## _afr1401

## Source details

**Canonical URL:** [_afr1401](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/dp/2014/_afr1401.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/dp/2014/_afr1401.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/dp/2014/_afr1401.pdf.json)

---

### Introduction and key summary findings
- Over the last three years, the frontier market economies of sub-Saharan Africa have received growing amounts of portfolio capital flows.
- Since 2010, continued positive macroeconomic performance, coupled with unprecedented accommodative monetary policies in advanced economies, renewed foreign investors’ interest on a much larger scale, resulting in sub-Saharan African frontier markets becoming more integrated with international capital markets.
- The number of sub-Saharan African countries with international credit ratings has increased; a large number of countries issued sovereign bonds—many for the first time—and foreign investors have become active players in some domestic bond and equity markets.
- Compared with FDI, portfolio capital flows and cross-border bank loans tend to be more volatile and more sensitive to changing conditions in global financial markets.
- Recommendation highlights for sub-Saharan frontier markets:
  - Improve data.
  - Enhance macroeconomic and financial policies.
  - Improve capacity to effectively use macroprudential policies.
  - Exercise caution in the use of capital flow management measures.

### Short-term capital flow trends and magnitudes
- Net private flows to sub-Saharan African countries doubled over the 2010–12 period compared with the 2000–07 period; in the case of sub-Saharan African frontier markets, there has been a fivefold increase.
- Over 2010–12, net portfolio flows to sub-Saharan African frontier markets in terms of country GDP outstripped those of emerging and developing economies: about 1.8 percent versus 0.9 percent for emerging and developing countries.
- Aggregate figures and examples:
  - Sub-Saharan African frontier markets surpassed the US$17 billion mark in private portfolio and cross-border bank flows in 2012.
  - Ghana, Nigeria, and Zambia recorded portfolio flows estimated at about 1.9 percent of GDP, 2.7 percent of GDP, and 1.6 percent of GDP, respectively.
  - SSA - Frontier Markets (Table 1): 2000–07 Total 2.0; FDI 3.3; Portfolio 0.1; Other −1.4. 2010–12 Total 2.2; FDI 4.2; Portfolio 1.8; Other −3.9.

### Composition and cross-border bank activity
- Portfolio and cross-border bank flows increased alongside FDI; cross-border bank lending to sub-Saharan African frontier markets also rose, while net deposit positions remained stable as deposits in foreign banks grew steadily.
- Cross-border bank-related flows (Table 2):
  - Cross-Border Bank-Related Flows to SSA (Billions U.S. dollars): 2000–04 Total 50; 2010–12 Total 109; Loans to Nonbanks 39 (for SSA overall).
  - For SSA Frontier Markets: 2000–04 Total 8; 2010–12 Total 36.

### Market reaction and volatility episodes (2013)
- Since June 2013, capital flows in emerging market economies became particularly volatile after the Federal Reserve chairman’s announcement about possible tapering.
  - Outflows from emerging market mutual funds reached US$50 billion (3.9 percent of total) through mid-September 2013.
  - Roughly US$23 billion came from bond funds during the selloff episode.
- Market impact comparisons:
  - Between May 22 and late August, the broader emerging market bond index (EMBIG) declined by 8.5 percent, while the average frontier market bond price declined by 5 percent.
  - Some frontier market currencies (with the exception of the Ghanaian cedi) experienced relatively little pressure compared to some emerging markets, but Ghana and Tanzania experienced widened bond spreads comparable to those of emerging markets.

### Drivers of portfolio flows (Box 1 summary)
- Push factors (outside sub-Saharan Africa):
  - Interest rates in advanced economies
  - Global risk aversion
  - International portfolio diversification
  - Potential growth differential between advanced and sub-Saharan African frontier markets
- Pull factors (inside sub-Saharan Africa):
  - Commodity prices
  - Domestic interest rates
  - Domestic inflation
  - Exchange rate stability
  - Fiscal and external balance sheets
  - Trade openness and other reforms
  - External capital account openness
- Econometric analysis note:
  - The global risk aversion variable (VIX) was negative, significant, and robust across many model specifications; none of the other push or pull factors were statistically significant, likely due to a combination of factors including poor data quality.

### Box 2 — Ghana’s recent experience in portfolio flows (key points)
- Portfolio and FDI flows:
  - Ghana attracted high and stable levels of foreign direct investment of about 8 percent of GDP, underpinned by the discovery and subsequent production of oil in 2011.
  - Oil-related foreign direct investment flows account for two-thirds of total foreign direct investment.
  - Portfolio investments increased in significance to 2.8 percent of GDP in 2012.
  - Portfolio inflows are mostly in medium-term investments because nonresidents are only allowed to purchase bonds with maturity of at least three years.
  - In August 2013, Ghana sold a bond of US$1billion, with 10-year maturity at a yield of 8 percent.
  - Ghana’s first 10-year Eurobond was issued in 2007, which was four times oversubscribed and yielded as little as 4½ percent in April 2013. That bond had a yield of 8½ percent at issuance but in an environment of high international interest rates.
  - About one-third of domestic debt is held by foreigners.
- Sensitivity and liquidity:
  - Frontier markets are very illiquid; when investors need liquidity they may sell the most liquid assets first and retain frontier positions longer.
  - If foreign investors liquidate positions in illiquid frontier markets, the effect could amplify due to insufficient domestic participants to substitute for the foreign investor.
  - More liquid frontier market currencies faced relatively more pressures in the second stage of the selloff (June 24–mid-September) than in the first stage (May 22–June 21).
  - Ghana underperformed other frontier currencies in the period due to rising inflation pressures, deteriorating fiscal balance, and relatively low international reserves.
- Domestic debt, reserves, and macro effects:
  - High domestic interest rates have sustained foreign interest in the medium-term domestic market.
  - Increased participation of foreigners in government debt markets could be an additional source of vulnerability due to rollover risks.
  - Reserves have been kept roughly stable in recent months but fell below estimates of the optimal reserve level in percent of imports in 2012.
  - The average exchange rate depreciated by about 17 percent in the first half of 2012; Bank of Ghana measures in 2012 stabilized the currency but resulted in double-digit real interest rates.
- Policy implications for Ghana:
  - Given Ghana’s much lower reserves, large fiscal and current account deficits, the recommended policy mix in the event of capital outflows is to:
    - allow the currency to adjust, and
    - substantially tighten monetary and fiscal policies to attenuate capital outflow and reduce the need for foreign financing.
  - Managing volatile capital flows requires a combination of:
    - macroeconomic policies as primary adjustment tools,
    - macroprudential policies to limit systemic financial risk, and
    - capital flow measures (CFMs) as temporary measures to buy time for fundamental adjustment (CFMs should not substitute for policy adjustment).
  - Policy choices should account for: (i) exchange rate and competitiveness assessment; (ii) foreign exchange reserve adequacy in terms of import cover and short-term liabilities; (iii) the country’s cyclical position and inflation pressures; (iv) risks for asset bubbles; and (v) balance sheet conditions of banks, corporates, and financial intermediaries.

### Box 5 — Capital flow management measures (CFMs) in frontier markets (types and examples)
- Measures aimed at managing capital inflows:
  - Restrictions on nonresident purchase of government securities:
    - Tanzania: restrictions on all purchase of government securities by nonresidents.
    - Ghana: restricts only purchases of government securities of less than three years.
    - Nigeria and Kenya: apply selective restrictions on nonresident purchases of securities.
  - Maximum total primary issuance of bonds held by nonresidents:
    - Zimbabwe: maximum cap of 35 percent.
  - Maximum total share of national companies held by nonresidents:
    - Kenya: 60 percent.
    - Tanzania: 60 percent.
    - Zimbabwe: 35 percent at primary issuance, and approval required for participation in the secondary market.
  - Maximum on share purchased by individual investors:
    - Tanzania: individual investors may not acquire more than 1 percent of an issue and institutional investors no more than 10 percent.
  - Minimum holding period:
    - Tanzania: minimum holding period of three months for all shares and securities purchased locally by nonresidents.
  - Limits to direct or indirect foreign exchange exposure:
    - Tanzania: banks no longer able to enter into swaps or forward sales with nonresidents; even spot transactions in foreign exchange must reflect an economic interest.
    - Kenya: swaps are restricted to maturities longer than one month.
  - Approval required prior to capital transactions with nonresidents:
    - Mozambique: all transactions with nonresidents require approval by the central bank.
    - Zimbabwe: approval from the Exchange Control to invest new inflows of funds in money market instruments.
- Measures aimed at managing capital outflows:
  - Approval prior to transaction:
    - Senegal: approval by the Ministry of Finance is required for virtually all capital outflows except for amortization of debt and repayments of short-term loans.
  - Request of accounting and/or tax records:
    - Tanzania and Zambia: request presentation of audited accounts and/or compliance with tax obligations for the repatriation of capital and associated income.

### Institutional capacity, macroprudential policy, and implementation constraints
- Financial system complexity and supervisory gaps:
  - Nonbank financial institutions (pension funds, insurance companies) are becoming systemically important and can account for a significant share of the financial system’s total assets.
  - Linkages among financial institutions are increasing; cross-border banking activities are growing rapidly.
  - Supervisory processes have traditionally focused on compliance for individual banks instead of monitoring systemic risks.
  - Supervisory resources, including qualified staff and availability of analytical tools, are limited.
  - Limited high-frequency data on the concentration of risks within the system limit efficient monitoring of systemic vulnerabilities.
- Strengthening supervisory capabilities for macroprudential policy requires:
  - Developing ways to monitor a risk buildup.
  - Setting in place indicators and analytical tools to detect when risks are about to materialize.
  - Using macroprudential tools in a timely manner.
- Risks of applying sophisticated macroprudential measures in limited-capacity settings:
  - Examples include countercyclical capital provisioning or countercyclical limits on banks’ net open foreign exchange positions without timely and high-quality data, which could amplify business cycles and create foreign exchange liquidity crunches that negatively affect bank stability.

### IMF institutional view and practical guidance on CFMs
- Institutional view:
  - CFMs can play a complementary role in supporting macroeconomic adjustment and safeguarding financial stability in certain circumstances.
  - CFMs may be temporarily appropriate if room for adjusting monetary, fiscal, and exchange rate policies is constrained or if balance sheet effects in the financial sector could destabilize.
  - CFMs may buy time where policy adjustments take time to implement.
  - CFMs should be temporary and not substitute for necessary macroeconomic adjustment or financial stability measures.
- Practical considerations for frontier markets:
  - Most sub-Saharan African countries have only partially liberalized capital accounts; remaining restrictions may affect flows even if not designed for that purpose.
  - Costs of introducing new CFMs include damaging effects on future inflows and reputational risks; costs are particularly large for CFMs that limit outflows.
  - For frontier markets with reputations for relatively open capital accounts (for example, Nigeria and Mauritius), the costs of introducing new CFMs may be relatively high.
  - In unsophisticated financial market structures, controls may be more effective than in developed markets, but institutional capacity and information constraints can prevent effective monitoring and enforcement, undermining CFM effectiveness.
  - Any new CFMs should be temporary and scaled back when capital flow pressures abate; CFMs should avoid leading to external payments arrears or sovereign default.
  - In crisis or near crisis situations, temporary CFMs on outflows could be part of a broader policy package to address the underlying cause of the crisis.

### Examples of macroprudential policies (country snapshots)
- Ghana (In 2009):
  - Foreign and local currency deposits are subject to a 9 percent reserve requirement in domestic currency.
  - Reserve balance must be held with the Bank of Ghana.
  - Daily single foreign currency exposure limit was reduced from 15 to 10 percent of the capital base.
- Kenya:
  - Capital requirements are relatively high to prepare banks for overall risks.
  - Households do not borrow in foreign currency, except for a small segment of the mortgage market.
- Mozambique:
  - Limited impact of private capital inflows on overall liquidity conditions has not called for specific macroprudential measures.
- Nigeria:
  - If exposure to a particular industry within a sector is in excess of 20 percent of total credit facilities of a bank the risk weight of the entire portfolio in that industry shall be 150 percent.
- Senegal:
  - No traditional macroprudential regulations have been employed.
- Tanzania:
  - In late 2011 net open position on foreign exchange was reduced.
- Uganda:
  - Ongoing design of contingency plans to deal with excessive inflows through macroprudential means.

### Conclusions and recommended priorities for frontier markets
- Key conclusions:
  - Foreign portfolio flows to sub-Saharan Africa grew considerably in the past three years; sub-Saharan African frontier markets were the main beneficiaries.
  - Most frontier markets have so far been relatively unscathed by recent global financial market turbulence, partly reflecting illiquid financial markets and relatively strong fundamentals.
  - Greater integration with global financial markets implies increasing vulnerability to global financial shocks and possible reversals.
- Priority areas to strengthen:
  - Improving data:
    - Need to improve data quality and timeliness, including balance-of-payments data and coverage of cross-border private capital flows and stocks.
    - Official data on private capital portfolio flows and stocks are, in a number of countries, based on defective surveys with partial coverage or poor quality.
  - Enhancing macroeconomic and financial policies:
    - Managing capital flows is more likely to succeed if supported by sound fiscal, monetary, and exchange rate policies and adequate fiscal and international reserve buffers.
    - Strong fundamentals have helped limit investor punishment during global turbulence.
  - Improving capacity to effectively use macroprudential policies:
    - Macroprudential measures can prevent buildup of systemic risks from volatile capital flows.
    - Effectiveness requires progress in monitoring systemic risk buildup and using tools timely; supervisory resources, qualified staff, high-frequency data, and analytical tools are limited and need strengthening.
  - Improving the toolkit of capital flow management measures:
    - When macroeconomic policy room is limited, CFMs may be considered to temper volatile flows, but effective implementation requires significant institutional improvements.
    - Imposition of new CFMs, especially on outflows, must consider negative effects on future inflows and financial deepening; CFMs on outflows should be a last resort in financial crisis situations.

### Annex — econometric analysis (method and key points)
- Regression setup and data:
  - Regressions reported in Table A.1 are run on an annual panel covering the period 1991–2012 for selected emerging and frontier markets.
  - Dependent variable: ratio of portfolio liabilities (nonresident purchases of domestic assets net of sales) to GDP.
  - Lagged regressors used to minimize endogeneity.
  - Push factors included: U.S. Treasury bill/10-year yields, VIX (global risk).
  - Pull factors included: inflation level, fiscal balance, gross public debt, output growth, change in reserves, stock market development.
  - Random effects used to allow inclusion of a dummy for sub-Saharan African frontier markets.
  - Note: A table note states: "The sample period covers 2001–12 years."
- Key empirical observations:
  - Running regressions for sub-Saharan African frontier markets alone produced no statistically significant relationships, likely because of poor data quality and small sample size.
  - Another set of regressions (Table A.2) focused on sub-Saharan African frontier markets using quarterly data and disaggregating portfolio flows into equity and bond flows; dependent variables based on EPFR data, defined as flows into equity and bond funds of the country divided by the stock at the beginning of the quarter.
  - A crisis dummy equals 1 from 2008:Q4 onward.
  - Quarterly estimation limited by lack of quarterly availability for most pull factors; quarterly output growth available only for three countries.
  - Ghana and Nigeria were selected as a subsample given larger portfolio inflows.

_Italic: Source: _afr1401 - 1.  Capital movements—Africa, Sub-Saharan.    I.  Gulde, Anne-Marie._

### 1.  Capital movements—Africa, Sub-Saharan.    I.  Gulde, Anne-Marie.

### _afr1401 - 1.  Capital movements—Africa, Sub-Saharan.    I.  Gulde, Anne-Marie.

### Introduction and key summary findings
- Over the last three years, the frontier market economies of sub-Saharan Africa have received growing amounts of portfolio capital flows.
- Since 2010, continued positive macroeconomic performance, coupled with unprecedented accommodative monetary policies in advanced economies, renewed foreign investors’ interest on a much larger scale, resulting in sub-Saharan African frontier markets becoming more integrated with international capital markets.
- The number of sub-Saharan African countries with international credit ratings has increased, a large number of countries issued sovereign bonds—many for the first time—and foreign investors have become active players in some domestic bond and equity markets.
- Compared with FDI, portfolio capital flows and cross-border bank loans tend to be more volatile and more sensitive to changing conditions in global financial markets.
- Recommendation highlights for sub-Saharan frontier markets:
  - Improve data.
  - Enhance macroeconomic and financial policies.
  - Improve capacity to effectively use macroprudential policies.
  - Exercise caution in the use of capital flow management measures.

### Short-term capital flow trends and magnitudes
- Net private flows to sub-Saharan African countries doubled over the 2010–12 period compared with the 2000–07 period; in the case of sub-Saharan African frontier markets, there has been a fivefold increase.
- Over 2010–12, net portfolio flows to sub-Saharan African frontier markets in terms of country GDP outstripped those of emerging and developing economies: about 1.8 percent versus 0.9 percent for emerging and developing countries.
- Aggregate figures and examples:
  - Sub-Saharan African frontier markets surpassed the US$17 billion mark in private portfolio and cross-border bank flows in 2012.
  - Ghana, Nigeria, and Zambia recorded portfolio flows estimated at about 1.9 percent of GDP, 2.7 percent of GDP, and 1.6 percent of GDP, respectively.
  - SSA - Frontier Markets (Table 1): 2000–07 Total 2.0; FDI 3.3; Portfolio 0.1; Other −1.4. 2010–12 Total 2.2; FDI 4.2; Portfolio 1.8; Other −3.9.

### Composition and cross-border bank activity
- Portfolio and cross-border bank flows increased alongside FDI; cross-border bank lending to sub-Saharan African frontier markets also rose, while net deposit positions remained stable as deposits in foreign banks grew steadily.
- Cross-border bank-related flows (Table 2):
  - Cross-Border Bank-Related Flows to SSA (Billions U.S. dollars): 2000–04 Total 50; 2010–12 Total 109; Loans to Nonbanks 39 (for SSA overall).
  - For SSA Frontier Markets: 2000–04 Total 8; 2010–12 Total 36.

### Market reaction and volatility episodes (2013)
- Since June 2013, capital flows in emerging market economies became particularly volatile after the Federal Reserve chairman’s announcement about possible tapering.
  - Outflows from emerging market mutual funds reached US$50 billion (3.9 percent of total) through mid-September 2013.
  - Roughly US$23 billion came from bond funds during the selloff episode.
- Market impact comparisons:
  - Between May 22 and late August, the broader emerging market bond index (EMBIG) declined by 8.5 percent, while the average frontier market bond price declined by 5 percent.
  - Some frontier market currencies (with the exception of the Ghanaian cedi) experienced relatively little pressure compared to some emerging markets, but Ghana and Tanzania experienced widened bond spreads comparable to those of emerging markets.

### Drivers of portfolio flows (Box 1)
- Push factors (outside sub-Saharan Africa):
  - Interest rates in advanced economies
  - Global risk aversion
  - International portfolio diversification
  - Potential growth differential between advanced and sub-Saharan African frontier markets
- Pull factors (inside sub-Saharan Africa):
  - Commodity prices
  - Domestic interest rates
  - Domestic inflation
  - Exchange rate stability
  - Fiscal and external balance sheets
  - Trade openness and other reforms
  - External capital account openness
- Econometric analysis note: The global risk aversion variable (VIX) was negative, significant, and robust across many model specifications; none of the other push or pull factors were statistically significant, likely due to a combination of factors including poor data quality.

### Policy responses and implications (summary from Intro)
- Improve data:
  - Strengthen balance-of-payments data, including coverage of cross-border private capital flows and stocks.
  - Article IV reports identified issues such as large net errors and omissions, challenges in data collection on transactions in nonresident securities, unreliable current/capital transfer splits for foreign aid, poor coverage of reinvested earnings, inconsistent debt stock and principal payment data, incomplete financial account recording, and flows and stocks of gross international reserves and net foreign assets often requiring substantial adjustments.
- Enhance macroeconomic and financial policies:
  - Sound fiscal, monetary, and exchange rate policies and adequate fiscal and international reserve buffers improve the ability to manage capital flows.
- Improve capacity for macroprudential policies:
  - Supervisory resources, including qualified staff, availability of high-frequency data, and analytical tools to assess systemic risks are limited and need strengthening.
- Exercise caution with capital flow management measures (CFMs):
  - New CFMs could jeopardize further financial sector deepening and thus should be considered as temporary measures in the context of managing a crisis or near crisis situation.

*Source: _afr1401 - 1.  Capital movements—Africa, Sub-Saharan.    I.  Gulde, Anne-Marie.*

### Box 2 . Ghana’s Recent Experience in Portfolio Flows

### Box 2 . Ghana’s Recent Experience in Portfolio Flows

### Portfolio and foreign direct investment flows
- Ghana attracted high and stable levels of foreign direct investment of about 8 percent of GDP, underpinned by the discovery and subsequent production of oil in 2011.
- Oil-related foreign direct investment flows account for two-thirds of total foreign direct investment.
- Portfolio investments increased in significance to 2.8 percent of GDP in 2012.
- Portfolio inflows are mostly in medium-term investments because nonresidents are only allowed to purchase bonds with maturity of at least three years.
- In August 2013, Ghana sold a bond of US$1billion, with 10-year maturity at a yield of 8 percent.
- Ghana’s first 10-year Eurobond was issued in 2007, which was four times oversubscribed and yielded as little as 4½ percent in April 2013. That bond had a yield of 8½ percent at issuance but in an environment of high international interest rates.
- About one-third of domestic debt is held by foreigners.

### Sensitivity to global shocks, liquidity, and investor behavior
- Portfolio flows into frontier markets can be sensitive to deteriorating fiscal and current account deficit positions and to global push factors that caused emerging market bond turmoil in mid-2013.
- Frontier markets represent a tiny fraction of the overall portfolio of dedicated emerging market investors and are very illiquid; when investors need liquidity they may sell the most liquid assets first and retain frontier positions longer.
- If foreign investors liquidate positions in illiquid frontier markets, the effect could amplify due to insufficient domestic participants to substitute for the foreign investor.
- Where liquidity conditions permitted, investors differentiated among sub-Saharan African frontier markets based on economic fundamentals. The more liquid frontier market currencies faced relatively more pressures in the second stage of the selloff (June 24–mid-September) than in the first stage (May 22–June 21).
- Ghana, facing rising inflation pressures, deteriorating fiscal balance, and relatively low international reserves, underperformed other frontier currencies in the period.

### Domestic debt, reserves, and macroeconomic effects
- High domestic interest rates have sustained foreign interest in the medium-term domestic market.
- The increased participation of foreigners in government debt markets could be an additional source of vulnerability due to rollover risks. The secondary market is rather illiquid, although early redemption or purchase of not-yet-matured three-year and five-year bonds is possible.
- Although reserves have been kept roughly stable in recent months, they have fallen below estimates of the optimal reserve level in percent of imports in 2012.
- The average exchange rate depreciated by about 17 percent in the first half of 2012. Consequently, the Bank of Ghana took a number of measures in 2012 to tighten domestic liquidity. Although these measures were successful in stabilizing the currency, they came at the cost of double-digit real interest rates.

### Policy implications and recommendations for Ghana
- Given Ghana’s much lower reserves, large fiscal and current account deficits, the recommended policy mix in the event of capital outflows is to:
  - allow the currency to adjust, and
  - substantially tighten monetary and fiscal policies to attenuate capital outflow and reduce the need for foreign financing.
- More broadly, managing volatile capital flows requires a combination of:
  - macroeconomic policies as primary adjustment tools (with emphasis on maintaining macroeconomic stability and adequate buffers),
  - macroprudential policies to limit systemic financial risk, and
  - capital flow measures (CFMs) as temporary measures to buy time for fundamental adjustment (CFMs should not substitute for policy adjustment).
- Policy choices should account for: (i) exchange rate and competitiveness assessment; (ii) foreign exchange reserve adequacy in terms of import cover and short-term liabilities; (iii) the country’s cyclical position and inflation pressures; (iv) risks for asset bubbles; and (v) balance sheet conditions of banks, corporates, and financial intermediaries.

_Italic: Source: Box 2 . Ghana’s Recent Experience in Portfolio Flows (extracted content)._

### Box 5 . Capital Flows Management Measures in Sub-Saharan African Frontier Markets

### Box 5 . Capital Flows Management Measures in Sub-Saharan African Frontier Markets

### Types of capital flows management measures in frontier markets
- Measures aimed at managing capital inflows:
  - Restrictions on nonresident purchase of government securities:
    - Tanzania: restrictions on all purchase of government securities by nonresidents.
    - Ghana: restricts only purchases of government securities of less than three years.
    - Nigeria and Kenya: apply selective restrictions on nonresident purchases of securities.
  - Maximum total primary issuance of bonds held by nonresidents:
    - Zimbabwe: maximum cap of 35 percent.
  - Maximum total share of national companies held by nonresidents:
    - Kenya: 60 percent.
    - Tanzania: 60 percent.
    - Zimbabwe: 35 percent at primary issuance, and approval required for participation in the secondary market.
  - Maximum on share purchased by individual investors:
    - Tanzania: individual investors may not acquire more than 1 percent of an issue and institutional investors no more than 10 percent.
  - Minimum holding period:
    - Tanzania: minimum holding period of three months for all shares and securities purchased locally by nonresidents.
  - Limits to direct or indirect foreign exchange exposure:
    - Tanzania: owing to regulations of the Bank of Tanzania (October 2011), banks are no longer able to enter into swaps or forward sales with nonresidents, and even spot transactions in foreign exchange have to reflect an economic interest.
    - Kenya: swaps are restricted to maturities longer than one month.
  - Approval required prior to capital transactions with nonresidents:
    - Mozambique: all transactions with nonresidents require approval by the central bank.
    - Zimbabwe: approval from the Exchange Control to invest new inflows of funds in money market instruments.
- Measures aimed at managing capital outflows:
  - Approval prior to transaction:
    - Senegal: approval by the Ministry of Finance is required for virtually all capital outflows except for amortization of debt and repayments of short-term loans.
  - Request of accounting and/or tax records:
    - Tanzania and Zambia: request presentation of audited accounts and/or compliance with tax obligations for the repatriation of capital and associated income.

### Institutional capacity, macroprudential policy, and implementation constraints
- Financial system complexity and supervisory gaps:
  - Nonbank financial institutions (pension funds, insurance companies) are becoming systemically important and can account for a significant share of the financial system’s total assets.
  - Linkages among financial institutions are increasing; cross-border banking activities are growing rapidly.
  - Supervisory processes have traditionally focused on compliance for individual banks instead of monitoring systemic risks.
  - Supervisory resources, including qualified staff and availability of analytical tools, are limited (Beck, Maimbo, Faye, and Triki, 2011).
  - Limited high-frequency data on the concentration of risks within the system limit efficient monitoring of systemic vulnerabilities.
- Strengthening supervisory capabilities for macroprudential policy requires:
  - Developing ways to monitor a risk buildup.
  - Setting in place indicators and analytical tools to detect when risks are about to materialize.
  - Using macroprudential tools in a timely manner (IMF, 2011b).
- Risks of applying sophisticated macroprudential measures in limited-capacity settings:
  - Example risks: countercyclical capital provisioning or countercyclical limits on banks’ net open foreign exchange positions without timely and high-quality data could amplify business cycles and create foreign exchange liquidity crunches that negatively affect bank stability.

### IMF institutional view and guidance on CFMs
- Institutional view (IMF, 2013b):
  - CFMs can play a complementary role in supporting macroeconomic adjustment and safeguarding financial stability in certain circumstances.
  - CFMs may be temporarily appropriate if room for adjusting monetary, fiscal, and exchange rate policies is constrained or if balance sheet effects in the financial sector could destabilize.
  - CFMs may buy time where policy adjustments take time to implement.
  - CFMs should be temporary and not substitute for necessary macroeconomic adjustment or financial stability measures.
- Practical considerations for frontier markets:
  - Most sub-Saharan African countries have only partially liberalized capital accounts; remaining restrictions may affect flows even if not designed for that purpose.
  - In restrictive capital account frameworks, additional CFMs may be more difficult to implement and could have negative feedback effects on investor confidence.
  - Costs of introducing new CFMs include damaging effects on future inflows and reputational risks; costs are particularly large for CFMs that limit outflows.
  - For frontier markets with reputations for relatively open capital accounts (for example, Nigeria and Mauritius), the costs of introducing new CFMs may be relatively high.
  - In unsophisticated financial market structures, controls may be more effective than in developed markets, but institutional capacity and information constraints can prevent effective monitoring and enforcement, undermining CFM effectiveness.
  - Any new CFMs should be temporary and scaled back when capital flow pressures abate; CFMs should avoid leading to external payments arrears or sovereign default.
  - In crisis or near crisis situations, temporary CFMs on outflows could be part of a broader policy package to address the underlying cause of the crisis.

### Examples of macroprudential policies (Box 6, country examples)
- Ghana (In 2009):
  - Foreign and local currency deposits are subject to a 9 percent reserve requirement in domestic currency.
  - Reserve balance must be held with the Bank of Ghana.
  - Daily single foreign currency exposure limit was reduced from 15 to 10 percent of the capital base.
- Kenya:
  - Capital requirements are relatively high to prepare banks for overall risks.
  - Households do not borrow in foreign currency, except for a small segment of the mortgage market.
- Mozambique:
  - The limited impact of private capital inflows on overall liquidity conditions has not called for specific macroprudential measures.
- Nigeria:
  - If exposure to a particular industry within a sector is in excess of 20 percent of total credit facilities of a bank the risk weight of the entire portfolio in that industry shall be 150 percent.
- Senegal:
  - No traditional macroprudential regulations have been employed.
- Tanzania:
  - In late 2011 net open position on foreign exchange was reduced.
- Uganda:
  - Ongoing design of contingency plans to deal with excessive inflows through macroprudential means.

### Conclusions and recommended priorities for frontier markets
- Key conclusions:
  - Foreign portfolio flows to sub-Saharan Africa grew considerably in the past three years; sub-Saharan African frontier markets were the main beneficiaries.
  - Most frontier markets have so far been relatively unscathed by recent global financial market turbulence, partly reflecting illiquid financial markets and relatively strong fundamentals.
  - Greater integration with global financial markets implies increasing vulnerability to global financial shocks and possible reversals.
- Priority areas to strengthen:
  - Improving data:
    - Need to improve data quality and timeliness, including balance-of-payments data and coverage of cross-border private capital flows and stocks.
    - Official data on private capital portfolio flows and stocks are, in a number of countries, based on defective surveys with partial coverage or poor quality.
  - Enhancing macroeconomic and financial policies:
    - Managing capital flows is more likely to succeed if supported by sound fiscal, monetary, and exchange rate policies and adequate fiscal and international reserve buffers.
    - Strong fundamentals have helped limit investor punishment during global turbulence.
  - Improving capacity to effectively use macroprudential policies:
    - Macroprudential measures can prevent buildup of systemic risks from volatile capital flows.
    - Effectiveness requires progress in monitoring systemic risk buildup and using tools timely; supervisory resources, qualified staff, high-frequency data, and analytical tools are limited and need strengthening.
  - Improving the toolkit of capital flow management measures:
    - When macroeconomic policy room is limited, CFMs may be considered to temper volatile flows, but effective implementation requires significant institutional improvements.
    - Imposition of new CFMs, especially on outflows, must consider negative effects on future inflows and financial deepening; CFMs on outflows should be a last resort in financial crisis situations.

### Annex — econometric analysis (method and key points)
- Regression setup and data:
  - Regressions reported in Table A.1 are run on an annual panel covering the period 1991–2012 for selected emerging and frontier markets.
  - Dependent variable: ratio of portfolio liabilities (nonresident purchases of domestic assets net of sales) to GDP.
  - Lagged regressors used to minimize endogeneity.
  - Push factors included: U.S. Treasury bill/10-year yields, VIX (global risk).
  - Pull factors included: inflation level, fiscal balance, gross public debt, output growth, change in reserves, stock market development.
  - Random effects used to allow inclusion of a dummy for sub-Saharan African frontier markets.
  - Note: A table note states: "The sample period covers 2001–12 years."
- Key empirical observations:
  - Running regressions for sub-Saharan African frontier markets alone produced no statistically significant relationships, likely because of poor data quality and small sample size.
  - Another set of regressions (Table A.2) focused on sub-Saharan African frontier markets using quarterly data and disaggregating portfolio flows into equity and bond flows; dependent variables based on EPFR data, defined as flows into equity and bond funds of the country divided by the stock at the beginning of the quarter.
  - A crisis dummy equals 1 from 2008:Q4 onward.
  - Quarterly estimation limited by lack of quarterly availability for most pull factors; quarterly output growth available only for three countries.
  - Ghana and Nigeria were selected as a subsample given larger portfolio inflows.

*Italic: Source — _afr1401 - Box 5 . Capital Flows Management Measures in Sub-Saharan African Frontier Markets (PDF chapter/section)._

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/dp/2014/_afr1401.pdf_
