## _wp12275

## Source details

**Canonical URL:** [_wp12275](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12275.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12275.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12275.pdf.json)

---

### I. Introduction — study scope and contribution
- Focus: short- to medium-term effects of liberalizing capital flows on macroeconomic performance and risks to financial stability.
- Indicators analyzed: economic growth, inflation, capital inflows, outflows and net flows, equity returns, and bank capital adequacy ratios (equity returns and bank capital adequacy ratios used as proxies for potential risks to financial stability).
- Sample: 37 countries that liberalized capital flows during 1995–2010.
- Approach: dynamic panel data specifications with country fixed effects and a lagged dependent variable to capture partial adjustment toward the steady state; relatively short time dimension considered as the transition period.
- China application: econometric method draws on experiences of other countries that liberalized in the past 15 years; results for China should be interpreted with caution because of China’s high saving and investment rates, long-existing low real deposit rates, already very high investment rate, and relative openness to Greenfield FDI.
- Key qualitative conclusions:
  - Liberalization encourages financial integration (strong positive links with gross capital flows).
  - Impact on net flows is not statistically significant.
  - Liberalization is associated with higher GDP per capita growth and lower inflation.
  - Liberalization is associated with higher equity returns and lower bank capital adequacy ratios, indicating potential risks to financial stability.
  - Sizes of effects for China depend upon the extent and pace of liberalization.

### II. Measuring capital flow restrictiveness — indices and data
- Two de jure measures constructed from AREAER for 185 countries over 1995–2010; higher values indicate more controls.
- Narrow de jure restrictiveness index:
  - Based on 21 categories of restrictions (equity, bond, money market and collective investment instruments, financial credit, and direct investment by direction).
  - Distinguishes inflows, outflows, and net flows.
  - Correlation with the Schindler index: 0.92 (period of availability).
- Broader de jure index:
  - Average of binary indicators across 62 categories (includes foreign exchange and domestic currency accounts of residents and nonresidents, regulatory measures related to the financial sector, repatriation and surrender requirements).
  - Range between zero and 1; higher values represent more restricted cross-border capital flows.
  - Correlation narrow vs. broad: 0.92 for 185 countries and 0.90 for the empirical sample.
  - Correlation with Chinn-Ito (2008): 0.78 (narrow) and 0.86 (broad) for the period of availability.
- De facto openness measure: gross stocks of foreign assets and liabilities as a ratio to GDP (FDI, equity, portfolio investment, external debt, official reserves); stock data up to 2007 from Lane and Milesi-Ferretti (2007) and 2008–2010 updates using IMF Balance of Payments flows (not controlled for valuation).
- Structural break: AREAER data have a structural break in 1995 (pre-1995 AREAER used a binary dummy; post-1995 more structured subcategory information).

### III. Recent trends in capital flow liberalization — stylized facts and sample selection
- Distributional changes (mid-1990s to 2010):
  - 2010: 17 out of 185 countries fully open (de jure restrictiveness index equal to zero) and 31 fully closed (index equal to one).
  - Mid-1990s: 19 out of 185 countries fully open and 39 fully closed.
  - About 50 percent of countries had an index lower than 0.5 in 2010, compared to about 43 percent in the mid-1990s.
- Geography: most largely closed countries are in Asia and Africa.
- Few significant reversals of liberalization over the past 15 years (examples cited include Argentina and Iceland).
- Sample selection for empirical analysis:
  - Countries retained if they liberalized by at least 0.1 point in the de jure index between 1995 and 2010.
  - For a given country, only years following the start of liberalization where the index declines by at least 0.01 point are retained.
  - Resulting sample: about 37 countries.
  - For those countries: mean liberalization between 1995 and 2010 was 0.4; maximum 0.83; minimum 0.1.
- Sequencing evidence:
  - More than half of the 37 countries increased exchange rate flexibility prior to capital flow liberalization (IMF de facto exchange rate classification).
  - One-fourth initiated interest rate liberalization before capital flow liberalization (interest rate liberalization data available only until 2005).

### IV. Empirical strategy — model, estimation issues, and threshold analysis
- Dependent variables in panel regressions: capital outflows, inflows, net flows, real GDP per capita growth, inflation, equity returns, and bank capital adequacy ratios.
- General specification: dynamic panel with country fixed effects and lagged dependent variable; estimated coefficients represent short-run effects; long-run effects derived by dividing coefficients by (1-β1).
- Endogeneity and estimation:
  - Addressed with System GMM (Arellano and Bover, 1995; Blundell and Bond, 1998) treating all right-hand variables as endogenous.
  - Panel unit root tests indicate series used are stationary.
- Threshold analysis (Kose and others 2009 method):
  - Composite threshold index made from four normalized indicators: financial development (market capitalization/GDP or private credit/GDP), quality of bureaucracy and corruption, fiscal balance ratio, and trade openness (exports+imports/GDP).
  - Median of composite index used to create "above threshold" and "below threshold" subsamples.

### V. Empirical results — main findings from panel estimations
- Broad findings (sample of 37 countries, Tables 2–7):
  - Liberalization associated with higher gross capital inflows and outflows.
  - Effect on net flows not statistically significant.
  - Liberalization associated with higher GDP per capita growth.
    - Numeric result reported: 0.14 percentage points increase in real GDP per capita growth (Table 2).
    - Additional numeric statements in document: “1.7 percentage points.” (presented in text as a listed specific value).
  - Liberalization associated with lower inflation.
    - Reported mapping: a 0.1 point decline in the index implies about a 0.7 percentage points decrease in inflation (Table 3).
  - Liberalization associated with higher equity returns.
    - Reported mapping: a 0.1 point decline in the index implies about a 2.9 percentage points increase in equity returns (Table 4).
    - Example projection: “the equity return in 2016 may increase by 17 percentage points.”
  - Liberalization associated with lower bank capital adequacy ratios.
    - Reported mapping: a 0.1 point decline in the index implies a decrease in the capital adequacy ratio of about 0.3 percentage points (Table 5).
    - Example projection: “capital adequacy ratio in 2016 may decrease by 0.7 percentage points.”
  - Capital flows (gross) — reported elasticities/coefficients in tables show large and statistically significant associations:
    - Inflows: a 0.1 point decline in the index implies a 1.2 percentage points increase in inflows (Table 6).
    - Outflows: a 0.1 point decline in the index implies a 0.8 percentage points increase in outflows (Table 7).
- Additional reported empirical relationships (selected coefficients as presented):
  - Growth regression: capital flow restrictiveness index coefficients include -1.845**, -1.526**, -1.500**, -1.695***, -1.385*, -1.387**, -1.382**, -1.980**, 0.608, -1.426**, -5.210***.
  - Inflation regression: capital flow restrictiveness index coefficients include 7.970**, 7.277**, 6.830**, 6.816**, 17.294*, 1.028, 7.851***; broader index 23.897**.
  - Equity returns: equity flow restrictiveness index coefficients include -49.939***, -48.057***, -47.806***, -36.612***, -29.125*, -29.428*, -29.002*, -15.741**, -5.584, -43.792***.
  - Bank capital adequacy: capital flow restrictiveness index coefficients include 4.207**, 2.378*, 2.568**, 2.513*, 3.458***, 3.175, 3.854**; broader index 4.383***.
  - Gross inflows (Table 6): coefficients include -17.458***, -12.653***, -12.823***, -12.808***, -12.687***, -12.938***, -12.923***, -9.802***, -21.679***, -0.185; broader index -9.824**.
  - Gross outflows (Table 7): coefficients include 12.773***, 7.319***, 7.443***, 7.379***, 7.494***, 7.920***, 8.332***, 7.733***, 17.349**, 0.003; broader index 7.828***.
  - Real credit to private sector growth positively associated with GDP per capita growth (examples: 0.050**, 0.047***, 0.054***, 0.055***, 0.071***, 0.016, 0.125***, 0.088***).
  - VIX (change) strongly negatively associated with equity returns (around -4.143*** to -4.078***).
  - NPLs negatively associated with capital adequacy ratios (examples: -0.177*, -0.194**, -0.108*, -0.119**, -0.296***, -0.036).

### Identification, index construction, and robustness
- Endogeneity control:
  - Internal instruments based on lagged values used; lagged levels as instruments in difference regressions and lagged differences as instruments in level regressions.
  - Number of lags restricted to a maximum of two.
  - Assumption of weak exogeneity: explanatory variables assumed uncorrelated with future realizations of the error terms.
- Index construction:
  - Normalization: Index = (actual value - minimum value) / (maximum value - minimum value).
  - Sub-indices aggregated by arithmetic mean.
- Robustness checks:
  - Results robust to System GMM, clustered standard errors (country level), alternative restrictiveness indices (broad index), pooled WLS weighting by countries’ GDP in U.S. dollar, and specifications aimed at larger countries (columns in Table 2).

### Threshold analysis: countries above vs. below threshold
- Countries “above threshold” (composite index above median):
  - Main findings generally confirmed.
  - Coefficients of liberalization larger than in full samples, indicating larger benefits.
  - Interpretation: countries with higher financial and institutional development derive greater benefits from capital flow liberalization.
- Countries “below threshold”:
  - Coefficients of liberalization not significant in most regressions, including the growth regression, indicating limited benefits for those countries.

### Simulation of capital flow liberalization in China — method and scenarios
- Two scenarios for 2012–2016:
  - Scenario without capital flow liberalization (equation 2).
  - Scenario with capital flow liberalization (equation 3).
- Calibration and key assumptions:
  - Constant terms (ˆjα) set to fit actual 2011 data; coefficients (ˆjβ) taken from panel regressions.
  - Capital flows in China assumed to be gradually liberalized to reach the 2010 average level of G-20 in 10 years (2010 G-20 average for index: 0.5; this equals the 185-country average for 2010).
  - Liberalization in scenario assumed partial but significant; specific Appendix II assumptions include:
    - Capital flow restrictiveness index: decline by 0.05 each year over 2012–2016 in the liberalization scenario; no change in no-liberalization scenario.
    - VIX: decline by 2 points annually from 2013 onward in both scenarios.
    - REER/NEER: misalignment disappears over the Medium Term in liberalization scenario; NEER calculated based on REER and inflation differentials.
    - Country risk: average of difference in country risk during 1995–2010 in both scenarios.
    - Credit: WEO projections in both scenarios.
    - Interest rates and U.S. interest rate: fixed at 2011 level in both scenarios.
    - NPL ratio: increase of one percentage point every year from 2011 onward in both scenarios.

### Simulation results for China (differences liberalization vs. no-liberalization)
- Capital flows (gross):
  - Inflows may increase by 3.3 percentage points in 2016.
  - Outflows may increase by 2.1 percentage points in 2016.
  - Representing a net increase of US$380 billion (inflows) and US$240 billion (outflows) in 2016, respectively.
  - Conclusion: liberalization would have a substantial impact on gross capital flows; effect on net flows depends on measures and sequencing.
- Growth and inflation:
  - The GDP growth rate in 2016 may increase by 0.4 percentage points under liberalization.
  - Simulation indicates lower inflation rates under liberalization (specific 2016 decrease not provided in excerpt).
- Financial stability indicators:
  - Example projections in text: equity return in 2016 may increase by 17 percentage points; capital adequacy ratio in 2016 may decrease by 0.7 percentage points.
- Assumed adverse credit trajectory: NPL ratio assumed to increase by one percentage point every year from 2011 to 2016.

### Conclusions and policy implications
- Using panel data on 37 countries over the past 15 years, the study finds:
  - Liberalization of capital flows is strongly linked to larger gross capital flows and financial integration.
  - Liberalization is associated with higher GDP per capita growth and lower inflation.
  - Liberalization is associated with higher equity returns and lower bank capital adequacy ratios, implying potential risks to financial stability and a need for stronger prudential regulation.
  - Countries meeting threshold conditions (financial/institutional development) are better able to reap growth and stability benefits of financial globalization.
  - Gross capital flows can contribute to global financial instability and warrant close scrutiny beyond net flows and composition.
- For China and other EMEs with remaining controls:
  - Gradual, well-sequenced liberalization is advised to harness benefits while mitigating risks.
  - Quantitative effects on China depend on pace and sequencing; cross-country extrapolations provide insights but have caveats given China’s unique features.

*Source: IMF staff estimates and text from the content unit "_wp12275 - References".*

### References .............................................................................................................

### References

### I. Introduction — study scope and contribution
- Focus: short- to medium-term effects of liberalizing capital flows on macroeconomic performance and risks to financial stability.
- Indicators analyzed: economic growth, inflation, capital inflows, outflows and net flows, equity returns, and bank capital adequacy ratios (equity returns and bank capital adequacy ratios used as proxies for potential risks to financial stability).
- Sample: 37 countries that liberalized capital flows during 1995–2010.
- Approach: dynamic panel data specifications to capture partial adjustment toward the steady state; relatively short time dimension considered as the transition period.
- China analysis: econometric method draws on experiences of other countries that liberalized in the past 15 years; results for China should be interpreted with caution due to country-specific features (high saving and investment rates, long-existing low real deposit rates, already very high investment rate, relative openness to Greenfield FDI).
- Key qualitative conclusions (from the panel estimates and China application):
  - Liberalization encourages financial integration (strong positive links with gross capital flows).
  - Impact on net flows is not statistically significant.
  - Liberalization is associated with higher GDP per capita growth and lower inflation.
  - Evidence of association with higher equity returns and lower bank capital adequacy ratios, indicating potential risks to financial stability.
  - Sizes of effects for China depend upon the extent and pace of liberalization.

### II. Measuring capital flow restrictiveness — indices and data
- Two new de jure measures of capital flow liberalization constructed from the IMF’s Annual Report on Exchange Arrangements and Exchange Restrictions (AREAER); indices computed for 185 countries over 1995–2010. Higher values indicate more controls.
- First de jure restrictiveness index:
  - Similar to the Schindler index (Schindler 2009).
  - Based on 21 categories of restrictions (including equity, bond, money market and collective investment instruments, financial credit, and direct investment by direction).
  - Distinguishes between inflows, outflows, and net flows.
  - Correlation with the Schindler index for the period of availability: 0.92.
- Second de jure (broader) index:
  - Average of binary indicators of restrictiveness in 62 categories of capital transactions (includes foreign exchange and domestic currency accounts of residents and nonresidents, regulatory measures related to the financial sector, repatriation and surrender requirements).
  - Range between zero and 1; higher values represent more restricted cross-border capital flows.
  - Correlation between narrow index and broader index: 0.92 for the 185 countries and 0.90 for the sample used in the empirical part of the paper.
  - Correlation with the Chinn-Ito (2008) index: 0.78 for the narrower index and 0.86 for the broader index (for the period of availability of the Chinn-Ito index).
- De facto openness measure also used (gross stocks of foreign assets and liabilities as a ratio to GDP, including FDI, equity, portfolio investment, external debt, and official reserves); stock data up to 2007 from Lane and Milesi-Ferretti (2007) and 2008–2010 updates using IMF Balance of Payments flows (not controlled for valuation).
- Structural break note: AREAER data have a structural break in 1995 (pre-1995 AREAER used a binary dummy; post-1995 more structured subcategory information).

### III. Recent trends in capital flow liberalization — stylized facts and sample selection
- Distributional changes between mid-1990s and 2010:
  - 2010: 17 out of 185 countries fully open (de jure restrictiveness index equal to zero) and 31 fully closed (index equal to one).
  - Mid-1990s: 19 out of 185 countries fully open and 39 fully closed.
  - About 50 percent of countries had an index lower than 0.5 in 2010, compared to about 43 percent in the mid-1990s.
- Geographic distribution: most largely closed countries are in Asia and Africa.
- Few significant reversals of liberalization over the past 15 years (examples cited include Argentina and Iceland).
- Sample inclusion criteria for empirical analysis:
  - Countries retained if they liberalized by at least 0.1 point according to the de jure index between 1995 and 2010.
  - For a given country, only years following the start of liberalization where the index declines by at least 0.01 point are retained.
  - Resulting sample: about 37 countries (Table 1).
  - For those countries: mean of capital flow liberalization between 1995 and 2010 was 0.4; maximum was 0.83; minimum was 0.1.
- Sequencing evidence:
  - More than half of the 37 countries had increased their exchange rate flexibility prior to capital flow liberalization (using IMF’s de facto exchange rate classification).
  - A fourth had initiated interest rate liberalization before liberalizing capital flows (data for interest rate liberalization available only until 2005 from Abiad and others (2008)).

### IV. Empirical strategy — model, estimation issues, and threshold analysis
- Dependent variables studied in panel regressions: capital outflows, inflows, net flows, real GDP per capita growth, inflation, equity returns, and bank capital adequacy ratios.
- General specification (dynamic panel with country fixed effects and lagged dependent variable):
  - The approach includes country fixed effects to account for unobserved heterogeneity; the lagged dependent variable captures inertia and implies estimated coefficients represent short-run effects; long-run effects can be derived by dividing each coefficient by one minus coefficient of the lagged dependent variable (1-β1).
  - Equation form presented in paper for panel specification (subscripts i for country, t for year, j for specific equation).
- Endogeneity and estimation:
  - Two econometric issues: potential endogeneity of some independent variables (simultaneity or reverse causality) and the endogeneity of the lagged dependent variable in fixed-effect estimators.
  - Robustness check: System Generalized Method of Moments (System GMM) estimators used with all right-hand variables treated as endogenous (Arellano and Bover, 1995; and Blundell and Bond, 1998).
  - Panel unit root tests indicate series used in empirical models are stationary (results available upon request).
- Threshold analysis (following Kose and others 2009):
  - Full sample separated into two sub-samples using a composite threshold composed of four normalized and averaged indicators: financial development (ratio of market capitalization to GDP or private sector credit to GDP), quality of bureaucracy and corruption, ratio of fiscal balances to GDP, and trade openness (ratio of exports and imports to GDP).
  - Median of composite index used as threshold: countries above median are “above threshold”; below median are “below threshold”.

### V. Empirical results — main findings from panel estimations
- Broad empirical findings reported in the paper (based on Tables 2 to 7 and the sample of countries that liberalized over the past 15 years):
  - Liberalization associated with higher gross capital flows (strong positive links).
  - Impact on net flows not statistically significant.
  - Liberalization associated with higher GDP per capita growth.
    - The coefficients of liberalization are significantly negative (a decline in the index means liberalization of capital flows).
    - Text in source begins to report a quantitative mapping (“The results indicate that a 0.1 point decline in the index implies about a”) but the statement is incomplete in the provided excerpt.
  - Liberalization associated with lower inflation.
  - Liberalization associated with higher equity returns.
  - Liberalization associated with lower bank capital adequacy ratios, suggesting potential risks to financial stability.
- China-specific application:
  - Applying panel coefficients to China qualitatively mirrors econometric results above.
  - Sizes of potential effects for China depend on the extent and pace of liberalization.
  - Caveats reiterated: China’s high saving and investment rates, low real deposit rates, already very high investment rate, and relative openness to Greenfield FDI may make cross-country extrapolations less informative.

*Source: IMF staff estimates and text from the content unit "_wp12275 - References".*

### 0.14 percentage points increase in real GDP per capital growth (Table 2).

### _wp12275 - 0.14 percentage points increase in real GDP per capital growth (Table 2)

### Key empirical findings
- 0.14 percentage points increase in real GDP per capital growth (Table 2).
- Lower inflation rates.
- The coefficients of the liberalization index are significantly positive:
  - A 0.1 point decline in the index implies about 0.7 percentage points decrease in inflation (Table 3).
- Higher equity returns:
  - The coefficients of the equity liberalization index are significantly negative. A decline in the index of 0.1 point implies about 2.9 percentage points increase in equity returns (Table 4).
- Lower bank capital adequacy ratios:
  - A decline in the index of 0.1 point implies a decrease in the capital adequacy ratio of about 0.3 percentage points (Table 5).
- Higher capital inflows and outflows:
  - A 0.1 point decline in the index implies a 1.2 percentage points increase in inflows and a 0.8 percentage points increase in outflows (Table 6 and 7).
  - The effect of liberalization on net flows is not statistically significant.

### Identification and index construction
- Endogeneity control:
  - Internal instruments based on lagged values of the explanatory variables are used.
  - Assumption of weak exogeneity: explanatory variables are assumed uncorrelated with future realizations of the error terms.
  - Lagged levels used as instruments in regressions in difference; lagged differences used as instruments in regressions in level.
  - Number of lags restricted to a maximum of two.
- Liberalization index construction:
  - Each variable normalized as: Index = (actual value - minimum value) / (maximum value - minimum value).
  - Sub-indices aggregated using the arithmetic mean.

### Threshold analysis (countries above vs. below threshold)
- For countries “above threshold”:
  - Main findings in the full sample are generally confirmed.
  - Coefficients of liberalization are larger than in full samples, indicating a larger role of capital flow liberalization in countries “above threshold.”
  - Interpretation: countries that have reached certain thresholds of financial and institutional development benefit the most from capital flow liberalization.
- For countries “below threshold”:
  - Coefficients of liberalization are not significant in most regressions, including in the growth regression, indicating a limited role of liberalization for those countries.

### Robustness checks
- Results robust to alternative estimation approaches and different capital flow liberalization measures:
  - Several panel data specifications including system GMM produced broadly similar results to fixed effects estimator.
  - Similar results when using clustered standard errors (at the country level).
  - Using a broad restrictiveness index of capital flows leads to similar results.
  - Pooled weighted least squares (WLS) estimation, weighting observations by countries’ GDP in U.S. dollar, yields results broadly similar to pooled (unweighted) least squares (LS).
  - Results for real GDP per capita growth presented in the last two columns in Table 2 suggest conclusions hold for larger countries as well.

### Simulation of capital flow liberalization in China (method and scenarios)
- Focus: apply coefficients from panel regressions to China’s variables during 2012–16.
- Two scenarios constructed:
  - Scenario without capital flow liberalization (equation 2).
  - Scenario with capital flow liberalization (equation 3).
- Differences between the two scenarios interpreted as additional effects of capital flow liberalization on China.
- Calibration and assumptions:
  - Constant term for each equation (ˆjα) set to fit actual 2011 data.
  - Coefficients of independent variables (ˆjβ) come from regressions in the previous section.
  - Capital flows in China assumed to be gradually liberalized to reach the 2010 average level of G-20 in 10 years (for 2010 the average for G-20 countries was 0.5; coincidently, this is also the average for the 185 countries).
  - Liberalization is assumed partial but significant.
  - Other assumptions summarized in Appendix II (not reproduced here).

### Simulation results for China (differences between liberalization and no-liberalization scenarios)
- More gross capital inflows and outflows:
  - Inflows may increase by 3.3 percentage points in 2016.
  - Outflows may increase by 2.1 percentage points in 2016.
  - Representing a net increase of US$380 billion (inflows) and US$240 billion (outflows) in 2016, respectively.
  - Conclusion: liberalization would have a substantial impact on gross capital flows; effect on net flows depends on measures and sequencing.
- Higher GDP per capita growth rates compared to no-liberalization scenarios:
  - The GDP growth rate in 2016 may increase by 0.4 percentage points.
- Lower inflation rates:
  - The simulation indicates lower inflation rates under liberalization (specific 2016 decrease not provided in the excerpt).

*Source: _wp12275 - 0.14 percentage points increase in real GDP per capital growth (Table 2).*

### 1.7 percentage points.

### _wp12275 - 1.7 percentage points.

### Conclusions and policy implications
- Using panel data techniques on 37 sample countries over the past 15 years, the paper finds that liberalization of capital flows can partially explain macroeconomic performance and risks to financial stability.
- Liberalization is strongly positively linked with gross capital flows, indicating that liberalization encourages financial integration.
- Liberalization is associated with higher GDP per capita growth and lower inflation.
- Liberalization is associated with higher equity returns and lower bank capital adequacy ratios, indicating potential risks to financial stability and the need for stronger prudential regulation.
- Countries meeting threshold conditions are better able to reap growth and stability benefits of financial globalization.
- Gross capital flows can lead to global financial instability and warrant close scrutiny, beyond the traditional focus on net flows and composition of flows (short-term financing and exchange rate risk).

### Empirical findings — macroeconomic performance and stability risks
- Growth and inflation:
  - Liberalization associated with higher real GDP per capita growth (see Table 2 coefficients: e.g., Capital flow restrictiveness index coefficients include -1.845**, -1.526**, -1.500**, -1.695***, -1.385*, -1.387**, -1.382**, -1.980**, 0.608, -1.426**, -5.210***).
  - Liberalization associated with lower inflation (see Table 3: Capital flow restrictiveness index coefficients include 7.970**, 7.277**, 6.830**, 6.816**, 17.294*, 1.028, 7.851***; broader index 23.897**).
- Equity returns:
  - Liberalization linked to higher equity returns (Table 4: Equity flow restrictiveness index coefficients include -49.939***, -48.057***, -47.806***, -36.612***, -29.125*, -29.428*, -29.002*, -15.741**, -5.584, -43.792***).
  - Example projected change: “the equity return in 2016 may increase by 17 percentage points.”
- Banking sector and capital adequacy:
  - Liberalization associated with lower bank capital adequacy ratios (Table 5: Capital flow restrictiveness index coefficients include 4.207**, 2.378*, 2.568**, 2.513*, 3.458***, 3.175, 3.854**; broader index 4.383***).
  - Example projected change: “capital adequacy ratio in 2016 may decrease by 0.7 percentage points.”
- Capital flows (gross):
  - Liberalization strongly associated with larger capital inflows (Table 6: Capital flow restrictiveness index coefficients include -17.458***, -12.653***, -12.823***, -12.808***, -12.687***, -12.938***, -12.923***, -9.802***, -21.679***, -0.185; broader index -9.824**).
  - Liberalization associated with larger capital outflows (Table 7: Capital flow restrictiveness index coefficients include 12.773***, 7.319***, 7.443***, 7.379***, 7.494***, 7.920***, 8.332***, 7.733***, 17.349**, 0.003; broader index 7.828***).
- Other relevant empirical relationships reported in tables:
  - Real credit to private sector (growth) positively associated with GDP per capita growth (e.g., coefficients 0.050**, 0.047***, 0.054***, 0.055***, 0.071***, 0.016, 0.125***, 0.088***).
  - VIX (change) strongly negatively associated with equity returns (Table 4: VIX change coefficients around -4.143*** to -4.078***).
  - Nonperforming Loans (NPLs) negatively associated with capital adequacy ratios (Table 5: NPLs coefficients include -0.177*, -0.194**, -0.108*, -0.119**, -0.296***, -0.036).

### Policy implications and cautions
- Benefits of liberalizing capital flows: reducing misallocation of resources, risk diversification, deepening and development of financial markets.
- Risks and challenges: increased macroeconomic volatility, vulnerability to crises, complications for macroeconomic management (especially under fixed exchange rate regimes), booms in domestic demand and potential overheating, asset price bubbles, increased banking sector risks—necessitating stronger prudential regulation.
- For China and other EMEs with remaining capital controls, authorities have called for gradual liberalization to harness benefits while mitigating risks; internationalization of the renminbi would benefit from smooth liberalization.
- Quantitative effects on China depend on pace and sequencing; econometric results from the 37-country sample provide insight into possible short- to medium-term effects but may not fully capture China’s unique features.

### Simulation assumptions for China (Appendix II)
- Capital flow restrictiveness index:
  - Scenario without Liberalization: No changes.
  - Scenario with Liberalization: Decline by 0.05 each year over 2012–2016.
- VIX:
  - Both scenarios: Decline by 2 points annually from 2013 onward.
- REER:
  - Scenario without Liberalization: No changes.
  - Scenario with Liberalization: The misalignment, as computed by CGER in 2011 Article IV staff reports, disappears over the Medium Term. Simple average of the results of three approaches (ERER, ES, and MB) is used.
- NEER:
  - Scenario without Liberalization: No changes.
  - Scenario with Liberalization: Calculated based on the REER and inflation differential between China and weighted trading partner during 2011–2016.
- Country risk: Average of difference in country risk during 1995–2010 in both scenarios.
- Credit: WEO projections in both scenarios.
- Interest rates and U.S. interest rate: Fixed at 2011 level in both scenarios.
- NPL ratio:
  - Both scenarios: An increase of one percentage point every year from 2011 onward.
  - Note: “The NPL ratio in China is assumed to increase by one percentage point every year from 2011 to 2016 due to the recent credit expansion, including loans to local government financing platforms. This assumption is also in line with the market estimate.”

### Selected specific numeric projections and illustrative changes mentioned in the text
- “1.7 percentage points.”
- Equity return in 2016 may increase by 17 percentage points.
- Capital adequacy ratio in 2016 may decrease by 0.7 percentage points.
- NPL ratio assumed to increase by one percentage point every year from 2011 to 2016.

*Source: IMF staff estimates as presented in the supplied document.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12275.pdf_