Abstract
This Selected Issues paper analyzes spillover risks for Colombia. It highlights that external shocks could spill over to the Colombian economy through the country’s important and growing trade and financial linkages with the rest of the world. Colombia would be most exposed to a decline in oil prices, which could have a sizable adverse impact on the balance of payments, the fiscal accounts and growth. Growth shocks in key trading partners could also have a negative impact, particularly in the United States, which is Colombia’s main trading partner. Colombia’s fiscal rule and adjustment in the context of resource wealth is also analyzed.
Colombia: Assessing Spillover Risks1
External shocks could spill over to the Colombian economy through the country’s important and growing trade and financial linkages with the rest of the world. Colombia would be most exposed to a decline in oil prices, which could have a sizable adverse impact on the balance of payments, the fiscal accounts and growth. Growth shocks in key trading partners could also have a negative impact, particularly in the United States, which is Colombia’s main trading partner. The projected rise in U.S. Treasury yields is estimated to increase Colombian government debt yields by about the same amount. Stress in the global financial system is estimated to have moderate direct effects on Colombia, while shocks in the Central American banking systems could have a significant impact on Colombian banks. The main channel of transmission of outward spillovers from Colombia to the region would be through the financial system given the significant market position of Colombian banks in Central America. Potential real outward spillovers through trade to the region are estimated to be small.
A. Key Linkages
1. Colombia has important and growing trade and financial linkages with the rest of the world.
Trade. Colombia’s total trade of goods (exports and imports) with the rest of the world rose from 25 percent of GDP in 2000 to an estimate of 31 percent of GDP in 2013. In recent years, the authorities have pursued an active policy to bolster trade. Colombia has signed or is negotiating free trade agreements with a number of countries, including the United States (implemented in 2012), Canada, Chile, Mexico, Switzerland, the European Union, South Korea, Turkey, Japan, China, Costa Rica, Panama, and Israel. In addition, Colombia is looking to increase trade with its Latin American neighbors through the “Pacific Alliance,” an economic integration effort launched in 2011 comprising Chile, Colombia, Mexico and Peru. 2
Exports. About 70 percent of Colombia’s total exports are commodities, including petroleum products (accounting for over 50 percent of total exports), coal, gold, emeralds, coffee, nickel, flowers, and bananas. Colombia is the third largest Latin American exporter of oil (after Venezuela and Mexico), and the largest exporter of coal to the United States. The main destinations of exports are the United States (receiving about 40 percent of total exports), the European Union (15 percent), China (5 percent), and some Latin American countries, including Venezuela, Ecuador, Chile, Brazil, Peru, Mexico, and Central America. Venezuela used to be Colombia’s second largest trading partner, but bilateral trade ties have significantly diminished since 2009 due to political disputes. It remains, however, an important trading partner for manufacturing products.
Imports. Colombia’s main import products are industrial equipment, transportation equipment, consumer goods, chemicals, paper products, and electricity. The main sources of Colombia’s imports are the United States (30 percent of total imports), China (11 percent), Mexico (10 percent), and Brazil (5 percent).
Remittances. Colombia receives a significant amount of workers’ remittances. Net remittances amounted to an estimated 1.2 percent of GDP in 2013. The relative importance of remittances as a share of GDP has declined, however, in recent years, from a peak of 3.5 percent of GDP in 2003.
Foreign direct investment (FDI). Inward FDI grew rapidly in recent years, notably in the oil and mining sectors, reaching a record high of nearly US$17 billion in 2013 (4.4 percent of GDP). As of end-2013, the stock of foreign direct amounted to US$128 billion (33 percent of GDP). The United States, the United Kingdom, and Spain constitute the main sources of the investment (accounting for a combined 45 percent of the total stock). Colombia’s outward direct investment also rose significantly in recent years, most notably due to the purchase of banks in Central America by the largest Colombian banks.
External credit lines. According to data from the Bank of International Settlements (BIS), international banks have significant claims on Colombian borrowers. As of September 2013, these claims reached US$47 billion (12 percent of GDP), originating mostly from European banks (US$23 billion or 49 percent of the total)—of which US$18 billion were from Spanish banks (equivalent to 79 percent of all the European banks’ claims)—, U.S. banks (US$11 billion or 23 percent of the total), and Japanese banks (US$2.7 billion or about 6 percent of the total). Of these claims, the bulk, 61 percent of the total (US$28.8 billion), was on the non-bank private sector, while 18 percent (US$8.5 billion) were on banks. In addition to these claims, BIS reporting banks registered other exposures to Colombia, including derivatives, guarantees, and credit commitments, for about US$25 billion (6 percent of GDP).
Bond issuance. The Colombian government and state-owned oil company, Ecopetrol, have fluid access to international capital markets, and have issued a significant amount of external debt at favorable terms in recent years. Colombia is rated BBB (two notches into investment grade level) by all three international credit rating agencies. In 2013, the government placed external bonds for US$2.6 billion and Ecopetrol issued US$2.5 billion. In 2014, the government placed US$2 billion 30-year external bonds at a yield of 5.65 percent, 190 basis points over U.S. Treasuries. As of end-2013, the stock of external Colombian government bonds outstanding was US$20.6 billion, and Ecopetrol’s amounted to US$4 billion. By contrast, external bond issuance by private sector companies has been relatively small, concentrated in a few large corporates.
Portfolio flows. Portfolio inflows to Colombia are relatively low as a share of total external inflows and vis-à-vis FDI flows, and are mostly directed to government bonds. These flows, however, have risen rapidly in recent years driven by increasing inflows by non-residents. In 2013, net portfolio inflows (excluding pension funds) are estimated to have totaled US$3.2 billion (0.8 percent of GDP).
International investment position. Colombia’s net international investment position (NIIP) improved over the past decade. At end-2013, the Colombia’s net IIP position stood at -27 percent of GDP (-US$104.2 billion), improving from -31 percent of GDP in 2003. Gross foreign assets stood at 33 percent of GDP, while external liabilities amounted to 60 percent of GDP, of which about 55 percent were FDI liabilities (up from a share of 35 percent in 2003).
Merchandise Trade
(In percent of GDP)
Citation: IMF Staff Country Reports 2014, 167; 10.5089/9781498302128.002.A001
Sources: WEO.Inward Foreign Direct Investment, by sector
(In billions of US$)
Citation: IMF Staff Country Reports 2014, 167; 10.5089/9781498302128.002.A001
Source: Banco de la República.Claims of BIS reporting banks on Colombia as of September 2013
Claims of BIS reporting banks on Colombia as of September 2013
| USD bn | Share of total (percent) |
||||
|---|---|---|---|---|---|
| Cross-border and local claims in all currencies | |||||
| Total | 47.1 | ||||
| Of which: Cross-border claims | 15.4 | 32.7 | |||
| Banks | 8.5 | 18.0 | |||
| Public sector | 9.8 | 20.8 | |||
| Non-bank private sector | 28.8 | 61.2 | |||
| By nationality of reporting banks | |||||
| European banks | 23.0 | 48.8 | |||
| Of which: Spain | 18.1 | 38.4 | |||
| UK | 2.0 | 4.3 | |||
| Germany | 1.3 | 2.7 | |||
| United States | 11.1 | 23.5 | |||
| Japan | 2.7 | 5.7 | |||
| Chile | 2.1 | 4.4 | |||
| Others | 8.3 | 17.6 | |||
| Other exposures | 24.6 | ||||
| Derivatives | 1.2 | ||||
| Guarantees | 18.8 | ||||
| Credit commitments | 4.6 | ||||
Claims of BIS reporting banks on Colombia as of September 2013
| USD bn | Share of total (percent) |
||||
|---|---|---|---|---|---|
| Cross-border and local claims in all currencies | |||||
| Total | 47.1 | ||||
| Of which: Cross-border claims | 15.4 | 32.7 | |||
| Banks | 8.5 | 18.0 | |||
| Public sector | 9.8 | 20.8 | |||
| Non-bank private sector | 28.8 | 61.2 | |||
| By nationality of reporting banks | |||||
| European banks | 23.0 | 48.8 | |||
| Of which: Spain | 18.1 | 38.4 | |||
| UK | 2.0 | 4.3 | |||
| Germany | 1.3 | 2.7 | |||
| United States | 11.1 | 23.5 | |||
| Japan | 2.7 | 5.7 | |||
| Chile | 2.1 | 4.4 | |||
| Others | 8.3 | 17.6 | |||
| Other exposures | 24.6 | ||||
| Derivatives | 1.2 | ||||
| Guarantees | 18.8 | ||||
| Credit commitments | 4.6 | ||||
Portfolio flows to Colombia (excluding pension funds)
(In billions of USD)
Citation: IMF Staff Country Reports 2014, 167; 10.5089/9781498302128.002.A001
Source: Banco de la República.B. Inward Spillovers
2. As a result of its important linkages with the rest of the world, the Colombian economy is significantly exposed to external shocks. Colombia is particularly exposed to a sharp decline in commodity prices, especially oil, negative growth shocks in key trading partners, and a deterioration in global financial conditions. The prospect of rising interest rates in the United States, as the Federal Reserve normalizes monetary policy, could also negatively impact Colombia if it were to lead to global financial volatility and lower growth in emerging markets.
A sharp decline in oil prices would cut export receipts (which account for over one half of total exports), reduce fiscal revenue (oil-related revenues amount to about 4½ percent of GDP or 16 percent of total government revenue), and negatively impact economic activity. It could also significantly reduce FDI inflows (considering that a significant share of them goes to the oil sector), with a further negative effect on the balance of payments. A drop of US$10 in the price of oil would reduce exports by about US$3.3 billion (0.9 percent of GDP) and fiscal revenue by about 0.4 percent of GDP.
A negative growth shock in key trading partners would reduce the demand for Colombian exports, weakening economic growth. Colombia could be particularly affected by a growth shock in the United States, the European Union, China, and a set of Central and South American countries, which account for the bulk of Colombia’s foreign trade. Despite trade linkages with neighboring Venezuela have diminished in recent years, spillovers from any potential political or economic stress in the latter country could still be significant (particularly in border areas), given that Venezuela remains an important partner for manufacturing trade.
Deterioration in global financial conditions (stemming from any event that could increase global risk aversion, such as geopolitical tensions or a revision in market expectations about macroeconomic fundamentals in the US, Europe, Japan, China, or the emerging markets) could reduce bank credit lines to Colombian borrowers and portfolio flows, negatively affecting the external accounts and economic activity. This shock could also operate through the channels described above by weakening global economic growth and/or triggering a decline in oil prices.
The expected rise in US interest rates projected for the coming years is poised to negatively affect Colombia by increasing external borrowing costs (for both the government and the private sector) and reducing capital inflows. This could negatively affect the fiscal balance and corporate balance sheets, weakening investment spending and growth. This impact, however, could be mitigated (or even offset) by stronger growth in the United States (as is to be expected as monetary policy normalizes) and subsequent greater demand for Colombian exports (considering that the U.S. is the main destination of Colombian exports). In the event that the rise in U.S. interest rates were not accompanied by a corresponding increase in U.S. growth or led to large capital outflows from emerging markets, weaker global economic growth and/or a decline in oil prices, Colombia could see a net negative impact through the channels described above. In turn, financial volatility in emerging markets could also trigger dislocations in the domestic capital market and a sharp decline in the value of domestic financial assets, with negative implications for the real economy.
3. Staff has developed simulation exercises to estimate the potential impact of some of the shocks that could adversely affect the Colombian economy. These exercises are meant to be illustrative, and do not cover the full effect of the potential shocks. Staff has estimated the impact on Colombia’s real GDP growth of a decline in oil prices and a slowdown in the pace of growth of key trading partners. It has also studied the effect that the increase in 10-year U.S. Treasury yields could have on Colombian government bond yields; simulated the potential impact from stress in the international banking system; and quantified the exposure of the Colombian banking system to shocks in the Central American banking systems.
Impact from a decline in oil prices
4. Colombia depends heavily on oil exports, making it vulnerable to a drop in oil prices. As a result, a decline in oil prices would likely lead to lower growth of investment and consumption. The correlation between oil prices and investment and private consumption growth in the 2000–2013 period was 0.93 and 0.45, respectively. Empirical analyses suggest that oil prices Granger-cause and have statistically significant effects on the growth of investment and private consumption with a three-quarter lag.
Oil prices and investment
Citation: IMF Staff Country Reports 2014, 167; 10.5089/9781498302128.002.A001
Sources: Fund staff estimates.Oil prices and private consumption
Citation: IMF Staff Country Reports 2014, 167; 10.5089/9781498302128.002.A001
Sources: Fund staff estimates.5. A decline of US$10 in oil prices could reduce Colombia’s real GDP growth by about 1 percentage point cumulatively over one year after the shock. A VAR model, including the international price of oil, Colombia’s exchange rate vis-à-vis the U.S dollar, oil export volumes and Colombia’s real GDP growth was used to assess the extent to which a decline in oil prices could affect Colombia’s growth. Cholesky decomposition was used as shock identification strategy, with the ordering of the variables as listed above. The impulse response functions to the oil price shock suggest a slight depreciation of the peso, and a reduction in real GDP growth. The cumulative impact of the shock on real GDP growth amounted to ½ of a percentage point of GDP after six months after the shock.
Impact from a slowdown in growth in key trading partners
6. A multi-country vector autoregressive (VAR) model was used to assess the impact on Colombia’s GDP growth of a slowdown in growth in key trading partners. The exercise followed the approach described in Poirson and Weber (2011). The model was estimated with quarterly real GDP data for the sample period 2000Q1 to 2019Q4 (using IMF staff projections for 2014–2019) for Colombia, the United States, China, the Euro Area, Venezuela, Argentina, Brazil, Chile, Costa Rica, Ecuador, Mexico, Panama, and Peru, accounting for about 75 percent of Colombia’s total exports. Identification was obtained using the Cholesky ordering of the countries in the sample. For the ordering we distinguished two set of countries: countries with large economies (the U.S., the Euro Area, and China) that can be leading countries (i.e., not contemporaneously affected by the other countries), and countries with smaller economies (Venezuela, Argentina, Brazil, Chile, Costa Rica, Ecuador, Mexico, Panama, and Peru), which were ordered after the former. The orderings were arranged so that the three countries in the first group would be ordered either first, second or third, and the countries in the second group would always come after them. This procedure resulted in 24 different orderings.
7. Five separate shock scenarios were analyzed. Each scenario assumed a reduction of 1 percentage point in the projected rate of growth for 2014 of the United States, the Euro Area, China, Venezuela, and the other key Latin American trading partners (ex-Venezuela) relative to baseline projections. The exercise showed the following results:
United States growth shock. An adverse shock to U.S. growth would cut Colombia’s GDP growth by about 0.1 percentage points in 2014, with the adverse impact diminishing to almost zero in 2015. The low sensitivity stemming from the exercise despite the U.S. being Colombia’s largest trading partner possibly owes to the fact that a significant share of Colombia’s trade with the U.S. is commodities (e.g., coal), which is possibly relatively more inelastic to adverse growth shocks.
Growth shock in the Euro Area. A growth shock in the Euro Area would reduce Colombia’s GDP growth by about 0.2 percentage points both in 2014 and 2015.
China growth shock. The growth shock in China would have a similar intensity as the shock in Europe, of -0.1 percentage points in 2014 and -0.3 percentage points in 2015. The shock would propagate to Colombia directly as well as through its impact on the U.S., Euro Area, and Latin America.
Venezuela growth shock. The exercise suggests that a slowdown in Venezuela would have no significant impact on Colombia. The results seems counter-intuitive given the still-significant (though greatly diminished) manufacturing trade between the two countries. The low sensitivity stemming from the exercise is probably due to the significant variability in growth in Venezuela in recent years relative to Colombia’s more stable growth pattern.
Combined growth shock in key Latin American trade partners (ex-Venezuela). A combined adverse growth shock in Colombia’s main Latin American trade partners (ex-Venezuela) would have a substantial impact on Colombia’s growth, of -0.2 percentage points in 2014 and -0.5 percentage points in 2015 respectively. The results highlight the important (and growing) trade links between Colombia and its Latin American neighbors, particularly in manufacturing, the close association of their economic cycles, and the fact that most are subject to similar shocks.
Effects on a 1-percentage point slowdown in 2014 GDP growth of selected countries on Colombia’s real GDP growth 1/
Results of multi-country VAR model.
Comprises Argentina, Brazil, Chile, Costa Rica, Ecuador, Mexico, Panama, and Peru
Effects on a 1-percentage point slowdown in 2014 GDP growth of selected countries on Colombia’s real GDP growth 1/
| 2014 | 2015 | |
|---|---|---|
| United States | −0.1 | 0 |
| Euro area | −0.2 | −0.2 |
| China | −0.1 | −0.3 |
| Venezuela | 0 | 0 |
| Selected Latin American countries 2/ | −0.2 | −0.5 |
Results of multi-country VAR model.
Comprises Argentina, Brazil, Chile, Costa Rica, Ecuador, Mexico, Panama, and Peru
Effects on a 1-percentage point slowdown in 2014 GDP growth of selected countries on Colombia’s real GDP growth 1/
| 2014 | 2015 | |
|---|---|---|
| United States | −0.1 | 0 |
| Euro area | −0.2 | −0.2 |
| China | −0.1 | −0.3 |
| Venezuela | 0 | 0 |
| Selected Latin American countries 2/ | −0.2 | −0.5 |
Results of multi-country VAR model.
Comprises Argentina, Brazil, Chile, Costa Rica, Ecuador, Mexico, Panama, and Peru
Impact from the normalization of monetary policy in the United States
8. The expected normalization of monetary policy in the United States will affect the Colombian economy through different channels. Economic growth is projected to strengthen and unemployment to decline in the U.S. in the coming years. The strengthening of the U.S. economy should benefit the Colombian economy by increasing the demand for Colombian exports. However, as the economic picture in the U.S. improves, the Federal Reserve is expected to normalize the monetary policy stance, initially by tapering asset purchases (as it has already begun doing), eventually by ending them, and ultimately by increasing short term interest rates. This will push up interest rates in Colombia, increasing borrowing costs and curtailing investment and growth. The net impact on Colombia from stronger growth and higher interest rates in the U.S. is thus uncertain. To assess the effect of an increase in U.S. interest rates on Colombia, staff estimated a model of the yield on 10-year peso-denominated government bonds.
9. Staff estimates suggest that changes in U.S. Treasury yields have a significant short-term impact on local Colombian bond yields. A vector error correction model (VECM) was estimated to examine the relationship between yields on 10-year Colombian peso-denominated government bonds, and the short-term interest rate, the yield on 10-year U.S. Treasury bills, a measure of global risk aversion, exchange rate risk, and idiosyncratic sovereign credit risk. Using data from mid-2010 to late 2013, the model showed that a 100 basis points-increase in 10-year U.S. Treasury yields translates into an increase of 91 to 123 basis points in 10-year Colombian government bond yields after 6 months. Hence, the analysis suggests that Colombian interest rates could rise sharply in the event there were new bouts of market volatility caused by a shift in expectations or surprises about U.S. monetary policy. The same analysis for other countries in the region shows that the response of domestic government bond yields to an increase in U.S. Treasury yields would be larger in Brazil and Peru, similar to the one estimated for Colombia in Mexico, and lower in Chile.
Response of Colombian 10-year government bond yields to a 100-basis point increase in 10-year U.S. Treasury yields
(Cumulative responses after 6 months, in basis points)
Citation: IMF Staff Country Reports 2014, 167; 10.5089/9781498302128.002.A001
Source: IMF staff estimates.Response of 10-year government bond yields to a 100-basis point increase in 10-year U.S. Treasury yields
(Cumulative responses after 6 months, in basis points)
Citation: IMF Staff Country Reports 2014, 167; 10.5089/9781498302128.002.A001
Source: IMF staff estimates.10. The model suggests that 10-year government bonds yields are broadly in line with fundamentals. After a significant rally in 2012, Colombian government bond yields had fallen about 60 basis points below the level implied by fundamentals (i.e., the predicted value from the VECM) by March 2013. Since then, bond yields started rising. After May 22, bond yields rose sharply, initially overshooting, but eventually converging to the fundamental values estimated by the model. In early 2014, at about 7 percent, 10-year peso-denominated government bond yields were broadly in line with fundamentals as predicted by the model. The analysis does not cover the most recent decline in government bond yields following the announcement by JP Morgan (in late March) of the increase of Colombian bonds’ weight in its emerging market bond indices.
Colombian 10-year government bond yields
(Actual yields and model estimates, in percent per annum)
Citation: IMF Staff Country Reports 2014, 167; 10.5089/9781498302128.002.A001
Source: IMF staff estimates.Spillovers from stress in international banks
11. A spillover analysis was conducted to estimate the effects from stress in international banks that are involved in lending to Colombian borrowers. The analysis covers cross-border lending to Colombian borrowers by international banks, as well as lending through their affiliates in Colombia. The exercise drew on the RES Bank Contagion Module, based on the banking statistics of the BIS. In the simulation, a first round considered losses on asset holdings of international banks that reduce (partially or fully) their capital, based on assumptions of a decline in value of different types of assets (e.g., claims on the public sector, banking sector, and non-bank private sector of an individual country or group of countries). In the second round, if losses are large, banks were assumed to restore their capital adequacy ratios through deleveraging (i.e., sale of assets and refusal to roll-over existing loans), thus reducing credit lines to all borrowers, including those in Colombia as well as those in other countries. In the third round, banks were assumed to reduce their lending to other banks, potentially triggering fire sales, further deleveraging, and additional losses at other banks. Final convergence is achieved when no further deleveraging occurs.
12. The model suggests that foreign credit availability to Colombian borrowers could be significantly affected by losses in claims of international banks on selected economies. Based on the assumed decline in value (10 percent in the exercise) of private and public sector assets of a certain country or group of countries, the model estimates the losses of foreign banks and their implications for credit to Colombian borrowers. The largest direct impact of this shock in terms of reduction in international banks’ credit to Colombian borrowers would stem from combined losses in European assets (6.6 percent of GDP). Considering individual countries, the shocks that would pose the largest adverse affects in credit to Colombian borrowers would be losses in Spanish assets (5.3 percent of GDP), Canadian assets (2.6 percent of GDP), U.K. assets (1.2 percent of GDP), Japanese (0.8 percent of GDP) and U.S. and German assets (0.4 percent of GDP).
Spillovers to Colombia from International Banks’ Exposures as of September 2013
Magnitude denotes the percent of on-balance sheet claims (all borrowing sectors) that lose value.
Reduction in foreign banks credit to Colombia due to the impact of the analyzed shock in their balance sheets, assuming a uniform deleveraging across domestic and external claims.
Greece, Ireland, Portugal, Italy, Spain, France, Germany, Netherlands, and the UK.
Spillovers to Colombia from International Banks’ Exposures as of September 2013
| Country | Shock originating from: | Impact on Credit Availability |
|---|---|---|
| Magnitude 1/ | (percent of GDP) 2/ | |
| USA | 10 | −0.4 |
| Canada | 10 | −2.6 |
| UK | 10 | −1.2 |
| Germany | 10 | −0.4 |
| France | 10 | −0.3 |
| Spain | 10 | −5.3 |
| Italy | 10 | 0.0 |
| Greece | 10 | 0.0 |
| Portugal | 10 | 0.0 |
| Selected European Countries 3/ | 10 | −6.6 |
| Japan | 10 | −0.8 |
Magnitude denotes the percent of on-balance sheet claims (all borrowing sectors) that lose value.
Reduction in foreign banks credit to Colombia due to the impact of the analyzed shock in their balance sheets, assuming a uniform deleveraging across domestic and external claims.
Greece, Ireland, Portugal, Italy, Spain, France, Germany, Netherlands, and the UK.
Spillovers to Colombia from International Banks’ Exposures as of September 2013
| Country | Shock originating from: | Impact on Credit Availability |
|---|---|---|
| Magnitude 1/ | (percent of GDP) 2/ | |
| USA | 10 | −0.4 |
| Canada | 10 | −2.6 |
| UK | 10 | −1.2 |
| Germany | 10 | −0.4 |
| France | 10 | −0.3 |
| Spain | 10 | −5.3 |
| Italy | 10 | 0.0 |
| Greece | 10 | 0.0 |
| Portugal | 10 | 0.0 |
| Selected European Countries 3/ | 10 | −6.6 |
| Japan | 10 | −0.8 |
Magnitude denotes the percent of on-balance sheet claims (all borrowing sectors) that lose value.
Reduction in foreign banks credit to Colombia due to the impact of the analyzed shock in their balance sheets, assuming a uniform deleveraging across domestic and external claims.
Greece, Ireland, Portugal, Italy, Spain, France, Germany, Netherlands, and the UK.
13. The indirect effects on the Colombian economy associated with the analyzed shocks, however, could be much larger. The model estimates do not consider the negative effects of deleveraging on market confidence, balance sheets of corporates, and output growth, which could have a sizable adverse impact on the economy. The impact stemming from these factors could be potentially more damaging for the Colombian economy than the estimated direct spillovers estimated in the exercise.
Spillovers from losses in Colombian banks’ investments abroad
14. In recent years, Colombia’s three largest financial institutions expanded aggressively abroad. The aim of their expansion plans was to diversify and complement their home operations. With this purpose, they acquired financial institutions, primarily in Central America, where they attained a significant market position. As of end-2013, the assets of Colombian banks’ subsidiaries abroad reached US$54 billion, accounting for 26.8 percent of the total assets of the Colombian banking system.
15. The international expansion of the Colombian banks has left them significantly exposed to shocks in the Central American region. As of end-2013, the assets of Colombian banks’ subsidiaries in Central America amounted to US$44.4 billion, equivalent to 82 percent of the total assets of Colombian banks’ subsidiaries abroad. These accounted for about 22 percent of the total assets of the Colombian banking system, and about a third of the total assets of the three Colombian banks with investments in Central America. Hence, a shock to the Central American banking systems could have a significant impact on the Colombian banking system. Ranked by the amount of the assets of their subsidiaries in the region, Colombian banks are most exposed to shocks in Panama (concentrating 55 percent of Colombian banks’ assets in the region), El Salvador (17 percent), Costa Rica (12 percent), and Honduras (8 percent). Individual country exposures, however, may not give a completely accurate picture, as some of the subsidiaries of the Colombian banks are domiciled in one country but operate across the Central American region, particularly in the case of the Panamanian banks. Hence, the above figures may underestimate exposure to individual countries. Exposure to shocks in Central America is mitigated by the strength of the balance sheets and soundness indicators of the subsidiaries in the region, which for the most part boast adequate capitalization and robust profitability.
Assets of Colombian banks’ subsidiaries abroad (2013)
Assets of Colombian banks’ subsidiaries abroad (2013)
| In USD bn | As a share of total investments abroad |
As a share of Colombia’s banking system’s assets |
|
|---|---|---|---|
| Panama | 24.5 | 45.3 | 12.1 |
| El Salvador | 7.6 | 14.0 | 3.8 |
| Costa Rica | 5.4 | 9.9 | 2.7 |
| Honduras | 3.6 | 6.6 | 1.8 |
| British Virgin Islands | 3.0 | 5.6 | 1.5 |
| Guatemala | 2.2 | 4.0 | 1.1 |
| Chile | 1.8 | 3.4 | 0.9 |
| Peru | 1.7 | 3.2 | 0.9 |
| Cayman Islands | 1.3 | 2.4 | 0.6 |
| Nicaragua | 1.2 | 2.2 | 0.6 |
| Mexico | 1.1 | 2.1 | 0.6 |
| Puerto Rico | 0.3 | 0.6 | 0.2 |
| Barbados | 0.2 | 0.3 | 0.1 |
| Dom Rep | 0.1 | 0.1 | 0.0 |
| Bahamas | 0.1 | 0.1 | 0.0 |
| Uruguay | 0.03 | 0.05 | 0.0 |
| USA | 0.02 | 0.03 | 0.0 |
| TOTAL | 54.0 | 100.0 | 26.8 |
| Central America | 44.4 | 82.1 | 22.0 |
Assets of Colombian banks’ subsidiaries abroad (2013)
| In USD bn | As a share of total investments abroad |
As a share of Colombia’s banking system’s assets |
|
|---|---|---|---|
| Panama | 24.5 | 45.3 | 12.1 |
| El Salvador | 7.6 | 14.0 | 3.8 |
| Costa Rica | 5.4 | 9.9 | 2.7 |
| Honduras | 3.6 | 6.6 | 1.8 |
| British Virgin Islands | 3.0 | 5.6 | 1.5 |
| Guatemala | 2.2 | 4.0 | 1.1 |
| Chile | 1.8 | 3.4 | 0.9 |
| Peru | 1.7 | 3.2 | 0.9 |
| Cayman Islands | 1.3 | 2.4 | 0.6 |
| Nicaragua | 1.2 | 2.2 | 0.6 |
| Mexico | 1.1 | 2.1 | 0.6 |
| Puerto Rico | 0.3 | 0.6 | 0.2 |
| Barbados | 0.2 | 0.3 | 0.1 |
| Dom Rep | 0.1 | 0.1 | 0.0 |
| Bahamas | 0.1 | 0.1 | 0.0 |
| Uruguay | 0.03 | 0.05 | 0.0 |
| USA | 0.02 | 0.03 | 0.0 |
| TOTAL | 54.0 | 100.0 | 26.8 |
| Central America | 44.4 | 82.1 | 22.0 |
Assets and market share of Colombian banks’ subsidiaries abroad as of December 2013
Assets and market share of Colombian banks’ subsidiaries abroad as of December 2013
| Colombia | Panama | Costa Rica | Guatemala | Honduras | El Salvador | Nicaragua | Central America |
|
|---|---|---|---|---|---|---|---|---|
| Assets, in billions of USD | ||||||||
| Total System | 201.7 | 97.7 | 41.6 | 28.1 | 17.0 | 14.2 | 5.3 | 203.9 |
| Bancolombia | 46.7 | 15.4 | 4.0 | 19.4 | ||||
| Banco de Bogota | 29.8 | 8.0 | 3.8 | 2.2 | 2.4 | 1.5 | 1.2 | 19.0 |
| Davivienda | 24.3 | 1.2 | 1.6 | 1.2 | 2.1 | 6.0 | ||
| Colombian banks | 24.5 | 5.4 | 2.2 | 3.6 | 7.6 | 1.2 | 44.4 | |
| Assets as a share of parent bank’s assets | ||||||||
| Bancolombia | 32.9 | 8.6 | 41.5 | |||||
| Banco de Bogota | 26.8 | 12.8 | 7.2 | 8.0 | 5.2 | 3.9 | 63.9 | |
| Davivienda | 4.7 | 6.4 | 5.0 | 8.4 | 24.6 | |||
| Colombian banks | 24.3 | 5.3 | 2.1 | 3.6 | 7.5 | 1.2 | 44.0 | |
| Market share (by assets), in percent | ||||||||
| Bancolombia | 23.2 | 15.7 | 28.1 | 9.5 | ||||
| Banco de Bogota | 14.8 | 8.2 | 9.1 | 7.7 | 13.9 | 10.8 | 22.4 | 9.3 |
| Davivienda | 12.1 | 1.2 | 3.8 | 7.1 | 14.4 | 2.9 | ||
| Colombian banks | 25.1 | 12.9 | 7.7 | 21.1 | 53.2 | 22.4 | 21.8 |
Assets and market share of Colombian banks’ subsidiaries abroad as of December 2013
| Colombia | Panama | Costa Rica | Guatemala | Honduras | El Salvador | Nicaragua | Central America |
|
|---|---|---|---|---|---|---|---|---|
| Assets, in billions of USD | ||||||||
| Total System | 201.7 | 97.7 | 41.6 | 28.1 | 17.0 | 14.2 | 5.3 | 203.9 |
| Bancolombia | 46.7 | 15.4 | 4.0 | 19.4 | ||||
| Banco de Bogota | 29.8 | 8.0 | 3.8 | 2.2 | 2.4 | 1.5 | 1.2 | 19.0 |
| Davivienda | 24.3 | 1.2 | 1.6 | 1.2 | 2.1 | 6.0 | ||
| Colombian banks | 24.5 | 5.4 | 2.2 | 3.6 | 7.6 | 1.2 | 44.4 | |
| Assets as a share of parent bank’s assets | ||||||||
| Bancolombia | 32.9 | 8.6 | 41.5 | |||||
| Banco de Bogota | 26.8 | 12.8 | 7.2 | 8.0 | 5.2 | 3.9 | 63.9 | |
| Davivienda | 4.7 | 6.4 | 5.0 | 8.4 | 24.6 | |||
| Colombian banks | 24.3 | 5.3 | 2.1 | 3.6 | 7.5 | 1.2 | 44.0 | |
| Market share (by assets), in percent | ||||||||
| Bancolombia | 23.2 | 15.7 | 28.1 | 9.5 | ||||
| Banco de Bogota | 14.8 | 8.2 | 9.1 | 7.7 | 13.9 | 10.8 | 22.4 | 9.3 |
| Davivienda | 12.1 | 1.2 | 3.8 | 7.1 | 14.4 | 2.9 | ||
| Colombian banks | 25.1 | 12.9 | 7.7 | 21.1 | 53.2 | 22.4 | 21.8 |
C. Outward Spillovers
16. The main channel of transmission of shocks from Colombia to other countries would be through the financial sector (e.g., via cross-border bank lending and foreign direct investment links). This channel could be significant for some Central American countries, where Colombian banks have expanded significantly in recent years, in the event these came under financial stress. The countries most exposed to shocks in Colombian banks are El Salvador, Panama, Nicaragua, and Honduras, where Colombian banks have attained a significant market position. These risks, however, are mitigated by the strength of the capital positions and other soundness indicators of the Colombian banks.
Market share of Colombian banks’ subsidiaries in Central America (by assets)
Citation: IMF Staff Country Reports 2014, 167; 10.5089/9781498302128.002.A001
Source: Fund staff estimates based on Financial Superintendence.17. Another possible channel of transmission of shocks would be through trade. A slowdown in growth in Colombia could affect other countries by reducing their exports to Colombia. Trade data suggests that, in such an event, the most affected countries could be Bolivia, Bahamas and Ecuador, for which Colombian demand accounts for 4.8, 3.5 and 3.4 percent of their total exports, respectively. For the rest of trading partners in the region, the share of exports to Colombia in their total exports does not exceed 2 percent, suggesting that the potential for real spillovers is small.
18. There could be potential for pure market contagion through investor perception of regional risk, e.g., in response to adverse economic or political developments in Colombia, but the magnitude that such a shock could have is difficult to gauge.
Exports to Colombia as a share of total exports (2013)
Citation: IMF Staff Country Reports 2014, 167; 10.5089/9781498302128.002.A001
Sources: IMF COMTRADE data.References
Kamil et al, “U.S. Fed Tapering and Financial Spillovers to Emerging Portfolio Markets: The Effect on Sovereign Yields in Brazil and Mexico,” IMF Working Paper (forthcoming; Washington: International Monetary Fund).
Perreli and Góes, “Tapering Talks and Local Currency Sovereign Bond Yields: How South Africa Performed Relative to Its Emerging Market Peers,” IMF Working Paper (forthcoming; Washington: International Monetary Fund).
Perrelli, Roache, and Góes, “The Impact of the Tapering Tantrum on the Slope of the Yield Curve of Emerging Market Economies,” IMF Working Paper (forthcoming; Washington: International Monetary Fund).
Poirson and Weber, 2011, “Growth Spillover Dynamics from Crisis to Recovery,” IMF Working Paper (Washington: International Monetary Fund)
International Monetary Fund, 2014, Regional Economic Outlook, April 2014: “Taper Tantrum or Tedium: How Will the Normalization of U.S. Monetary Policy Affect Latin America and the Caribbean?”, Regional Economic Outlook for the Western Hemisphere (Washington).
