2. Outlook and Policy Issues for Latin America and the Caribbean
- International Monetary Fund. Western Hemisphere Dept.
- Published Date:
- May 2011
The region continues to grow at a robust pace, led by rapidly expanding domestic demand. Growth is projected to moderate this year, though a quicker withdrawal of monetary and fiscal policy stimulus may be necessary to guard against rising overheating risks, particularly in economies experiencing strong “tailwinds” from rising commodity export revenues and capital inflows. In countries with the strongest real links to advanced economies, economic weakness was more prolonged, but their recoveries are gaining traction. Throughout the region, rising global prices of food and fuel are adding to challenges of containing inflation and protecting the poor. Downside risks to the global outlook highlight the need to build policy buffers and to guard against an eventual reversal of capital inflows.
Real GDP growth is moderating yet remains robust and above trend rates in much of the region, led by strong private demand (Figure 2.1). Output gaps have closed or are quickly closing for much of the region and overheating pressures are emerging. Inflation is rising, current account deficits are widening, and in some countries credit and asset prices are growing rapidly. Double tailwinds from high commodity prices and easy external financing conditions, coupled with stillaccommodative macroeconomic policies are lifting economic activity above its potential level. Meanwhile, the recovery is gaining strength in Central America on the back of a gradual pickup of the U.S. economy and of domestic demand. In the Caribbean, economic activity is starting to recover after a deep recession, with output still well below precrisis levels in most countries.
Figure 2.1.Real GDP growth is moderating, yet domestic demand continues to grow strongly
Sources: National authorities; and IMF staff calculations.
1Includes Argentina, Brazil, Chile, Colombia, the Dominican Republic, Ecuador, Mexico, Peru, and Venezuela. Data are through 2010:Q4, except for Ecuador (2010:Q3).
Output growth is estimated to have reached about 6 percent in 2010, slightly higher (¼ percent) than projected in the October 2010 Regional Economic Outlook: Western Hemisphere. Domestic demand growth was somewhat stronger than anticipated, likely reflecting more favorable terms of trade and a slow pace of withdrawal of policy stimulus. Real GDP growth of the region is projected to moderate to 4¾ percent this year and to converge to its potential rate, about 4 percent, over the next two years. Growth will continue to be led by domestic demand, consistent with a further deterioration in the current account deficit. This baseline scenario assumes a significant withdrawal of the policy stimulus and a deceleration of private sector demand growth, particularly for countries that are large commodity exporters.
Economic performance will be less uneven within the region in 2011, compared with the previous two years. There are signs that the recovery is finally gaining traction in economies with closer real links to advanced economies, where recovery has lagged (Figure 2.2). South America’s commodity exporters will continue to lead the expansion, though growth is expected to moderate this year, in a healthy way, with the assumed withdrawal of the policy stimulus. The recovery in most of Central America and the Caribbean countries is projected to strengthen, though countries will continue to be constrained by the effects of weak employment conditions in advanced economies, less favorable terms of trade, and in some cases by high public debt.
Figure 2.2.The expansion will be less uneven, with output gaps closed or closing for much of the region. Differences in terms of trade persist
Sources: National authorities; and IMF staff calculations.
1 Central America includes the Dominican Republic and Panama.
2 Caribbean excludes the Dominican Republic.
Although downside risks dominate for the world economy, risks for the near-term outlook in Latin America are tilted to the expansionary side as a whole. A distinct risk is that domestic demand and GDP growth could be higher than the baseline scenario if the assumed policy tightening does not materialize or proves insufficient to slow domestic demand. Under this scenario, inflation and current account deficits also would turn out higher than we project, raising the risk of boom-bust dynamics down the road. In that sense, the “upside” risks for 2011 could turn out to be downside risks for subsequent years.
However, downside tail risks remain for the near term. Increased and continued political tensions in the Middle East and North Africa or a more protracted crisis in Japan could compromise the global recovery currently under way. A large further increase in oil prices driven by oil supply conditions would adversely affect global growth; such a global slowdown would likely bring also a decline in the price of other commodities, having an added negative impact on the region’s non-oil commodity exporters. Moreover, the situation in Europe remains fragile, as many sovereigns face large refinancing needs and banks require further capitalization. Although recent bouts in market volatility (Greece, Ireland, and Portugal) have not had large effects in Latin America, renewed turbulence in the euro area, particularly if the epicenter were to shift to Spain or to the core European economies, could have a stronger impact on the region.
The rest of this chapter is organized as follows. The next section discusses the policy challenges for the different subregions of Latin America, referring to specific countries when appropriate. The following sections provide more detailed views on policies for dealing with (i) rising commodity prices; (ii) accelerating credit growth and asset prices; and (iii) significant and volatile capital inflows.
2.2. Policy Challenges
With output gaps already closed in much of the region, the primary policy challenge is not very different from six months ago: first to normalize and eventually tighten the stance of fiscal and monetary policies, to avoid boom-bust dynamics down the road. Increasing world prices of food and fuels are adding to challenges facing the net commodity importing regions, particularly Central America and the Caribbean, where recovery from the crisis generally has been slower. For all countries, rising commodity prices create social challenges and strain the poor especially, as well as complicating the task of containing inflation.
South America—Managing Double Tailwinds
Macroeconomic policies should be tightened to prevent growing overheating pressures, particularly in the face of external stimulus to demand coming from strong terms of trade and easy external financing conditions. Consideration will need to be given to expanding the policy toolkit to limit the proyclicality of credit and excesses by the private sector. In some countries, addressing supply bottlenecks that are holding back trend growth remains a key priority.
Growth in most commodity exporting countries of South America is projected to moderate this year, after reaching an average of 6½ percent in 2010, as the policy stimulus and natural inventory cycle are unwound. Commodity exporters will continue to benefit from very favorable terms of trade, and countries with strong links to financial markets (Brazil, Chile, Colombia, Peru, and Uruguay) will also benefit from easy external financing conditions. Meanwhile, countries with strong trade links to Brazil (Argentina, Bolivia, Paraguay, and Uruguay) will continue to benefit from robust growth in that South American giant.
With continued rapid growth and output already above or around its potential level, inflation is projected to continue increasing this year in much of South America. Headline inflation is trending upward, though it still remains near targets or within target ranges in most countries (Figure 2.3). These developments reflect both incipient demand pressures and the (direct and indirect) effects of higher prices of food and fuel products. Higher inflation has been broadening into services and pushing core inflation and inflation expectations up in many countries, particularly those with tight labor market conditions (Brazil and Uruguay). In this light, many central banks have been raising policy rates (from low levels), yet only gradually, perhaps out of concern for attracting more capital flows and further strengthening exchange rates.
Figure 2.3.Headline inflation has picked up, with core and inflation expectations slowly trending upward, in the context of tight labor market conditions
Sources: National authorities; and IMF staf f calculations.
After appreciating sharply during much of last year, exchange rates have stabilized since October 2010, in tandem with some slowdown in portfolio flows to emerging economies worldwide (Figure 1.2). Although more positive news on the U.S. economy and higher asset prices in emerging economies themselves likely acted to slow capital flows, various policy measures also may have played a role. Countries not only stepped up foreign exchange market intervention, but also increased taxes on inflows, and strengthened or adopted additional macroprudential measures. Real exchange rates are in the upper part of the ranges observed over the past decade, though this largely reflects stronger fundamentals and improvements in their terms of trade.
Robust private sector demand is leading to a widening in the current account deficit in most countries (Colombia is an exception).1 The current account deficit for the more financially integrated economies of South America is projected to average about 2 percent of GDP in 2011, ¼ percentage point larger than in 2010, despite higher commodity export prices. A simple exercise of keeping export and import prices constant at 2005 levels, and controlling for the sensitivity of foreign investors’ profits and dividends to commodity prices, suggests that the current account would be 3 percentage points of GDP weaker this year if terms-of-trade gains had not occurred (holding volumes constant). The deterioration in the current account is being driven mainly by the private sector, where investment is outstripping saving (Figure 2.4).2 It is worth noting that although the increased geographic diversity of exports is a welcomed development; this has come at the expense of an increase in the commodity dependence of the region (Box 2.1).
Figure 2.4.Despite improved terms of trade, the current account deficit has widened further, led by a strong growth in private sector demand
Sources: IMF, International Financial Statistic; and IMF staff calculations.
Credit growth accelerated during much of last year and continues to do so; asset prices remain high after large gains (Figure 2.5). Private sector credit growth in real terms reached an average of 10–15 percent by end-2010, very strong though well below the 20–25 percent observed in the years prior to the global crisis. The credit expansion is now more broad-based and even across countries in the region, though Brazil and Peru are clear outliers since their credit slowdowns during the crisis were small and short lived.3 Equity prices increased sharply in 2010, and despite some normalization and reversals in recent months, they are up by an average of more than 20 percent in U.S. dollars over the past year. A more detailed discussion can be found in Section 2.4 on credit markets and asset prices.
Figure 2.5.Amid strong capital inflows, real exchange rates remain high, and credit is expanding fast.
Sources: National authorities; and IMF staff calculations.
Box 2.1.Changing Export Patterns in Latin America and the Caribbean
Increased trade openness over the past decade has been accompanied by significant change in the region’s trade structure. Emerging economies are playing a more significant role in trade relations, and commodities represent an increasing share of the region’s exports. Despite regional integration efforts, the share of exports within the region has been broadly unchanged.
Rising power of emerging economies. The share of exports to emerging economies has increased by 10 percentage points over the past decade, reaching 35 percent of total exports in 2010. Although this trend has been common across regions, it has been particularly strong for Latin America. Exports to emerging economies outside the region, particularly China, explain the bulk of this increase, with intraregional trade playing a more muted role. The increased reliance on trade with Asia and other emerging regions allowed many countries in the Latin America and Caribbean (LAC) region (particularly in South America) to bounce back more quickly from the global crisis.
LAC: Exports to Emerging and Developing Economies
Sources: IMF, Direction of Trade Statistics; and IMF staff calculations.
Increased reliance on commodities. The share of commodity exports increased from 40 percent in 2000 to 52 percent by 2008, mainly reflecting increased demand from emerging market countries. The sharp increase in exports has been driven not only by positive price effects, but also higher volumes (total export volumes increased by 35 percent between 2000–10). Moreover, a bulk of this increase reflects increased exports of metals and agricultural products, with oil playing a minor role. The dependence on commodity exports makes the region especially vulnerable to a reversal in Asia’s economic performance and a downward adjustment in global commodity prices. Increased reliance on commodities may also raise issues in terms of growth and development prospects, including Dutch disease and deindustrialization, though the empirical evidence remains inconclusive.
Little action on the intraregional trade front. On average, 20 percent of all exports are to countries within the region, well above the level observed in other emerging regions. However, despite regional integration efforts, the share of exports within Latin America and the Caribbean has remained relatively flat over the past decade. This is in stark contrast to emerging Asia—where intraregional trade rose by 40 percent since 2000 largely reflecting increased openness from China.
Trading patterns vary somewhat across the existing customs unions.
In CAPDR (which includes Central America, Panama, and the Dominican Republic), the share of intraregional trade has been on an upward trend since the late 1990s, and today is the region that trades the most within itself.
The share of trade within Mercosur (which includes Argentina, Brazil, Paraguay, and Uruguay) is up since 2002, with Brazil emerging as an increasingly important trade destination for Argentina, Paraguay, and Uruguay. However, trade within the region is well below the levels reached in the mid- to late 1990s, where exports represented 25 percent of total exports. Associate members include Bolivia, Chile, Colombia, Ecuador, and Peru. Venezuela signed a membership agreement in 2006 that has yet to be ratified.
Trade within the Andean Community of Nations (which currently includes Colombia, Bolivia, Ecuador, and Peru; Venezuela withdrew its membership in 2006) is down; exports within the regional bloc remains low, at about 8 percent of total exports.
The share of exports within Caricom (which includes The Bahamas; Barbados, Belize, ECCU, Guyana, Haiti, Jamaica, Trinidad and Tobago and Suriname) has also fallen from an average of about 20 percent in 2000 to about 15 percent currently. Exports from Trinidad and Tobago and Barbados represent over 80 percent of all intraregional exports, the bulk of which are in the form of fuel and lubricants and agricultural commodities.
LAC Export Trends: Several Perspectives, 2000–10
Sources: IMF, Direction of Trade Statistics; World Bank, World Integrated Trade Solutions Database; and IMF staff calculations.
1 Percent of total merchandise exports.
LAC Exports within Regional Trading Blocs
Sources: IMF Direction of Trade Statistics, and IMF staff calculations.
1 Includes Venezuela for full period.
2 Includes Argentina, Brazil, Paraguay, and Uruguay.
3 Data do not include the Bahamas and Haiti.
On monetary policy, countries have taken steps to raise interest rates and—in a few cases—reserve requirements, though further tightening is warranted in most cases. Policy rates in most countries are still below estimates of “neutral” rates (Figure 2.6), whereas real sector developments and the stages of their business cycles would suggest that rates should be around or above neutral levels.4 In addition, there is a need for some additional tightening to maintain (ex ante) real interest rates given predictable second-round effects from recent commodity price increases to headline inflation as well as to contain these spillovers (see Section 2.3 for a fuller discussion).5 To achieve inflation objectives, another consideration will be the stance of fiscal policy, which should avoid placing an excessive burden on monetary policy in the context of overheating risks, capital inflows, and currency appreciation pressures. To avoid undermining the credibility of inflation-targeting regimes, countries should avoid adjusting their established framework when being tested, including the target (even if that target is temporarily missed).
Figure 2.6.Output gaps have closed (or are rapidly closing) amid still accommodative policies
Sources: National authorities; and IMF staff calculations.
Fiscal policy, after being expansionary in most countries in 2010, needs to “downshift” this year, at least into a neutral gear. Last year, fiscal policy generally added to demand pressures, in the sense that real expenditures grew considerably faster than potential output (Figure 2.6). Amid closed or rapidly closing output gaps,6 the projected slowing of expenditure growth in several countries this year is welcome, and will avoid adding a further burden to monetary policy. That said, while primary balances are projected to improve this year amid strong revenue growth, real expenditure plans will in most cases preserve, rather than begin to unwind, the higher expenditure levels established last year. In the event that revenue growth were to turn out higher than projected, governments should avoid letting this translate into higher expenditure. Such discipline would also help build fiscal space to cope with adverse shocks in the future.7
The budget cuts recently approved in Brazil (1¼ percent of GDP) and Chile (about ½ percent of GDP) are steps in the right direction; it will be critical to ensure their timely implementation. In the case of Brazil, it will also be important to undertake the announced reduction in policy lending by the state-owned bank (BNDES). During the course of the year, if governments incur higher fiscal costs as a result of higher food or fuel prices, it would be best to take offsetting actions.
Exchange rate flexibility should remain an important component of the macroeconomic policy package. In the context of monetary tightening, some additional currency appreciation may occur (although it is likely that appreciation based on expected further interest rate hikes already has occurred). Such appreciation, by reorienting private demand to external suppliers, would help temper inflation. Although concerns have been raised about the possible negative impact on growth and exports of real exchange appreciation, the empirical evidence on the link between “Dutch disease” and overall growth is inconclusive (Magud and Sosa, 2010). This is particularly the case for commodity exporting countries, where the export sector is known to have relatively low labor intensity. That said, although currency appreciation is desirable on cyclical grounds, foreign exchange market intervention could continue to be part of the policy toolkit, particularly after a substantial degree of appreciation has been allowed and “one-sided bets” have subsided. Foreign exchange intervention needs also to be mindful of potentially large sterilization costs. A discussion on the modalities, effectiveness, and costs of foreign exchange intervention can be found in Chapter 3.
In the context of loose external financial conditions and strong credit growth, macroprudential policies remain an essential component of the policy reaction (see the October 2010 Regional Economic Outlook: Western Hemisphere for a fuller discussion). Many countries continue to actively use macroprudential instruments, including by raising reserve requirements, tightening foreign borrowing limits, and strengthening capital requirements on certain loan types (Box 2.2). These measures should be aimed at preserving financial sector stability and, although recognizing their effects on the macroeconomy, not be deployed as a substitute for basic macroeconomic policy adjustments. Moreover, actions on the regulatory front should be complemented by efforts to quantify their impact and to improve coordination among supervisors and regulators to monitor systemic risks. Capital controls may also be part of the policy reaction on a temporary basis, although as in the case of macroprudential policies they should not play a substitute role for fiscal and monetary policies.8
South America—The Less Financially Integrated Commodity Exporters
The less financially integrated economies of South America will continue to benefit from high commodity export prices, though differences in growth performance and outlook within this group will persist. Argentina and Paraguay are projected to record strong growth in 2011—albeit at lower rates than in 2010— reflecting also close real linkages to Brazil as well as procyclical policies. In Bolivia, output is projected to expand in line with potential. Supply-side constraints and weak investment climate will continue to constrain growth in Ecuador and Venezuela.
Highly expansionary macroeconomic policies in most countries are stimulating demand and fueling inflation (Figure 2.7). Monetary aggregates are growing at an annual rate of more than 20 percent. Real primary expenditures are projected to continue to grow well above the rate of potential output in most countries. Real credit growth is accelerating, exceeding 20 percent in some countries, adding to concerns over boom-bust dynamics. The external current account is projected to deteriorate, though it will remain in surplus in most countries buoyed by large terms-of-trade gains.
Figure 2.7.In the less financially integrated economies of South America, policies generally remain expansionary
Sources: National authorities; and IMF staff calculations.
Box 2.2.An Update on Macroprudential Policies
An increasing number of emerging economies, including from the region, continue to adopt prudential measures (see table below) in an effort to: (i) slow credit growth (by introducing speed limits or targeting specific sectors such as housing or foreign exchange); and (ii) improve the management of liquidity and credit risk. Policies introduced thus far generally have not focused on the cross-sectional dimension of systemic risk such as spillovers, or interconnectedness.
|Policy tool||Recent examples or proposals||Motivation/Objective|
|●||Countercyclical capital requirements||Basel III; Brazil (auto loans-December 2010)||Buffer ranging between 0–2.5 percent to be introduced when aggregate credit is growing too fast.|
|●||Dynamic provisioning||Bolivia (2008); Colombia (2007); Peru (2008); and Uruguay (2001)||Countercyclical tool that builds up a cushion against expected losses in good times so that they can be released in bad times.|
|●||Leverage ratios||Basel III||Constrain the leverage in the banking sector, to mitigate the risk of the destabilizing deleveraging processes; and supplement the risk-based measure with a simple, transparent, independent measure of risk.|
|●||Loan-to-value (LTVs) ratios||Canada (Mortgage market-April 2010, March/April 2011);||Regulatory limit to moderate cycles in specific sectors by limiting loan growth and leaning on asset demand.|
|●||Debt-to-income (DTIs) ratios||Korea (August 2010)||Measure to limit the leverage of borrowers and manage credit risk.|
|●||Liquidity requirements||Colombia (2008); New Zealand (2010); and Basel III||Tools to identify, measure, monitor, and/or control liquidity risk under conditions of stress.|
|●||Reserve requirements on bank deposits||Peru (January and April, 2011); Brazil (December 2010); China (January 2011); and Turkey (2009–11)||Countercyclical tool that acts as: i) speed limit on credit; ii) tool for credit allocation; and iii) complement to monetary policy to achieve macroprudential goals.|
|●||Tools to manage foreign exchange credit risk||Peru (July 2010); Uruguay||Tool to internalize foreign exchange credit risks associated with lending to unhedged borrower.|
|●||Limits to foreign exchange positions||Colombia (2007); Israel (restrictions on banks derivatives transaction-2011)||Measures to manage foreign exchange risk in on- and off-balance-sheet FX-denominated assets and liabilities. Also useful for dealing with surges in capital inflows, which may pose systemic risks to the financial system when they create “bubbles” in certain economic sectors.|
|●||Others||Brazil (tax on consumer credit-April 2011)||Curb credit expansion|
Brazil and Peru have been particularly active on this front since October 2010.
Brazil: (i) increased by 50 percentage points the risk-weighting on consumer and automobile loans depending on their loan-to-value ratio and maturity; (ii) introduced a 60 percent reserve requirement on short U.S. dollar positions; and (iii) increased the tax on consumer credit from 1.5 percent to 3.0 percent.
Peru: (i) raised by 100 basis points the implicit reserve requirement rates on domestic and foreign currency deposits, and the unremunerated portion of reserve requirements (currently 9 percent of deposits); (ii) reduced reserve requirements on external FX liabilities with maturities under 2 years (from 75 percent to 60 percent), but extended their application to credit channeled through off-shore branches of domestic financial institutions; and (iii) established limits on the net FX derivative position of banks (40 percent of capital or S/.400 million, whichever is higher).Note: This box was prepared by Camilo E. Tovar.
The withdrawal of the policy stimulus is necessary to avoid overheating and accelerating inflation in most countries. Policy frameworks need to be strengthened to insulate the economies from large fluctuations in commodity export prices, including through the adoption of fiscal rules that effectively limit spending growth during commodity price booms. Consideration should be given to gradually reducing domestic fuel subsidies, which are highly distortionary in encouraging excessive fuel consumption, costly, and typically not well targeted to the poor (if they are targeted at all). Improving the business climate to address supply bottlenecks remains critical to boost growth potential in some countries.
Mexico and Central America— Rebuilding Policy Buffers
Growth is strengthening in Mexico and Central America largely reflecting an improved U.S. outlook. In Mexico, where the output gap is closing, the task ahead involves a gradual adjustment of the policy stance and continued efforts to ensure fiscal consolidation. Central America should regain the policy space used during the crisis, though this could proceed at a somewhat slower pace should terms of trade deteriorate more sharply.
The recovery is strengthening in Mexico in line with the improved performance of the U.S. economy and more recently a pickup in domestic demand responding in part to easy external financing conditions. Output expanded by about 5½ percent during 2010; growth is projected to moderate somewhat this year to about 4½ percent, as fiscal consolidation proceeds. Mexico’s output gap is expected to continue closing apace this year, as growth remains above trend.
Given Mexico’s cyclical position, the task ahead for monetary policy involves gradually adjusting to a closing output gap. Although inflation is near the 3 percent target (and within the 2–4 percent variability range), and expectations remain well anchored, the authorities need to remain vigilant for potential second-round effects of food and fuel price inflation. Fiscal consolidation is envisaged to continue this year, helping to rebuild buffers. Consensus on a strategy to further consolidate public finances needs to be developed given the prospects of declining oil revenues and pressures from age-related spending over the medium term.
Risks to Mexico’s outlook are linked to the downside risks in the United States noted in Chapter 1. Slower U.S. growth would imply weaker demand for Mexican exports as well as lower remittances. Renewed turbulence in the euro area could have adverse effects in Mexico, in part through banking channels, but such an impact should be limited given a robust subsidiary model and a tighter regulation limiting liquidity drains.
In Central America,9 the recovery gained strength in 2010, led by a rebound in domestic demand in the context of stimulative macroeconomic policies (Figure 2.8). Exports and remittances have picked up, yet remittances remain below precrisis levels reflecting weak U.S. employment conditions (Figure 1.6). Growth has been particularly strong in the Dominican Republic and Panama, which are benefiting from strong FDI and private demand. In contrast, growth has been more anemic in El Salvador, given its close ties to the U.S. economy and fiscal consolidation efforts. Output in Central America is projected to expand by an average of about 4 percent in 2011, but rising energy prices represent a downside risk to growth given the region’s reliance on imported oil.
Figure 2.8.In Mexico, fiscal consolidation is proceeding. Meanwhile, in much of Central America, policies remain generally stimulative
Sources: National authorities; and IMF staff calculations.
With relatively small (and closing) output gaps, it continues to be appropriate for Central American countries to regain policy space, particularly by slowing the pace of government spending, which remains above potential output growth in many countries. Although measures should be adopted to shield the poor from increasing food and energy prices, these should take place within the established budget envelopes. To boost growth in the medium term, priority should be given to supply-side policies that improve business climate and strengthen competitiveness, rather than demand policies.
Rising commodity prices also present a challenge to monetary policy. Headline inflation has picked up since October 2010, as would be expected given the high weights of food and fuel in CPI baskets. So far, core inflation has been subdued, but the risk of acceleration needs to be watched carefully. Countries moving toward inflation targeting and more flexible exchange rates have kept policy rates low (though the Dominican Republic and Guatemala raised their policy rates more recently) perhaps fearing further currency appreciation pressures, and may need to raise rates more aggressively to limit second-round effects—which have traditionally been strong in these countries. For countries with limited currency flexibility, the burden falls on fiscal policy, and ensuring wage policies remain prudent.
The Caribbean—Recovery Beginning Amid High Debt
The recession has been protracted in most Caribbean countries. The region is projected to exit from recession in 2011 as the recovery in advanced economies proceeds, but rising food and fuel import prices could weaken prospects. Because high public debt is an ever-present obstacle to growth as well as stability, fiscal consolidation plans will need to proceed. Efforts will need to be made to protect the poor in that context.
The Caribbean is beginning to turn the corner after a long and deep recession. Weak external demand and high public debt levels have held back economic activity in much of the region for the last two years, which has also been adversely affected by natural disasters. Tourism is recovering gradually (Figure 1.6), with larger islands (The Bahamas, Barbados, Dominican Republic, and Jamaica) observing a faster and earlier pickup in tourist arrivals than the smaller islands of the Eastern Caribbean Currency Union.
The Caribbean economy (excluding the Dominican Republic and Haiti) is projected to expand by an average of about 2 percent in 2011, following a contraction of about ½ percent last year. The recovery is supported by a mild improvement in labor market conditions in advanced economies (critical for tourism and remittances), as fiscal consolidation proceeds in most of the region. However, rising commodity import prices present a clear downside risk for most countries (with the exception of Trinidad and Tobago, which stands to benefit from higher oil export prices). Meanwhile, reconstruction efforts in Haiti are expected to take growth above 8 percent this year (Box 2.3).
Given the high debt burden in most countries (Figure 2.9), fiscal consolidation should proceed. In this, efforts will be needed to reallocate spending to protect the poor from the impact of higher food and energy prices. Special efforts will be required to contain expenditure growth and in particular wages to strengthen overall competitiveness. As discussed in the next section, price subsidies should be avoided, particularly in the case of energy, as they tend to be costly and limit substitution. In countries where a well-functioning social safety net is not in place, temporary subsidies on staples consumed by the poor could be considered, with the fiscal cost offset by adjustments in other fiscal outlays.
Figure 2.9.In most of the Caribbean, high debt level and oil dependency will constrain recovery
Sources: National authorities; and IMF staff calculations.
The financial system remains vulnerable to shocks from cross-border financial conglomerates. The resolution of the insurance subsidiaries of the CL Financial Group is still pending, with potentially large fiscal costs. Contingent liabilities could reach 10 percent of GDP in Trinidad and Tobago, where investors are being paid in full up to a threshold, while the remaining amounts are restructured. In the ECCU, where insurance claims exceed 17 percent of GDP, the restructuring of the failed subsidiary is proceeding more slowly, and in Barbados a judicial manager is being appointed to oversee the resolution process. Regulators should aim to complete the resolution while containing fiscal costs and to continue efforts to strengthen the legislative, supervisory, and regulatory framework.
Box 2.3.Haiti: Economic Developments since the January 2010 Earthquake
Haiti is gradually recovering from the devastating earthquake of January 2010. International support is aiding reconstruction efforts, though political uncertainties and rising commodity prices are posing additional challenges.
In the aftermath of the January 2010 earthquake, the IMF provided significant support.
An initial emergency fund disbursement in the amount of US$ 110 million, two weeks after the earthquake. This was followed by the cancellation of all of Haiti’s debt to the IMF, releasing an amount of US$ 268 million to speed the reconstruction process. Similar debt relief was provided by other multilaterals (Inter-American Development Bank and World Bank) as well as bilaterals (Canada, France, Italy, and Venezuela).
A three-year US$ 65 million program was approved in July 2010, which aims at providing a framework to boost growth and combat poverty. Emergency technical assistance is being provided to restore essential state functions (Treasury and revenue administration), and maintain financial sector stability.
Since the earthquake, the macroeconomic situation has improved, reflecting the authorities’ efforts to quickly restore state institutions and prudent macroeconomic policies.
Real GDP is estimated to have contracted by about 5 percent in FY2010 (Oct-09 to Sep-10). The impact of the earthquake on growth was mitigated by resilient agricultural and manufacturing activity, and a pickup in public investment.
The fiscal deficit (excluding grants and foreign financed capital spending) increased to 5.2 percent of GDP from 4.3 percent in FY2009, reflecting higher capital spending.
The current account deficit widened somewhat, as higher reconstruction-related imports more than offset a rebound in exports and higher official transfers. However, net international reserves are up US$ 400 million since September 2009, reaching 1.1 billion at end-January 2011 owing to sizable donor inflows and debt relief.
Haiti’s economy is projected to rebound sharply in 2011, though political uncertainties, unfavorable terms of trade, and further delays in donor disbursements could complicate the recovery.
Real GDP is projected to grow by about 8½ percent in 2011, as reconstruction efforts intensify. This projection assumes an acceleration in the pace of donor disbursements, which in turn is based on the resolution of election-related uncertainties. Thus far, only one-fourth of the total US$ 5.5 billion (20 percent of GDP) pledged for 2010–11 at the New York Donor’s Conference of March 2010 has been disbursed.
The rise in global food and energy prices is pushing inflation, with possibly negative consequences on growth and the poor. The government recently froze domestic fuel prices, though it intends to partially pass-through further increases in global fuel prices. Fuel subsidies are being financed through concessional financing from Petro-Caribe, though this could pose risks for external debt sustainability.
Moreover, continued weakness in administrative capacity could hamper implementation of reconstruction efforts and programs to protect the poor. It will be critical that financial support continues to be accompanied by capacity building, to build a stronger nation less vulnerable to natural disasters and recurrent health epidemics.
Box 2.4.Caribbean Offshore Financial Centers: Opportunities and Challenges
Some Caribbean countries are considering diversifying their economies through the further development of offshore financial services. Not only do offshore financial centers (OFCs) provide employment opportunities for local labor, they can also generate spillovers to other sectors in the economy, including tourism and infrastructure, because they require improved telecommunication and transportation. Indeed the empirical evidence suggests that OFC-related portfolio flows contribute to economic growth (for more details see González, Khosa, Liu, Schipke, and Thacker, forthcoming). Currently, the Caribbean accounts for more than half of global OFC-related flows, though much of it is dominated by nonsovereign territories, such as the Cayman Islands (see figure). Although OFCs account for a relatively smaller share of GDP in sovereign Caribbean countries, in some countries (Bahamas and Barbados) their economic/fiscal contribution is significant.
OFCs in the Caribbean
|Number of Agreements|
|Apr. 09||Jan. 11||Latest Status|
|Antigua and Barbuda||7||12+||White|
|St. Kitts and Nevs||0||12+||White|
|St. Vncent and the Grenadines||0||12+||White|
|British Vrgin Islands||3||12+||White|
|U.S. Vrgin Islands||12+||12+||White|
|Turks and Caicos Islands||0||12+||White|
Given the increasingly large volume of financial flows handled by OFCs, international pressure has built in recent years to ensure these centers follow stricter prudential and supervisory financial standards, control money-laundering activities, and limit opportunities for tax evasion. These initiatives are spearheaded by a number of different global institutions, including the Global Forum on Transparency and Exchange of Information, the Financial Stability Board, and the Financial Action Task Force (with support from the IMF). A strong understanding of the details of these different initiatives is necessary to limit reputational risks (via the possibility of black/grey listing) in cases of noncompliance.
Countries and jurisdictions hosting OFCs are taking steps to demonstrate their commitment to adhere to these international standards. For example, all countries except one have signed the required Tax Information and Exchange Agreements to be moved to the “white” list by April 2011. However, continued efforts will be needed to ensure compliance with all of the initiatives, especially as they are moving toward the mutual assessment/effectiveness stages. Indeed the empirical evidence suggests that countries/territories that adopt good regulatory standards benefit from higher inflows.Note: This box was prepared by Alfred Schipke.
Improved export performance is crucial to boosting and sustaining growth in the Caribbean over the medium term. Efforts to further develop and diversify Caribbean exports are needed, including by refreshing the tourism product (by diversifying markets, for example, to South American visitors), and developing new high value-added export services to take advantage of the new and emerging growth poles, including offshore financial centers (Box 2.4). For a more in-depth discussion of growth issues, see Chapter 5 of the October 2010 Regional Economic Outlook: Western Hemisphere.
2.3. Dealing with the Oil and Food Price Shock
Rising global food and energy prices present a serious challenge to efforts to anchor inflation and are a particular hardship to the poor. Monetary policy may need to be tightened to contain higher food and fuel prices from spilling over into core inflation and inflation expectations, particularly in countries with weaker policy frameworks and track records. Steps should be taken to protect vulnerable groups from higherfood prices, ideally through proven targeted income transfers, while avoiding costly and distortive generalized price subsidies.
International commodity prices, as reviewed in Chapter 1, are back near peaks observed prior to the global crisis. Although weather-related shocks and political tensions in the Middle East and North Africa are responsible for the more recent surge in food and oil, continued strong growth in the demand for commodities is expected to keep prices near current highs (see the April 2011 World Economic Outlook). In the case of food, although futures markets indicate a gradual reduction in some prices over the course of this year (as supply shocks dissipate), prices will remain near precrisis highs.
Impact of Global Commodity Price Shocks on Inflation
The increases in domestic food and fuel prices in much of the region are pushing headline inflation upward (Figure 2.10). Central banks are increasingly concerned that the recent (possibly one-off) increase in food and fuel inflation may unhinge inflation expectations and spill over into core inflation, similar to the 2007–08 episode. Although inflation has been relatively contained thus far and remains within the target range in much of the region (in Brazil and Uruguay, it is close to or above the top of the band), core inflation and inflation expectations are slowly trending upward, and prices are likely to increase further in the months ahead as lagged pass-through effects play out.
Figure 2.10.The global commodity price shocks are gradually transmitting into higher domestic food and fuel prices, pushing headline inflation up and affecting the poor in the region, who spend a larger share of their income on food.
Sources: National authorities; Robles and others (2008); and IMF staff calculations.
To assess the potential impact on inflation of the recent commodity price shock, we reestimate what has been the typical pass-through from global to domestic food and fuel prices (first-round effects), and also the historical rate of pass-through from those domestic food and fuel prices to core prices (second-round effects) for a group of 10 countries in the region. The selected countries (Brazil, Chile, Colombia, Costa Rica, Dominican Republic, Guatemala, Mexico, Paraguay, Peru, and Uruguay) have either inflation-targeting regimes or otherwise have scope for monetary policy. It should be noted that the estimated historical passthrough coefficients are also dependent on the historical monetary policy responses and the credibility of the monetary policy regime. This is particularly important, because most of these countries have strengthened their monetary policy frameworks during the period of analysis. Annex 2.1 includes technical details of the estimation as well as country-specific results.
First-round effects: World commodity price shocks are found to have a nonnegligible impact on the corresponding food and fuel subcomponents of domestic consumer price indices (Figure 2.11). A 10 percent increase in the index of key global food prices leads to an increase in domestic food prices by an average of almost 3 percent, over an horizon of three quarters. Most of the effect arrives with a lag, but with little if any additional effect after three quarters. This pass-through tends to be somewhat larger for the poorer countries in the sample, probably reflecting the fact that food processing, distribution, and marketing typically account for a smaller share of the food cost structure. A similar average pass-through is found for fuel prices, though differences across countries are much greater, likely reflecting differences in subsidy policies. In some cases (for example, Mexico and Colombia), where fuel prices are administered, domestic prices are less sensitive to international price changes, but in countries where prices are fully market determined (for example, Chile), the pass-through is much larger (Annex 2.1).
Figure 2.11.The pass-through from global commodity price shocks varies depending on the strength of the policy framework and other structural factors.
Source: IMF staff calculations.
1 Each t represents one quarter. World prices are in local currency.
2 Includes countries with inflation-targeting regimes since early 2000 (Brazil, Chile, Colombia, Mexico, and Peru).
3 Includes countries that have adopted inflation targeting more recently (Costa Rica, Dominican Republic, Guatemala, and Uruguay), plus Paraguay.
4The impact is not statistically significantly at longer horizons.
Second-round effects: Increases in domestic food and fuel prices also tend to show up, to varying degrees, in core inflation in the region. A 10 percent increase in the domestic price of food would increase core prices by an average of 1¾ percent, though the impact is generally smaller in countries with longer inflation-targeting track records. The average pass-through from domestic fuel to core prices is much lower, roughly half that of food. The transmission to core prices is generally fast, with most of the impact typically occurring within one quarter after the shock to domestic food and fuel prices. The higher passthrough from food to core prices is likely related to the fact that food (given its higher weight in the consumption basket) plays a more important role in the wage-setting process and in influencing inflation expectations.
Combining these two effects, we find that—on average—the projected 24 percent average increase in global food prices this year could add 2½ percentage points to inflation in 2011, whereas the projected 36 percent increase in global fuel prices would add another 1½ percentage points.10 The impact would be smaller to the extent that these estimations do not fully capture the gains in monetary policy credibility that some countries have achieved to date. The impact of the global food price shock could be smaller this time around to the extent that local firms are able to absorb part of the price increase through reduced profit margins.11
Monetary Policy in the Face of Commodity Price Shocks
How should monetary policy react in response to the recent surge in commodity prices? To limit the risk that the recent increases in food and fuel prices spill over excessively into core inflation and unhinge inflation expectations, policy rates may need to be tightened in addition to what would have been necessary on cyclical grounds (see previous discussion) and to avoid a decline in ex-ante real interest rates (in light of predictable second-round effects of higher food and fuel prices on headline inflation). Monetary policy decisions should not take too much comfort from developments in core inflation to date, as it is likely to increase following persistent commodity price increases, albeit with some lag (see Box 2.5 on core inflation).
Monetary policy adjustments will need to take into account country circumstances, ensuring consistency with the policy regime. Countries with shorter monetary policy horizons and/or narrower target bands may need to tighten relatively more than those with more forward-looking policy horizons or broader bands. Moreover, the timing and extent of the policy reaction depends critically on the credibility of the policy framework. Countries with weaker frameworks and inflation track records may need to tighten earlier and more aggressively to limit the pass-through from higher food and fuel prices to core inflation and to keep inflation expectations anchored. As noted earlier, in countries with limited scope for monetary policy (dollarized economies and countries with limited exchange rate flexibility), the burden should fall on fiscal and wage policies.
Social Policies and Commodity Price Shocks: Protecting the Poor
Higher food prices will most negatively affect the poor across the region (with the exception of those rural poor who benefit from higher agricultural prices), as they typically spend a larger share of their budget on food. To protect the poor, efforts should be centered on scaling up proven social safety net programs (that is, targeted income transfers, school lunches and child nutrition programs). Generalized price subsidies (or controls) and indirect tax reductions should be avoided as these measures tend to be more costly, not well targeted and distort price setting (see the October 2008 Regional Economic Outlook: Western Hemisphere for a fuller discussion of policies in this context).
Box 2.5.The Information Value of Core Inflation
We perform a series of tests to assess the information value of core inflation in 10 Latin American economies.
Does core inflation drive headline inflation or vice versa? Using a methodology developed by Marques, Neves, and Sarmiento (2003), we assess whether core inflation is a “magnet” for headline inflation. Such assessment has a direct bearing on how useful is the chosen definition of core inflation. Here, we use the authorities’ definitions of core inflation, which in most cases excludes food, fuel, and administered prices. Of course, the results are not independent of how monetary policy is conducted; they reflect past associations rather than a deep relation.
Are differences between headline and core inflation rates temporary (that is, are they cointegrated, with unitary coefficient)? Passing this test ensures that both measures are related over the medium term and do not grow apart indefinitely.
Does headline inflation converge to core inflation? Passing this test is essential, otherwise there would be no point in knowing whether core inflation exceeds or is below headline.1 Moreover, one should also ensure that core inflation does not converge to headline (otherwise, if headline and core inflation differ, it would be unclear which one would converge to the other).
The findings, based on track records during 1996–2010, suggest that Chile, Mexico and Peru have been somewhat more successful in shielding core inflation from noncore price developments. Although differences between headline and core inflation rates are found to be temporary, in most countries it is difficult to determine, with statistical confidence, that headline inflation converges to core.
Is core inflation a good predictor of headline inflation? Preliminary findings for the region suggest that core inflation is a better predictor of headline inflation (for horizons of 12 months and 24 months) than headline inflation itself.2 The findings are consistent with those in the literature (see Crone and others, 2008); by excluding items that are particularly volatile, core inflation usually contains more information about future headline inflation. (Of course, a simple measure of core inflation alone is unlikely to give the “best” possible forecast of headline inflation, in part because it does not take into account expected developments on noncore items, nor the state of the economy.)
Measuring core inflation. Despite its widespread use, there is no generally agreed measure of core inflation. Some core measures focus on persistent inflation, stripping out any volatile or transitory shocks (Bryan and Cecchetti, 1993; and Cutler, 2001), and others exclude exogenous or supply-driven price shocks. The common practice of excluding most or all food and fuel prices is generally consistent with both, as those prices are typically volatile and sensitive to world markets. However, this division of the CPI is not a clean one: for example, many “core” goods and services are likely to have components that rely on fuel inputs (for example, transportation services). Moreover, most food items in the CPI have a large element of domestic labor inputs associated with processing, and are not simply agricultural raw materials.Note: This box was prepared by Alexander Klemm.1 The test on whether headline inflation converges to core inflation is implemented by a t-test on γ in:
The lack of fiscal space in some countries (mainly in Central America and the Caribbean) and cyclical considerations in others (mainly South America) will constrain the design of income policies. In this context, in much of the region, the fiscal cost of measures to protect the poor will need to be offset elsewhere in the budget, again highlighting the value of targeting. In countries where capacity constraints prevent the scaling up of safety net programs, subsidies on certain food items could be temporarily considered. However, fuel subsidies should be avoided, as these tend to be 4–5 times more costly to the budget, disproportionately benefit the rich, and discourage energy conservation.12
In the case of food-exporting countries, supply-side measures to boost agricultural production could be helpful over the medium term. Export taxes or restrictions should be avoided as they distort production incentives, reducing future supply and, moreover, have negative spillovers at the global level. In food importing countries (Central America and the Caribbean), steps could be taken to further liberalize trade, by lifting nontariff restrictions to imports, or reducing import tariffs on food items—especially if part of a broader trade reform that enhances overall economic efficiency and does not compromise fiscal sustainability.
2.4. Credit Markets and Asset Prices—Are Bubbly Conditions Taking Hold?13
Strong terms of trade and capital inflows are boosting credit and asset prices in many countries of the region. Although commonly used metrics do not yet suggest the presence of credit booms or clear evidence of asset bubbles, financial oversight needs to be strengthened and monitoring extended to include the corporate sector where indebtedness is also growing. Further consideration should be given to adopting prudential measures to prevent excessive proyclicality of credit, and information systems require strengthening to detect systemic risks in the housing and corporate sectors
Bank Credit Growth
Bank credit is accelerating in much of the region, raising concerns as to whether this credit expansion is becoming excessive and eventually unsustainable. This is particularly the case in countries benefiting from improved terms of trade and strong capital inflows, where bank credit growth has been fastest. Though commonly used metrics suggest that the current expansion does “not yet” rise to the level of a credit boom, it would if the expansion were sustained for a prolonged period.14 Credit levels are currently at or above their underlying trend, though deviations appear to be still below danger thresholds (Figure 2.12 and Panel 2.1). However, it is worth noting that standard methods and thresholds to identify credit booms are somewhat arbitrary, and that a more holistic approach is required, particularly since the estimated trend may also be growing too fast.
Figure 2.12.Credit growth has picked up, but remains below both precrisis levels and danger thresholds
Sources: National authorities; and IMF staff calculations.
Credit developments have differed somewhat across sectors. Corporate credit—which accounts for the largest share of total bank credit in most countries—and consumer credit experienced the sharpest growth rate decline during the crisis, yet are recovering fast. Mortgage credit, which unlike advanced economies did not suffer much during the crisis, continues to expand at a very fast pace, especially in Brazil, Colombia, and Peru.15 Though the mortgage market is still relatively small in these countries (below 4 percent of GDP, compared with 20 percent in Chile), monitoring of the housing sector needs to be strengthened (see section on housing prices below).
Banks in the region remain sound and standard financial sector indicators have improved some since the crisis—capital adequacy ratios are up from already comfortable levels, nonperforming loans returned to a downward trend, and bank profitability recovered somewhat. However, the rapid rebound in credit growth has been associated with a small increase in the loan-to-deposit ratio and liabilities to nonresidents. Although funding ratios remain generally healthy (bank foreign liabilities do not exceed 10 percent of total liabilities), authorities need to remain vigilant, as nonresident liabilities are typically a more volatile source of funding and bank assets tend to be overvalued in good times (Figure 2.13).
Figure 2.13.Financial sector remains sound though credit expansion has been funded more recently from external sources
Sources: National authorities; and IMF staff calculations.
As discussed in the October 2010 Regional Economic Outlook: Western Hemisphere, much of the expansion of bank credit since 2005 could be attributed to improved fundamentals and a process of financial deepening (credit has been growing on average faster in countries that started from lower credit-to-GDP levels). However, policymakers should not be complacent, particularly given the region’s long history of credit boom-bust cycles and the current easy global financing conditions. The main challenge is to avoid the excessive procyclicality of credit that could end later in a financial collapse. Supervision needs to be further strengthened, monitoring and perhaps regulation of nonbank intermediaries needs to be stepped up, and prudential regulation upgraded to mitigate credit excesses and asset price bubbles.
Cheap financing conditions, both external and internal, are fueling corporate sector borrowing outside the banking system. Total issuance of securities is up relative to precrisis highs, led by increased corporate and quasi-sovereign borrowing (Figure 2.14).
Figure 2.14.Nonbank corporate borrowing has surged, surpassing precrisis levels. Though leverage remains in check, vulnerabilities are building.
Sources: Dealogic; Worldscope; and IMF staff calculations.
1 Includes Argentina, Brazil, Chile, Colombia, Mexico, Peru, and Venezuela.
2 Emerging economies average includes Argentina, Brazil, Chile, China, Colombia, Egypt, Hungary, India, Indonesia, Malaysia, Mexico, the Philippines, Poland, Russia, South Africa, Sri Lanka, Taiwan Province of China, Thailand, and Turkey.
Although the region’s corporate sector leverage is still below the precrisis peak, and is broadly in line with that in other emerging market regions (except for Brazil, where debt-to-equity ratios have been higher), several trends are a source of concern. A greater share of corporate financing is increasingly in the form of foreign bond placements, reversing the trend observed prior to the crisis and with the potential of increasing currency mismatch risk. Moreover, leverage ratios can be a misleading measure of corporate vulnerabilities, particularly in the context of low global interest rates and increasing equity prices. A fuller discussion can be found in the April 2011 Global Financial Stability Report.16
Monitoring of the corporate sector needs to be strengthened, particularly in the context of easy financing conditions. The U.S. financial crisis showed the importance of pockets of lightly regulated sectors—the shadow banking system in that case. Even if prudential measures can dampen the cycle-amplifying effect of the financial sector, firms could bypass the domestic financial system, and be prone to overleveraging and to excessive currency and maturity mismatches. Improving information systems to monitor the financing structure of firms remains a key priority (see Cubeddu and Tovar, 2011). And although the region has made progress in compiling and disseminating firm balance sheet data, relatively little information is available on their debt structure (currency and maturity) and exposure to derivative instruments. Shortcomings of this nature were brought to light in Brazil and Mexico in 2008, when firms experienced severe losses from operations in the foreign exchange derivative market.17
Equity prices in the region rebounded sharply after the global crisis and in most countries are above precrisis levels, outperforming other emerging markets (Figure 1.2). Although systemic bubbles are not clearly evident and are difficult to measure, stock prices are currently above trend levels in most countries, with signs of stretched valuations in a few countries (Chile, Colombia, and Peru), where price-to-earnings ratios are well above historical averages and levels observed in other emerging markets (Figure 2.15 and Panel 2.1).
Figure 2.15.Equity prices recovered sharply, with signs of stretched valuations in a few countries.
Sources: Datastream; and IMF staff calculations.
1Shows 95th percentile and 5th percentile for the period from January 2001 to March 2011. Gray box covers range between 25th and 75th percentile, and red dots are latest observation.
2 Weighted average for Argentina, Brazil, Chile, Colombia, Mexico, Peru, and Venezuela.
The surge in commodity prices is attracting new investments and pushing up valuations of commodity-related firms (Box 2.6). Prices of basic material firms have increased sharply since 2005; their price-to-earnings ratios were relatively stable until more recently (early 2010) when they started to move up sharply. It is worth noting that part of the increase in these ratios is of a temporary nature, because earnings from new investments in oil and minerals extraction take time to come to fruition. That said, authorities should remain vigilant and consider providing their evaluation to help markets in their “price discovery” process.18
Despite the growing importance of mortgage credit, data on housing indicators remain scant in the region. Currently, and in contrast to other regions in which data availability is somewhat greater, only four countries in Latin America produce data on housing prices (Colombia, Mexico, Peru, and Uruguay), and many of these series are of short duration, with varying coverage and methodologies. Real housing prices in these four countries have increased by an annual average rate of 10 percent since 2005, with the crisis having had only a small and short-lived impact on values. Home price inflation has been somewhat faster than in emerging Asia, yet far below that observed in emerging Europe in the years leading up to the crisis (Figure 2.16).
Figure 2.16.Mortgage credit is expanding at a fast pace, and housing prices have been increasing somewhat faster than in emerging Asia.
Sources: Bank for International Settlements; national authorities; and IMF staff calculations.
The absence of systematic housing price series in the region hampers risk monitoring and assessment and, by extension, timely decision making to avoid excesses related to the real estate sector. Improving information on the mortgage market, where credit has been growing fast, is of critical importance. Housing price indicators should be developed and relationships between construction firms and the banking sector closely monitored. Consideration could be given to tightening loan-to-value ratios, a measure which generally seems to have worked well in Asia’s emerging markets.
2.5. Capital Inflows—Guarding Against the Reversal of Favorable Global Conditions19
Net capital flows to Latin America experienced a fast rebound after the collapse in the wake of the global financial crisis, driven by strong portfolio flows. This trend of strong net flows is expected to continue as long as the current conditions of low global interest rates, low risk aversion, and a relatively weak outlook in advanced economies persist. However, changes in these global conditions, including a tightening in U.S. monetary policy, may be associated with a larger reduction of capital inflows to the region, relative to other emerging economies given Latin America’s larger direct financial exposure to the United States.
Net capital flows to Latin America recovered rapidly after their retrenchment at the wake of the global financial crisis, driven by low global interest rates, improved fundamentals, and stronger growth prospects in the region. Compared with other emerging market economies, the post–global-crisis rise in net flows (in percent of GDP) was smaller in Latin America, but these flows recovered from a higher base. Net flows in the first three quarters of 2010 were already above the 2004–07 average, although below that observed during 1991–96, when the region financed larger current account deficits (Figure 2.17).
Figure 2.17.Despite rebounding quickly following the crisis, capital flows to the region are sensitive to changes in global conditions.
Sources: CEIC; IMF, Balance of Payments Statistics; national sources; and IMF staff calculations.
1The Latin America group includes: Argentina, Brazil, Chile, Colombia, Ecuador, El Salvador, Guatemala, Mexico, Paraguay, Peru, and Uruguay. See Appendix 4.1 of the April 2011 World Economic Outlook for the list of economies included in the other emerging market economies group.
2 1991–94 corresponds to a period of low global rates. 1996 presents a period of low VIX and high growth.
The recovery in net flows to Latin America during the first three quarters of 2010 was largely driven by debt creating flows, while the share of FDI flows has fallen. The recent composition of net flows is similar to that of 1991–97, when more than half of the flows was debt creating. This trend warrants special attention from policymakers, given evidence that portfolio debt flows are among the least persistent.
The Spring 2011 World Economic Outlook shows that net capital flows to emerging economies, particularly non-FDI flows, are sensitive to favorable global conditions (low interest rates and low risk aversion and large growth differentials between emerging and advanced economies). In Latin America, the dynamics of net inflows around these episodes are driven by portfolio debt flows, while in other emerging economies, bank and other private flows play a bigger role. Additionally, the findings in the Spring 2011 World Economic Outlook suggest that net inflows to Latin America can be more sensitive to changes in U.S. monetary policy than in other emerging regions, in part reflecting closer financial ties to the United States.20
A summary of these results follows:
Latin America’s financial exposure to the United States (average 25 percent) is larger than that of other emerging economies (17 percent).21
For a country with a financial exposure to the United States equal to that of the average Latin American economy, a one standard deviation unanticipated rise in the U.S. real interest rate— about 5 basis points—would reduce net inflows relative to the average emerging market economy by an additional 0.2 percentage point of GDP in the first quarter, and by a cumulative 1.0 percentage point of GDP after two years.
The sharper response in Latin America could reflect not only the higher financial exposure to the United States, but also the larger share of portfolio debt flows in net flows compared with other emerging economies, which are generally more sensitive to U.S. monetary policy changes. The high sensitivity of inflows to changes in global conditions suggests that concerns about the duration and stability of capital flows in the recent recovery are not unreasonable, particularly for Latin America. In fact, the recent slowdown of portfolio and equity inflows to emerging markets and the region (likely resulting from positive surprises in the U.S. outlook, inflation risks, and higher asset prices in emerging markets), reinforces the need to prepare against a sudden reversal of global conditions.
This concern is particularly relevant for countries in the region that already have current account deficits. Their macroeconomic policies will need to adjust to ensure a moderation of domestic demand growth, before current account deficits become too large. To limit bubbles and excessive risk taking, prudential measures should form part of the policy response. In some circumstances, capital account restrictions could be needed also, though these cannot substitute for basic macroeconomic policies. Particularly in light of Latin America’s high degree of financial openness, amid a continuation of unusually easy external financing conditions, a temporary use of some capital account restrictions could be part of a package of policies that prudently seeks to prevent a boom-bust cycle. For a more detailed discussion on policy challenges and options for dealing with capital inflows in the Latin American context, see Eyzaguirre and others (2011), as well as Ostry and others (2010 and 2011), and IMF (2011c).
Box 2.6.Stock Market Developments in the Region: The Role of Commodities
Equity markets in Latin America have been undergoing an important transformation. Stock market capitalization has grown from an average of 25 percent of GDP in 2000 to about 75 percent by end-2010. The deepening of equity markets has been fairly broad-based, though the increase in capitalization has been more muted in Mexico. Today Brazil is home of the fourth largest stock market in the world, and represents about 55 percent of the region’s equity market.
Stock Market Capitalization
Sources: Datastream; and IMF staff calculations.
1 Commodities include oil and gas and basic materials.
2 Simple average for Brazil, Chile, Colombia, Mexico, and Peru.
3 Simple average for China, India, Indonesia, Malaysia, Taiwan Province of China, Thailand, Singapore, South Korea, and the Philippines.
4 Simple average for Czech Republic, Hungary, Poland, Russia, and Turkey.
The deepening of equity markets has taken place in tandem with the expansion of the commodity sector. The share of commodity-related firms has more than doubled from an average of 15 percent in the early 2000s to close to 35 percent by end-2010. The share of finance firms has also increased (with the opening of the sector), whereas telecommunication and utility sectors have lost ground in most countries.Note: This box was prepared by Leandro Medina.
Panel 2.1.Credit and Asset Prices in Selected Latin American Economies: Deviations from Trend
Sources: National authorities; and IMF staff calculations.
1 Computed using a rolling, backward-looking Hodrick-Prescott filter for the period January 2000 to December 2010.
2 Threshold is equivalent to 1.5 times the standard deviation of credit fluctuations around the estimated trend.
Sources: Datastream; and IMF staff calculations.
1 Using a Hodrick-Prescott filter for the period January 1993 to December 2010.
2 Simple average.
This annex describes the methodology used to estimate pass-throughs from commodity price shocks. The analysis is done using quarterly data for the period 1996:Q1 to 2010:Q4, and on a sample of 10 countries in the Latin America and Caribbean region with either inflation-targeting regimes or otherwise with scope for monetary policy. These include: Brazil, Chile, Colombia, Costa Rica, Dominican Republic, Guatemala, Mexico, Paraguay, Peru, and Uruguay.
Food and Fuel Pass-Through
Step 1: The pass-through from international to domestic prices of food and fuel is estimated using country-by-country regressions of this form:
where Πf stands for the quarter-over-quarter log difference in food or fuel prices, respectively (the equations also include seasonal dummies). World prices are expressed in domestic currency.
We are interested in the impact of a one-period increase in the world food/fuel inflation rate (that is, a permanent increase in the price level). The immediate impact on domestic food and fuel inflation would be given by γ0, whereas the (not cumulated) impact j periods after would be:
Based on these coefficients, the pass-through from an increase in world food and fuel prices into domestic prices is computed for different time horizons, as presented in the previous section.
Step 2: The pass-through from domestic food and fuel to core inflation is estimated for each country using a simple Phillips curve specification:
where y andy* are quarterly, seasonally adjusted GDP and potential GDP in logs (seasonal dummies are also included). The immediate impact and the impact after j periods are calculated as under step 1 using δi and ϕi.22
Annex 2. Figure 1.Pass-Through from Global Commodity Price Shock to Domestic Prices
Source: IMF staff calculations.
1The impact is not statistically significant at longer horizons.
Note: This chapter was prepared by Luis Cubeddu and Sebastiàn Sosa, with contributions from Jaime Guajardo and Alex Klemm.
Import growth moderated slightly in recent months, yet remains strong (30 percent year over year in January 2011).
The projected private sector current account deficit for 2011 (1 percent of GDP) is still below the average observed during 1993–97 (2½ percent of GDP).
In the case of Brazil, public banks have also played a key role in the credit expansion (Chapter 3 of the Fall 2010 Regional Economic Outlook).
Given difficulties in determining neutral rates, IMF staff estimates a range using different approaches.
As of early April, surveys suggest that markets expect policy rates to increase by an additional 50 bps in Brazil and 200 bps in Chile over the next 12 months; expectations of U.S. monetary policy tightening during that period, if any, are generally much lower.
Real primary expenditure growth is generally a good proxy of changes in a country5 s fiscal stance, though a more complete picture requires assessing changes in revenue policy and administration.
In Colombia, the projected increase in public spending reflects in part reconstruction costs following floods in late 2010.
In October 2010, Brazil increased the tax on external inflows from 4 percent to 6 percent to discourage investments in the local fixed-income market. In March/April 2011, it also raised to 6 percent the tax on external short-term borrowing by firms and increased the maturity threshold from 90 days to 720 days.
Analysis includes the Dominican Republic and Panama.
The impact would be smaller in inflation-targeting countries, with the projected increases in global food and fuel prices adding 1½ percentage points and 1 percentage point to inflation, respectively.
Recent inflation reports for Chile and Peru make this argument, because domestic food prices did not decline as sharply as global food prices in the aftermath of the global crisis.
During the 2007–08 commodity price boom fuel subsidies in a large sample of low- and middle-income countries averaged about 1½ percent of GDP compared with ¼ percent in the case of food (IMF, 2008a).
The assessment focuses on credit and asset price developments in six of the more financially integrated countries in the region: Brazil, Chile, Colombia, Mexico, Peru, and Uruguay.
We consider here a definition of credit boom already used in the literature. A given credit expansion is identified as a boom if (i) the level of credit exceeds the underlying trend (estimated using a rolling, backward-looking Hodrick-Prescott filter) by a threshold equal to 1.5 times the standard deviation of credit fluctuations around trend (the exercise was also conducted using the credit-to-GDP ratio instead of credit levels); or (ii) the ratio of credit to GDP exceeds the underlying trend by more than 18 percent relative to trend (or, alternatively, if the absolute deviation between actual and trend credit-to-GDP ratio exceeds 4 percentage points of GDP). These approaches as well as the rationale for these thresholds are described in Mendoza and Terrones (2008) and Gourinchas, Valdés, and Landerretche (2001). The data frequency in those studies (annual) differs from the data frequency in this chapter (monthly).
In Brazil, for example, real annual mortgage credit growth exceeded 40 percent, more than tripling the level of early 2007.
Financing costs are down sharply. More than 40 percent of all corporate bonds in U.S. dollars have a yield to maturity of below 6 percent compared with below 20 percent in the period 2006–08.
Measures were adopted in 2009 requiring financial institutions to report their exposures to derivative contracts (Brazil) and firms that issue bonds or equity to document their market, credit, and liquidity risks related to financial derivative contracts (Mexico).
See the Financial Stability Report of Chile (December 2010) and Colombia (September 2010).
Jaime Guajardo of the IMF Research Department contributed to this section.
This analysis is based on a panel data set of 20 advanced and 30 emerging economies for the period 1989–2010, and analyzes whether differences in direct financial exposure to the United States affects the sensitivity of their net capital inflows to U.S. monetary policy changes (after controlling for country-fixed effects, time-fixed effects, and domestic drivers).
The direct financial exposure to the United States is defined by the share of its total foreign assets and liabilities that are U.S. related.
The pass-through from domestic food and fuel prices to core prices was also estimated using predicted values of domestic food and fuel inflation from the first step regressions. This implies that domestic food and fuel inflation reflect only variations owing to changes in international prices and own lagged effects, rather than changes in labor, transportation, and distribution costs that are explained by factors that also drive overall inflation. The results, however, do not change significantly.