduce over-borrowing.
- Efficiency implication: when source countries have incentives to restrict outflows, global efficiency requires equalizing the marginal cost of inflow controls in recipient countries to the marginal cost of outflow controls in source countries; convex costs pin down the optimal split of controls.

### Empirical survey: volume, composition, and spillovers
- Volume and composition of inflows:
  - Most evidence suggests capital controls have little effect on overall flows for countries with relatively open capital accounts.
  - Effectiveness depends on how extensive controls are, administrative/institutional capacity to enforce them, and incentives for investors to circumvent them.
  - Identification problems (e.g., countries facing large inflows are more likely to impose controls) complicate empirical inference.
  - Empirical studies find more consistent effects on the composition of flows (e.g., short-term debt) than on aggregate volume, suggesting compositional shifts are more credibly affected by controls.
  - Literature reviews and meta-studies (Magud et al. (2011), Habermeier et al. (2011), Ostry et al. (2010, 2011)) suggest capital account policies have traction on monetary policy independence, the exchange rate path, and compositional effects.
- Diversion and cross-border spillovers:
  - Direct evidence of cross-border effects from capital account policies is mixed.
  - Forbes et al. (2011) finds small average effects on flows to other countries from Brazilian capital controls: some countries saw increases in inflows, others decreases.
  - IMF (2011) using different methodology/data reaches similar mixed conclusions.

*Source: Box 5. Gains from Coordination under Costly Capital Controls (extracted content).*

### conclusion, that capital account policies have on occasion led to increases or decreases in

### _sdn1210 - conclusion, that capital account policies have on occasion led to increases or decreases in flows to other countries.

### Empirical findings on capital account policies and spillovers
- Empirical evidence on the impact of capital account policies is mixed, and generally stronger in terms of finding an effect on the composition of flows than on the aggregate level.
- Studies find capital account policies have on occasion led to increases or decreases in flows to other countries; explanations include:
  - search for yield (which pushes capital to countries with no controls);
  - amplified risks of the imposition of controls in other countries (deflecting capital toward countries deemed less likely to impose controls), which may lead total outflows from source countries to diminish.
- The deflection-away result implies that a capital control war between capital-receiving countries is unlikely to occur; where countries facing excessive inflows are more likely to want to impose controls, the deflection may be efficient.
- Advanced economy interest rates (or other proxies for monetary policy) are among the most important determinants of capital flows to EMEs.9
- Spillovers from a range of policies (monetary, financial, fiscal, regulatory, and structural) operate in part through cross-border financial flows and can have welfare implications when resulting flows exacerbate pre-existing distortions.
- Spillovers from macroprudential policies are evident where domestic prudential or regulatory deficiencies amplify risky cross-border capital flows (IMF, 2011).

### Key implications (overview)
- Global welfare may be enhanced by some multilateral coordination of country policies with the goal of internalizing spillovers from macro-economic, prudential, and capital control measures.
- Multilateral institutions such as the IMF could provide a forum for identifying spillovers and encouraging countries to internalize these external effects.
- Fund staff are in the process of crystallizing their views with the intention of proposing an institutional position; the present paper draws four implications from the analysis.

### Implication 1: Capital controls and related measures should not substitute for warranted external adjustment
- Capital controls—and policies that mimic their effects—should not act as a substitute for external adjustment when such adjustment is warranted.
- Discretion on their use is likely to be circumscribed, particularly when the currency is undervalued from a multilateral perspective.
- Example: Country A uses capital controls cum intervention to pursue an undervalued exchange rate with no obvious domestic distortion—this should be proscribed.
- Exceptions:
  - Measures to safeguard financial stability when inflow surges exacerbate financial-stability concerns (controls may be considered if another tool would not get the job done).
  - Genuine “learning-by-doing” externalities, but the bar for justification is much higher.
- There is a presumption that undervaluation combined with inflow controls should raise red flags in most cases.

### Implication 2: Countries should not seek to exploit market power
- Policies that produce strategic spillovers (e.g., terms of trade manipulation) merit multilateral scrutiny even if their primary purpose is not strategic.
- Examples of policies with cross-border effects:
  - Restrictions on outflows by large creditor countries that tend to raise world interest rates;
  - Quantitative easing by large debtor countries that tend to lower world interest rates.10
- The pros and cons of such measures should be evaluated from a multilateral perspective.

### Implication 3: Capital flows should be managed in both source and recipient countries
- Coordination between recipient and source countries is desirable because:
  - Spillovers between source and recipient countries are much larger than those among recipient countries.
  - Source countries may reduce risks of losses on foreign lending as crises in recipient countries become less frequent.
  - Terms of trade gains for international creditor source countries may offset some costs of curtailing outflows.
- Coordination objectives:
  - Ensure source countries internalize spillovers from monetary, prudential-regulatory, and other policies.
  - Require monetary policy to consider international ramifications for capital flows.
  - Extend regulatory mandates to cover activities of financial institutions that contribute to instability elsewhere and foster cooperation between regulators in recipient and source countries.
- Practical challenge: flows that are too small to pose systemic risks in the source country may nevertheless threaten recipient country financial stability.11

### Implication 4: Coordination may be needed to avoid capital control wars across recipient countries
- Multilateral efforts could help ensure that countries imposing or intensifying controls calibrate them appropriately, taking account of possible deflection of flows to other countries.12
- In practice, this implies less intensive controls than would occur if flows were concentrated in fewer countries.
- Distinguish between:
  - A capital control war: escalation of controls in response to deflected flows; and
  - Multiple countries imposing controls simultaneously as appropriate responses to excessive inflows (not necessarily a war).
- Policy advice should help recipient countries avoid escalation that would be against their own interests while recognizing legitimate simultaneous responses to common shocks.

### Conclusion and broader perspective
- Cross-border capital flows can cause extreme swings with macroeconomic and financial-stability risks, but also provide benefits (finance for productive projects, infrastructure, efficient allocation of resources, employment).
- The analysis suggests four implications:
  - Measures that restrict flows should not substitute for necessary external adjustment.
  - Measures by systemically important countries that distort the world interest rate merit multilateral scrutiny.
  - Recipient countries imposing inflow controls should be mindful of deflection to others and exercise restraint during general surges.
  - Greater coordination with systemically important source countries is needed so they take account of policy spillovers and supervisory responsibilities for cross-border activities.
- These implications are not hard and fast multilateral “rules of the road” but provide an analytical basis for multilateral institutions like the IMF to identify spillovers and suggest remedies to internalize external effects.
- Historical perspective: international cooperation to manage capital flows at both ends was advocated by Keynes and White and remains central to the IMF’s purposes.

*Source: _sdn1210 - conclusion, that capital account policies have on occasion led to increases or decreases in flows to other countries.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/sdn/2012/_sdn1210.pdf_
