## _wp15226 - Conclusions

## Source details

**Canonical URL:** [_wp15226 - Conclusions](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp15226.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp15226.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp15226.pdf.json)

---

### Macroeconomic effects of capital inflows: asset composition matters
- Extending asset choice to include both bonds (policy-rate-like) and “non-bonds” (stocks, bank deposits, etc.) that are imperfect substitutes reconciles standard Mundell-Fleming predictions with policymakers’ experience.
- For a given policy rate:
  - Bond inflows lead to an exchange rate appreciation and are contractionary.
  - Non-bond inflows lead to an exchange rate appreciation and a decrease in the rate of return on non-bonds; depending on which effect dominates, such flows may be expansionary.
- Mechanism: inflows into non-bonds can lower the return on these assets (effectively lowering the cost of financial intermediation), which can stimulate domestic demand and offset the negative external-demand effects of appreciation.

### Policy implications: tool effectiveness depends on inflow type
- Sterilized foreign exchange (FX) intervention:
  - If conducted through bonds (the usual case), sterilized FX intervention can fully offset bond inflows, leaving both the exchange rate and interest rates unchanged by taking the opposite position to foreign investors.
  - When used in response to non-bond inflows, sterilized FX intervention can prevent exchange rate appreciation but will cause a larger decrease in the rate of return on non-bonds.
- Capital controls:
  - If capital controls primarily target bond inflows, non-bond inflows will dominate, producing an appreciation and a decrease in non-bond returns.
  - If capital controls primarily target non-bond inflows, bond inflows will dominate, producing an appreciation, no change in interest rates, and thus a decrease in output.
- Policy-rate adjustments are blunt instruments relative to tools that can differentiate by asset type; tools that discriminate between bond and non-bond inflows (targeted sterilized intervention or differentiated capital controls) can be preferable.

### Choice of policy mix depends on welfare trade-offs and financial stability tools
- The “right” mix of FX intervention, capital controls, and policy rate changes depends crucially on whether lower interest rates on non-bonds (or higher credit growth) are desirable or likely to produce excessive credit booms and subsequent busts.
- If a country seeks higher output and possesses macroprudential tools to prevent excessive credit growth, non-bond flows are more attractive than bond flows because:
  - Both types of inflows cause appreciation, but
  - Non-bond flows reduce the return on non-bonds, while bond flows do not and tend to produce a larger appreciation.
- Therefore, policy tools that target one type of inflow (sterilized intervention in response to bond flows but not non-bond flows, or capital controls that differentiate between bond and non-bond flows) are preferable to nondiscriminating tools such as changes in the policy rate.

### Model highlights and key comparative-statics (selected analytical results preserved)
- Under simplifying assumptions R_B = R* = 1 and 1/e E+ = 1:
  - Rate on non-bonds: R_N = (1/11)(1 + 6 B_s − N_s).
  - Exchange rate: E = (1/11)(1 + 3 B_s + 6 N_s).
- With sterilized intervention size X:
  - R_N = (1/11)(1 + 6 B_s − N_s − X).
  - E = (1/11)(1 + 3 B_s + 6 N_s + 3 X).
  - To offset bond inflows (s_B > 0, s_N = 0): choose X = s_B → E = 1 and R_N = 1.
  - To offset non-bond inflows (s_N > 0, s_B = 0): choose X = (1/2) s_N → E = 1 and R_N = 1 − (1/4) s_N.
- Capital controls extremes:
  - If F_B = 0 (no foreign purchases of domestic bonds): R_N = 1 − (1/5) s_N and E = 1 + (1/5) s_N.
  - If non-bond purchases blocked: E = 1 + (2/3) s_B.
- Policy-rate trade-offs:
  - Using the policy rate to stabilize E versus stabilizing R_N produces different R_N and E outcomes; e.g., to keep E constant with non-bond inflows implies a larger decline in R_N (−1/3 under policy-rate adjustment versus −1/4 under sterilized FX intervention in the model).

### Empirical evidence: identification, sample, and main findings
- Identification strategy:
  - Use global flows to all emerging markets (excluding country i) as instruments for country-specific flows, interacted with country dummies (GBF and GNBF).
  - Sample: 19 major emerging market countries; Period: 2000 onwards, annual data.
  - Main regression for GDP growth ΔX_it includes bond flows/GDP, non-bond flows/GDP, partner growth X*_it, change in terms of trade, lagged GDP growth, country and time dummies.
- Key empirical results (IV estimates, numbers preserved):
  - An increase in exogenous non-bond flows of 1 percent of GDP increases GDP growth by 0.31 percentage points (0.312*** with [0.072] standard error).
  - Disaggregation of non-bond flows:
    - FDI Flows/GDP: 0.242** [0.103].
    - Portfolio equity Flows/GDP: 0.467*** [0.147].
    - Other Flows/GDP: 0.315*** [0.093].
  - Effects on private credit/GDP:
    - Other Flows: 0.642*** [0.224] — an increase of 1 percent of GDP in “other flows” (bank flows account for about half on average) leads to a 0.6 percentage point increase in credit to GDP.
    - FDI: −0.718** [0.291] (possible substitution away from bank-intermediated financing).
    - Portfolio equity: 1.103 [0.977] (large but not statistically significant in that specification).
- Policy controls (reserve flows/GDP and policy rate) included and instrumented (VIX and US 3-month T-bill rate interacted with country dummies):
  - Reserve Flows/GDP: −0.037 [0.058] (Col 4); −0.583*** [0.224] (Col 5).
  - Policy Rate: −0.223*** [0.070] (Col 4); −0.029 [0.257] (Col 5).
  - Authors note FX intervention coefficient in credit regression is unexpectedly negative and significant and has no ready interpretation.

- Selected table summary (coefficients exactly as reported):
  - Bonds Flows/GDP: −0.002 [0.124] (Col 1); 0.032 [0.108] (Col 2); 0.206 [0.279] (Col 3); −0.028 [0.098] (Col 4); 0.341 [0.295] (Col 5).
  - Non-Bond Flows/GDP: 0.312*** [0.072] (Col 1).
  - FDI Flows/GDP: 0.242** [0.103] (Col 2); −0.718** [0.291] (Col 3); 0.259*** [0.089] (Col 4); −0.667*** [0.248] (Col 5).
  - Port. Equity Flows/GDP: 0.467*** [0.147] (Col 2); 1.103 [0.977] (Col 3); 0.376** [0.153] (Col 4); 1.445 [0.928] (Col 5).
  - Other Flows/GDP: 0.315*** [0.093] (Col 2); 0.642*** [0.224] (Col 3); 0.278*** [0.077] (Col 4); 0.921*** [0.217] (Col 5).
  - Lagged dependent variable, partner growth, change in TOT, country/time fixed effects included.
  - R-squared: 0.522, 0.519, 0.13, 0.556, 0.142 (Cols 1–5 respectively).
  - N = 267, 267, 263, 256, 252; Countries = 19 in all columns.

### Synthesis and policy guidance (conclusions preserved)
- Theoretical and empirical synthesis:
  - For a given policy rate:
    - Bond inflows lead to appreciation and are contractionary.
    - Non-bond inflows lead to appreciation and reduce the cost of borrowing; net effect can be expansionary.
  - Empirical evidence supports the distinction:
    - Exogenous bond flows appear to have on average small negative effects on output.
    - Exogenous non-bond flows appear to have a positive effect on output.
    - Non-bond flows (excluding FDI) have a stronger positive effect on bank credit than bond flows.
- Policy implications:
  - FX intervention limits appreciation but increases inflows and amplifies financial-system effects.
  - Targeted capital controls can eliminate particular flow channels but magnify the effects of remaining flows.
  - Policy-rate adjustments face the policy dilemma: stabilizing E vs stabilizing R_N; using policy rate to hold R_N constant requires higher policy rate and larger FX intervention to offset increased inflows.
  - Combinations of instruments can offset effects on both E and R_N, but trade-offs remain (e.g., higher policy rate can directly harm output; capital controls avoid that direct effect).

*Source: _wp15226 - Conclusions (PDF content supplied).*

### Conclusions ............................................................................................................

### _wp15226 - Conclusions

### Macroeconomic effects of capital inflows: asset composition matters
- Extending asset choice to include both bonds (policy-rate-like) and “non-bonds” (stocks, bank deposits, etc.) that are imperfect substitutes reconciles standard Mundell-Fleming predictions with policymakers’ experience.
- For a given policy rate:
  - Bond inflows lead to an exchange rate appreciation and are contractionary.
  - Non-bond inflows lead to an exchange rate appreciation and a decrease in the rate of return on non-bonds; depending on which effect dominates, such flows may be expansionary.
- Mechanism: inflows into non-bonds can lower the return on these assets (effectively lowering the cost of financial intermediation), which can stimulate domestic demand and offset the negative external-demand effects of appreciation.

### Policy implications: tool effectiveness depends on inflow type
- Sterilized foreign exchange (FX) intervention:
  - If conducted through bonds (the usual case), sterilized FX intervention can fully offset bond inflows, leaving both the exchange rate and interest rates unchanged by taking the opposite position to foreign investors.
  - When used in response to non-bond inflows, sterilized FX intervention can prevent exchange rate appreciation but will cause a larger decrease in the rate of return on non-bonds.
- Capital controls:
  - If capital controls primarily target bond inflows, non-bond inflows will dominate, producing an appreciation and a decrease in non-bond returns.
  - If capital controls primarily target non-bond inflows, bond inflows will dominate, producing an appreciation, no change in interest rates, and thus a decrease in output.
- Policy-rate adjustments are blunt instruments relative to tools that can differentiate by asset type; tools that discriminate between bond and non-bond inflows (targeted sterilized intervention or differentiated capital controls) can be preferable.

### Choice of policy mix depends on welfare trade-offs and financial stability tools
- The “right” mix of FX intervention, capital controls, and policy rate changes depends crucially on whether lower interest rates on non-bonds (or higher credit growth) are desirable or likely to produce excessive credit booms and subsequent busts.
- If a country seeks higher output and possesses macroprudential tools to prevent excessive credit growth, non-bond flows are more attractive than bond flows because:
  - Both types of inflows cause appreciation, but
  - Non-bond flows reduce the return on non-bonds, while bond flows do not and tend to produce a larger appreciation.
- Therefore, policy tools that target one type of inflow (sterilized intervention in response to bond flows but not non-bond flows, or capital controls that differentiate between bond and non-bond flows) are preferable to nondiscriminating tools such as changes in the policy rate.

### Paper structure (as presented)
- Section 1: presents a simple version of the model.
- Section 2: extends the model to examine the role of sterilized FX intervention and capital controls.

*Conclusions from _wp15226 - Conclusions............................................................................................................*

### Section 3 looks at the empirical evidence in the light of the model. It confirms its main prediction,

### _wp15226 - Section 3 looks at the empirical evidence in the light of the model. It confirms its main prediction,

### II. A Portfolio Model — setup and demand for assets
- Model ingredients required:
  - Money (so the central bank can set the policy rate).
  - Two domestic assets: domestic bonds and domestic “non-bonds”; these two assets must be imperfect substitutes.
  - One foreign asset: foreign bonds; imperfect substitute for both domestic bonds and domestic non-bonds.
- Definitions and notation:
  - Domestic investor demands: D_M, D_B, D_N and D_B*.
  - Rates of return: R_B (domestic bonds), R_N (domestic non-bonds), R* (foreign bonds in foreign currency).
  - Exchange rate: E (increase in E is an appreciation of the domestic currency); expected exchange rate next period 1/e E+.
  - Wealth: W; initial domestic holdings D_M, D_B, D_N, D_B*; W = D_M + D_B + D_N + D_B*.
- Money demand:
  - Money demand depends only on the interest rate on bonds (policy rate): 0 1() DB MR α α = − .
- Domestic investors’ demands for bonds, non-bonds, and foreign bonds (proportional to wealth net of money demand) depend on relative returns and satisfy 1 a b c = .
- Foreign investors:
  - Do not hold domestic money; choose between foreign bonds and domestic bonds or domestic non-bonds.
  - Foreign demand shifts captured by B_s and N_s, the sources of capital flows.
  - Initial foreign holdings: F_B and F_N.
- Central bank:
  - Issues money against domestic bonds; chooses M and central bank holdings of domestic bonds C_B^B (and later allowed to hold foreign bonds).

### 2.2. Equilibrium conditions and main comparative static results
- Equilibrium conditions: domestic asset market clearing and capital account balance. With monetary policy setting R_B, the model solves for R_N and 1/e E+ (here taken equal to one for convenience in steps).
- Key analytical solutions (under simplifying assumptions R_B = R* = 1 and 1/e E+ = 1):
  - Rate on non-bonds: R_N = (1/11)(1 + 6 B_s − N_s)  (equation (3) as presented: 11 1 66 NBN Rss = + − — preserved linear form in text).
  - Exchange rate: E = (1/11)(1 + 3 B_s + 6 N_s)  (equation (4) preserved form: 11 1 36 BN Ess = + +).
  - Inflows: sum of bond and non-bond inflows: (equation (5) preserved text form) 11 22 FFFFBN BBNN ss β β − + − = + .
- Main implications for given policy rate:
  - An increase in bond inflows:
    - Leads to appreciation and an increase in R_N.
    - Contractionary effects likely (appreciation and higher R_N).
  - An increase in non-bond inflows:
    - Leads to appreciation and a decrease in R_N.
    - Net effect may be contractionary or expansionary:
      - In emerging markets with "primitive" financial systems, decrease in R_N may dominate → credit boom and output increase despite appreciation.
      - In advanced economies, appreciation may dominate → contractionary (example invoked: Switzerland).

### III. FX intervention, capital controls, and the policy rate — mechanisms and normative implications

- 3.1. Sterilized intervention
  - Central bank can buy foreign bonds and sterilize by selling domestic bonds.
  - Notation: M, C_B^B (domestic), C_B^{B*} (foreign); define X ≡ (C_B^{B*} / β) for normalized sterilized intervention size.
  - Modified equilibrium with sterilized intervention:
    - Non-bonds equilibrium unchanged (equation (6) preserved).
    - Capital account condition includes X (equation (7) preserved).
    - Solutions with X:
      - R_N = (1/11)(1 + 6 B_s − N_s − X)  (equation (8) preserved form: 111 1 666 NBN RssX = + − −).
      - E = (1/11)(1 + 3 B_s + 6 N_s + 3 X)  (equation (9) preserved form: 111 1 363 BN EssX = + + −).
      - Inflows: (equation (10) preserved) () 2 FFFFBN BBNN ss β − + − = + + X .
  - Two illustrative cases to keep E constant:
    - To offset bond inflows (s_B > 0, s_N = 0): choose X = s_B → implies E = 1 and R_N = 1. Sterilized intervention cancels effects of bond inflows on E and R_N.
    - To offset non-bond inflows (s_N > 0, s_B = 0): choose X = (1/2) s_N → implies E = 1 and R_N = 1 − (1/4) s_N. Sterilized intervention prevents appreciation but amplifies the decline in R_N (larger decrease than without intervention) and increases inflows.
  - Main characteristic: FX intervention limits appreciation but increases inflows and thus amplifies effects on the financial system.

- 3.2. Capital controls
  - If capital controls fully eliminate foreign purchases of domestic bonds (F_B = 0):
    - Foreigners choose only between foreign bonds and domestic non-bonds.
    - Solutions:
      - R_N = 1 − (1/5) s_N  (equation (11) preserved: 1 1 5 NN Rs = −).
      - E = 1 + (1/5) s_N  (equation (12) preserved: 1 1 5 N Es = +).
    - Capital controls eliminate bond-flow effects but increase the effects of non-bond flows on E and R_N relative to no controls (absence of bond flows removes dampening channel).
  - If capital controls fully eliminate purchases of domestic non-bonds:
    - Exchange rate: E = 1 + (2/3) s_B  (preserved as 2 1 3 B Es = +).
    - Controls on non-bond inflows magnify the effect of bond inflows on the exchange rate.
  - Summary: targeted capital controls can isolate the economy from particular flows; controls on one category magnify effects of the other category.

- 3.3. Policy rate
  - Reintroducing B_R (policy rate) and solving:
    - R_N = (1/11)(1 + 6 B_s + 6 N_s + 6 B_R ???) — the exact displayed equations in the text are:
      - 11 66 NBBN RRss = + + −  (equation (13) preserved as presented).
      - 11 36 BBN ERss = + +   (equation (14) preserved as presented).
    - Inflows do not depend on B_R in equilibrium: an increase in B_R raises R_N and E in ways that net out on flows.
  - Two policy uses in presence of non-bond flows:
    - To keep E constant: decrease B_R so 1 − 1/6 BN Rs = − (text: 1 1/6 BN Rs = −); then R_N = 1 − 1/3 N s.
    - To keep R_N constant: increase B_R so 1 + 1/6 BN Rs = + (text: 1 1/6 BN Rs = +); then E = 1 + 1/3 N s.
  - Policy dilemma: stabilize E (accept larger decline in R_N) or stabilize R_N (accept larger appreciation). Both policies leave inflows unchanged in equilibrium.
  - Comparison FX intervention vs policy rate to keep E constant under non-bond inflows:
    - Decline in R_N is larger when using policy rate: −1/3 under policy rate vs −1/4 under FX intervention.

- 3.4. Instrument choice remarks
  - Policy choice depends on distortions and objectives:
    - If decrease in R_N is associated with credit growth, central bank may want to limit decline in R_N (avoid excessive credit growth).
    - If appreciation risks “Dutch disease”, central bank may want to limit appreciation.
    - Nominal rigidities imply exchange rate appreciation and decrease in R_N may have opposing effects on aggregate demand.
  - Comparative preferences (illustrated by Figures 1 and 2 described in text):
    - For bond inflows: FX intervention or capital controls preferred to policy rate if the goal is to avoid appreciation with limited decline in R_N.
    - For non-bond inflows: to maintain exchange rate stability with minimum movement in R_N, capital controls best, FX intervention second, policy rate worst.
  - Combination of instruments can in principle fully offset inflow effects on both E and R_N; trade-offs remain (e.g., raising policy rate to keep R_N constant increases inflows and necessitates larger FX intervention).

### IV. Some empirical evidence — strategy and main findings
- Research question: non-bond flows, by decreasing the cost of credit for a given policy rate, are more likely than bond flows to increase output.
- Empirical challenges:
  - Need to focus on exogenous flows (use instruments).
  - Policy responses to inflows (policy rate, FX intervention) can cancel or reverse effects — policies must be controlled for and instrumented.
- Identification strategy:
  - Use global flows to all emerging market countries (excluding country i) as instruments for country-specific flows, interacted with country dummies to allow heterogeneous transmission (GBF and GNBF).
  - Sample: 19 major emerging market countries (Brazil, Chile, Colombia, Czech Republic, Hungary, India, Indonesia, Israel, Korea, Malaysia, Mexico, Peru, Philippines, Poland, Romania, Russia, Thailand, Turkey, South Africa).
  - Period: 2000 onwards, annual data.
  - Main regression for GDP growth ΔX_it includes: bond flows/GDP, non-bond flows/GDP, partner growth X*_it, change in terms of trade TOTΔ, lagged GDP growth, country and time dummies. Flows measured from IMF Financial Flows Analytics (FFA) database; outliers outside 1st–99th percentile dropped in baseline.
- Main empirical findings (first-stage IV results and extensions):
  - Column 1 (IV): Bond flows effect on GDP growth: negative and insignificant. Non-bond flows: positive and significant.
    - Quantitative estimate: An increase in exogenous non-bond flows of 1 percent of GDP increases GDP growth by 0.31 percentage points (0.312*** with [0.072] standard error in table).
  - Disaggregation of non-bond flows (column 2):
    - FDI flows/GDP: 0.242** [0.103].
    - Portfolio equity flows/GDP: 0.467*** [0.147].
    - Other flows/GDP: 0.315*** [0.093].
    - All three non-bond subcomponents positive, significant, and roughly similar magnitudes; bond flows remain negative and insignificant.
  - Effects on credit (column 3 and extensions):
    - Change in private credit/GDP:
      - Bond flows: positive but insignificant (0.206 [0.279]).
      - FDI flows: large negative and significant (−0.718** [0.291]); plausible reason: FDI may substitute for bank-intermediated financing.
      - Portfolio equity: large positive but statistically insignificant (1.103 [0.977] in column 3).
      - Other flows: positive, large, and statistically significant (0.642*** [0.224]); an increase of 1 percent of GDP in “other flows” (bank flows account for about half on average) leads to a 0.6 percentage point increase in credit to GDP.
  - Second-step: control for policy responses (FX intervention measured as reserve flows/GDP, and policy rate), instrumenting them using VIX and US 3-month T-bill rate interacted with country dummies (weaker instruments):
    - Column 4 (GDP IV with policy controls): Coefficients on bond and non-bond flows similar to column 2. FX intervention coefficient small and insignificant; policy rate has expected negative sign and is significant.
    - Column 5 (Change in credit with policy controls): Bond flows small/insignificant; FDI negative and significant; portfolio and other flows strongly positive. Policy rate negative but insignificant. Unexpectedly, FX intervention coefficient is negative and significant (contrary to model prediction that FX intervention should increase credit via larger non-bond inflows); authors note no ready interpretation.
- Table 1 (selected coefficients exactly as reported):
  - Bonds Flows/GDP: −0.002 [0.124] (Col 1), 0.032 [0.108] (Col 2), 0.206 [0.279] (Col 3), −0.028 [0.098] (Col 4), 0.341 [0.295] (Col 5).
  - Non-Bond Flows/GDP: 0.312*** [0.072] (Col 1).
  - FDI Flows/GDP: 0.242** [0.103] (Col 2); −0.718** [0.291] (Col 3); 0.259*** [0.089] (Col 4); −0.667*** [0.248] (Col 5).
  - Port. Equity Flows/GDP: 0.467*** [0.147] (Col 2); 1.103 [0.977] (Col 3); 0.376** [0.153] (Col 4); 1.445 [0.928] (Col 5).
  - Other Flows/GDP: 0.315*** [0.093] (Col 2); 0.642*** [0.224] (Col 3); 0.278*** [0.077] (Col 4); 0.921*** [0.217] (Col 5).
  - Reserve Flows/GDP: −0.037 [0.058] (Col 4); −0.583*** [0.224] (Col 5).
  - Policy Rate: −0.223*** [0.070] (Col 4); −0.029 [0.257] (Col 5).
  - Lagged dependent variable, partner growth, change in TOT, country/time fixed effects, R-squared and sample sizes as reported in table: R-squared 0.522, 0.519, 0.13, 0.556, 0.142; N = 267, 267, 263, 256, 252; Countries = 19 in all columns.

### V. Conclusions — synthesis and policy implications
- Theoretical and empirical synthesis:
  - For a given policy rate:
    - Bond inflows lead to appreciation and are contractionary.
    - Non-bond inflows lead to appreciation and reduce the cost of borrowing; net effect can be expansionary.
  - Empirical evidence supports the distinction:
    - Exogenous bond flows appear to have on average small negative effects on output.
    - Exogenous non-bond flows appear to have a positive effect on output.
    - Non-bond flows (excluding FDI) have a stronger positive effect on bank credit than bond flows.
- Policy implications:
  - Different instruments should be adapted to the nature of inflows:
    - FX intervention limits appreciation but increases inflows and amplifies financial-system effects.
    - Targeted capital controls can eliminate particular flow channels but magnify the effects of remaining flows.
    - Policy-rate adjustments face the policy dilemma: stabilizing E vs stabilizing R_N; using policy rate to hold R_N constant requires higher policy rate and larger FX intervention to offset increased inflows.
  - Combination of instruments can be used to offset effects of inflows on both E and R_N, but trade-offs remain (e.g., higher policy rate can directly harm output; capital controls avoid that direct effect).

### Data and measurement notes (preserved definitions)
- Capital flows source: IMF Financial Flow Analytics (FFA) database, drawing on Balance of Payments Statistics Database, BPM6.
- Flow definitions:
  - Bond Flows: Financial account, Portfolio investment, Debt securities.
  - Equity Flows: Financial account, Portfolio investment, Equity and investment fund shares.
  - FDI Flows: Financial account, Direct investment.
  - Other Flows: Financial account, Other investment, Non-official sector.
  - Non-Bond Flows = FDI Flows + Equity Flows + Other Flows.
  - Net Reserve Flows: Financial Account, Reserve Assets.
- Flows measured in gross terms (net purchases or sales of domestic assets by foreign residents), in current dollars and scaled by Nominal GDP in dollars.
- Additional data sources: GDP Volume (IFS), Partner Growth (WEO), Terms of Trade (WEO), Domestic Credit to Private Sector (WDI and IFS), Policy Interest Rate (IFS).
- Financial derivatives not included in bond or non-bond flow variables.

*Source: _wp15226 - Section 3 looks at the empirical evidence in the light of the model. It confirms its main prediction (PDF content supplied).*

### REFERENCES

### REFERENCES

### Key academic works on exchange rates and capital flows
- Blanchard, O., 2007, “Current Account Deficits in Rich Countries,” IMF Staff Papers, vol. 54(2), pp. 191–219.
- Branson, W., H. Halttunen, and P. Masson, 1977, “Exchange Rates in the Short-Run: The Dollar-Deutschemark Rate” European Economic Review, Vol. 10, pp. 303–324.
- Caballero, R. J., and G. Lorenzoni, 2014, “Persistent appreciations and overshooting: A normative analysis.” IMF Economic Review, 62, pp. 1–47.
- Dornbusch, R., 1976, “Expectations and Exchange Rate Dynamics” Journal of Political Economy, Vol. 84, No. 6, pp. 1161–1176.
- Fleming, J. M., 1962, “Domestic financial policies under fixed and floating exchange rates.” IMF Staff Papers Vol. 9, pp. 369–379.
- Gabaix, X., and M. Maggiori, 2015, “International Liquidity and Exchange Rate Dynamics,” Quarterly Journal of Economics, 130 (3), pp. 1369–1420.
- Kouri, P., 1976. “Capital Flows and the Dynamics of the Exchange Rate.” Seminar Paper 67. Stockholm: Institute for International Economic Studies.
- Mundell, R., 1963, “Capital mobility and stabilization policy under fixed and flexible exchange rates,” Canadian Journal of Economics and Political Science, Vol. 29, pp. 475–85
- Rey, H., 2013, “Dilemma not Trilemma: The Global Financial Cycle and Monetary Policy Independence,” Jackson Hole 2013.
- Krugman, P., 2014, “Currency Regimes, Capital Flows, and Crises.” IMF Economic Review, forthcoming.

### Research on capital flows, controls, and policy instruments
- Blanchard, O., G. Adler, and I. de Carvalho Filho, 2015, “Can Foreign Exchange Intervention Stem Exchange Rate Pressures from Global Capital Flow Shocks?” NBER working paper 21427, July 2015.
- Farhi, E., and I. Werning, 2012, “Dealing with the Trilemma: Optimal Capital Controls with Fixed Exchange Rates.” NBER Working Paper w18199.
- Farhi, E. and I. Werning, 2013, “A Theory of Macroprudential Policies in the Presence of Nominal Rigidities.” NBER Working Paper w18199.
- Friedman, B., 1978, “Crowding Out or Crowding In? The Economic Consequences of Financing Government Deficits.” Brookings Papers on Economic Activity, Vol. 3, pp. 593–654.
- Ghosh, A. R., M. S. Qureshi, J. Kim, J. Zalduendo, 2014, “Surges” Journal of International Economics, 92, pp. 266–285.
- Ostry, J. D., A. R. Ghosh, K. Habermeier, M. Chamon, M. S. Qureshi, and D. Reinhart, 2010, “Capital Inflows: The Role of Controls.” IMF Staff Position Note No. 10/04.
- Ostry, J. D., A. R. Ghosh, M. Chamon and M. S. Qureshi, 2011, “Capital Controls: When and Why?” IMF Economic Review 59, pp. 562–580.
- Ostry, J. D., A. R. Ghosh, M. Chamon, and M. S. Qureshi, 2012, “Tools for managing financial-stability risks from capital inflows.” Journal of International Economics, 88(2), pp. 407–421.
- Ostry, J. D., A. R. Ghosh, M. Chamon, 2012, “Two Targets, Two Instruments: Monetary and Exchange Rate Policies in Emerging Market Economies,” IMF Staff Discussion Note No. 12/1.
- Reinhart, C. and V. Reinhart, 2009, “Capital Flow bonanzas: an encompassing view of the past and present.” NBER Macroeconomics Annual, University of Chicago Press.
- International Monetary Fund, 2012, “The Liberalization and Management of Capital Flows: An Institutional View.” Available at: http://www.imf.org/external/np/pp/eng/2012/111412.pdf.

*Source: _wp15226 - REFERENCES*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp15226.pdf_