Chapter 11. Housing Markets in Latin America: Do We Need to Worry About a Bubble?

Dora Iakova, Luis Cubeddu, Gustavo Adler, and Sebastian Sosa
Published Date:
December 2014
  • ShareShare
Show Summary Details
Luis Cubeddu, Camilo E. Tovar and Evridiki Tsounta This chapter updates an earlier work published as Cubeddu, Tovar, and Tsounta (2012).

Easy financing conditions and favorable terms of trade have fueled credit and domestic demand in much of Latin America for more than a decade, with only a short interruption during the 2008–09 global crisis (Figure 11.1). The credit expansion has been particularly impressive for the mortgage sector, where legal reforms and government subsidies have also played a role. This strong growth—from generally low mortgage credit levels—to a large extent reflects a process of financial deepening necessary to address the significant housing deficits in the region (Figure 11.2).1

Figure 11.1Latin America: Private Sector Credit by Market Segment

(Index, 2007 = 100)

Sources: National authorities; and IMF staff calculations.

Note: Shows private sector credit in real terms. Simple average of Brazil, Chile, Colombia, Mexico, Peru, and Uruguay. Data through April 2014.

Figure 11.2Emerging Market Economies: Mortgage Credit, 2013

(In percent of GDP)

Sources: National authorities; and IMF staff calculations.

However, given the region’s long history of credit booms gone wrong, there are valid concerns about the potential buildup of financial sector excesses, even if current credit indicators appear manageable. Experience shows that credit-driven bubbles build slowly but can sour quickly. Indeed, even though housing market crashes have been rare in the region, Colombia’s experience in the late 1990s is a useful reminder of the systemic effects that even a small mortgage sector can have on the economy. Moreover, as seen in the recent U.S. housing crisis, problems in a small market (for example, the subprime sector) can become systemic, especially in new markets with significant data gaps.

The increase in mortgage credit in many countries of the region has been accompanied by an increase in home prices. In fact, the average real home price for the more financially integrated economies of the region (Brazil, Chile, Colombia, Mexico, Peru, and Uruguay) rose at an annual rate of over 7½ percent between 2009 and 2013. While the average price increase is well below that in emerging Europe in the run-up to the global financial crisis, it is somewhat above that in emerging Asia during the same time period.

This chapter documents developments in the housing and mortgage markets in Latin America during the past decade, comparing them to those of other emerging market economies. In addition, and despite serious data limitations, the chapter assesses whether (1) growth in mortgage credit is excessive compared to its long-term trend; (2) trends in house prices reflect changes in economic fundamentals; and (3) the extent to which vulnerabilities may be building in the household and banking sectors.

Relying on standard statistical and econometric techniques, the analysis finds little evidence of excessive growth in mortgage credit in much of the region. In Brazil, while mortgage credit has been growing very rapidly in recent years (supported by the expansion of government-sponsored housing credit programs), the mortgage-credit-to-GDP ratio is still low relative to that of other emerging market economies. Similarly, while there is evidence of minor house price misalignments in a few countries of the region (Peru and, to a lesser extent, Brazil), it cannot be concluded that there is a bubble in the real estate market. However, excesses and vulnerabilities are not always captured by contemporaneous indicators, only becoming visible if the current rates of mortgage credit and house price growth are sustained for an extended period.

To broaden the assessment of vulnerabilities, this chapter also examines the exposure of household and bank balance sheets to the real estate sector. The analysis finds that the number of nonperforming mortgage loans is still relatively low, and that mortgages continue to represent a small share of banks’ assets. Similarly, household indebtedness indicators (where available) suggest that financial burdens remain at manageable levels, although they are rising, especially for low-income households.

The overall findings should be interpreted with caution, since a proper assessment of the mortgage credit and housing situation is hindered by the fairly limited and weak information available for the real estate sector.2 House price data are only available for some countries—notably the largest Latin American economies—and even when available, time series are usually of short span and coverage is often limited to large metropolitan areas. In addition, there is little information on the stock and flows of housing and construction activity, as well as on housing-specific financial soundness indicators and household balance sheets. Addressing these data gaps remains an urgent priority in the region.

This chapter first documents recent developments in the mortgage credit and real estate sector in Latin America, then highlights the data gaps that hinder a comprehensive assessment of the risks and vulnerabilities of the real estate sector. Notwithstanding these constraints, the chapter then provides an assessment of misalignments in the levels of mortgage credit and house prices, followed by a discussion on the state of household and financial balance sheets in the region, policy recommendations, and conclusions.

Developments in the Mortgage Credit and Real Estate Sector

Mortgage Credit

During the past decade, much of Latin America has experienced an unprecedented expansion in overall credit, and mortgage credit in particular. Favorable external conditions, sustained economic growth, stronger fundamentals, and legal reforms have raised living standards and improved financing conditions, helping to unleash housing finance. Real mortgage credit in the more financially integrated economies of the region (Brazil, Chile, Colombia, Mexico, Peru, and Uruguay) grew by an annual average of 12 percent between 2003 and early 2014, and was less affected by the global financial crisis than other sectors (such as consumption and corporate credit). Growth in mortgage credit has been particularly strong in Brazil, where the inflation-adjusted stock of mortgage loans increased nine-fold since 2003, albeit from a low base (Figure 11.3).

Figure 11.3Real Mortgage Credit and Housing Prices

(12-month percent change, average from 2010-14:Q1)

Sources: Haver Analytics; national authorities; and IMF staff calculations.

Note: All data expressed in local currency and deflated by the consumer price index.

1 Data end in 2012 for house prices.

Structural reforms in property and credit markets as well as government efforts to broaden access to credit have been critical. For example, both Brazil (2005) and Mexico (2007) enacted bankruptcy reforms to strengthen creditor rights3 and overhauled their credit registries to enable banks to better gauge the creditworthiness of debtors.4 In addition, Brazil (where three-quarters of all housing credit has been provided by state-owned banks) relied extensively on mortgage credit subsidies, while Mexico provided mortgage insurance and guarantees through a government agency to support the residential mortgage-backed securities market (Scatigna and Tovar, 2007). The expansion of mortgage credit in Latin America has also gone hand-in-hand with the development of the domestic bond market in local currency and the lengthening of the term structure of yield curves, which in some cases has reached 20–30 years (Jeanneau and Tovar, 2008).5

Despite the rapid credit growth in recent years, however, financial intermediation levels in most of Latin America (with the notable exception of Chile) remain well below those of other emerging regions, even after accounting for income per capita. By end-2013, mortgage credit in Latin America averaged 7 percent of GDP (17 percent of total credit), well below the levels in other emerging market economies (18 percent of GDP and 29 percent of total credit). This gap reflects the shallowness of the credit market in general, the region’s long history of macroeconomic instability (often associated with credit booms and busts), and weak institutions, in particular those related to creditor and property rights (Cottarelli and others, 2005, Table 11.1).6

Table 11.1Data Availability on Select Financial and Housing Sector Indicators
Housing IndicatorsHousehold IndicatorsFinancial Soundness Indicators1Access to Credit
House Prices2HouseholdCommercial
CountryAvailable SinceFrequencyCoverageHousing Starts/PermitsConstruction Cost IndexHousehold Debt to IncomeDebt Service to IncomeReal Estate Loans to Total LoansMortgages to Total LoansNonperforming Mortgages3Legal Rights (0-10)4Credit Information (0-6)5
United States1987monthlynational96
Latin America
Memorandum items:
Emerging Asia
Emerging Europe
Sources: Eurostat; Global Property Guide; Haver Analytics; IMF, Financial Soundness Indicators (FSI); and World Bank, 2012Doing Business Indicators.

Construction Activity

Rising mortgage credit has gone hand-in-hand with the expansion of the construction sector. The share of construction in GDP has grown sharply during the past decade in the more financially integrated countries of Latin America, reaching levels above those of emerging Asia, yet remaining below the peaks observed in emerging Europe prior to the Lehman crisis (Figure 11.4).7 In Brazil, the number of construction companies and projects is well above 2010 levels, despite some recent correction (SECOVI, 2012), and employment in the construction sector has grown by over 13 percent since 2009, with the share of construction in total employment reaching 7.7 percent by end-2013. Similarly, in Colombia building permits almost doubled between 2009 and 2013, reflecting in part an expansion of government housing-related subsidies, including through the extension of loans and residential leasing for “social interest” (vivienda de interés social) housing programs, and the introduction in 2013 of an interest subsidy program targeting middle-income earners (IMF, 2014).

Figure 11.4Emerging Market Economies: Size of Construction Sector

(Percent of GDP)

Sources: Haver Analytics; and IMF staff calculations.

Note: EM = emerging market.

1 Average of China, Hong Kong SAR, Indonesia, Korea, Malaysia, and Thailand.

2 Average of Bulgaria, Lithuania, Russia, Slovak Republic, and Slovenia.

3 Average of Brazil, Chile, Colombia, Mexico, and Peru.

House Prices

The increase in mortgage credit and construction activity has coincided with a significant increase in house prices in many Latin American countries. The average home price in the six more financially integrated economies in the region rose by an annual real rate of more than 7½ percent between 2010 and early 2014. While the average price increase is well below that in emerging Europe in the run-up to the Lehman crisis, it is somewhat above that in emerging Asia. However, in level terms Latin America’s house prices remain low by international comparison.

In particular, according to the Global Property Guide (2012), home property prices in some metropolitan areas of the region generally do not appear out of line when measured relative to the country’s income per capita as well as rental prices (Figure 11.5), although they are approaching frothy levels in a few countries (Brazil, Chile, Colombia, and Peru).8

Figure 11.5Price-to-Rent Ratio, 2013

(Rent years to buy 120 m2 property in metropolitan center)

Sources: Global Property Guide; and IMF staff calculations.

Note: EM = Emerging market.

1 EM Asia includes China, India, Indonesia, Malaysia, Philippines, Singapore, Thailand. EM Europe includes Bulgaria, Croatia, Czech Republic, Estonia, Hungary, Latvia, Lithuania, Moldova, Poland, Romania, Russia, Serbia, Slovak Republic, Slovenia, Turkey, and Ukraine.

However, these measures need to be interpreted with caution since they only capture real estate prices in the higher-income segments of metropolitan areas, and do not necessarily reflect the situation at the national level and across the income distribution.9 The next section further examines the question of real estate overvaluation.

Making the Most of Imperfect Data

Assessing risks and vulnerabilities in Latin America’s real estate sector are hindered by weak and limited information. Data on housing prices and real estate activity remain scant in the region, despite recent progress. Only Brazil, Chile, Colombia, Mexico, Peru, and Uruguay publish housing price data, but in some instances these time series’ have short spans, and coverage is often limited to large metropolitan areas.10 Moreover, in several instances the data do not distinguish between new and existing homes, as well as between commercial and residential real estate. Little information is available on the stock and flows of housing, or on construction activity (including employment, prices of inputs, and land prices) (Table 11.1). Appropriate data on housing transactions are also limited. Moreover, an additional challenge is posed by the fact that trends can vary depending on whether prices are measured in local or foreign currency.11

Notwithstanding severe shortcomings in data, policymakers and analysts in the region still need to assess whether the pace of expansion of mortgage credit is excessive and whether movements in house prices can be explained by economic fundamentals. Despite the limitations, this section engages in this quixotic endeavor and tackles these questions with the data available.

Identifying Mortgage Booms

A credit expansion is identified here as a boom when the level of credit exceeds the underlying trend—estimated using end-adjusted rolling Hodrick Prescott filters—by a threshold equal to 1.5 times the standard deviation of the trend as used by Mendoza and Terrones (2008) and Gourinchas, Valdés, and Landerretche (2001). The analysis is implemented for a sample of nine countries (Brazil, Colombia, China, Hong Kong SAR, Indonesia, Korea, Malaysia, Peru, and Thailand) using monthly data between January 2000 and March 2014. Mortgage bank credit data are available from national sources, and include the mortgage claims on the private sector by deposit money banks as well as public finance corporations (where available).

The deviation from the long-term trend in the logarithm of real mortgage credit in country i, date t, is denoted as Mit, and the corresponding standard deviation of this cyclical component as σ(Mi). Country i is defined as having experienced a credit boom when Mit ≥ 1.5σ(Mi), that is, when the deviations from trend in mortgage credit exceed the typical expansion of credit over the business cycle by a factor of 1.5 or more. The long-term trend is calculated using the Hodrick-Prescott (HP) filter with the smoothing parameter set at 129,600, as is typical for monthly data. The filter is rolled month-over-month with the seed set in January 2006; the seed was adjusted accordingly for countries with data limitations. In other words, this expanding trend extends the sample over which the trend is computed by one month as each successive month in the sample is added.

The analysis suggests that mortgage growth is above levels dictated by past long-term trends in a few economies in the region, notably Brazil (Figure 11.6). In Brazil, the maximum deviation from trend credit growth in recent years seems particularly large, with moderate deviations recorded in other countries. However, the results should be viewed with caution, since the technique used here does not capture changes in trend credit growth or structural breaks that might arise due to the adoption of particular policies. In that regard, the significant increase in Brazilian mortgage credit coincides with the introduction and subsequent expansion of the housing program (Minha Casa, Minha Vida) aimed at low-income households.12 Despite these caveats, the rapid expansion of mortgage credit still warrants careful monitoring of this market segment. Moreover, it is important to ensure that credit risks remain under control, an issue that will be discussed later in the chapter.

Figure 11.6Episodes of Real Credit Surges

(Maximum deviation since mid-2009, percent)

Source: IMF staff calculations based on national sources.

Note: Estimates based on end-adjusted rolling Hodrick-Prescott filters estimated using monthly data since 2000 when available. The smoothing parameter, λ, was set at 129,600, and the filter was rolled month-over-month with the seed set in January 2006. In countries where sample size was an issue, the seed was adjusted accordingly. The threshold is defined as 1.5 times the standard deviation of the level relative to trend.

1 Average of sample that includes China, Hong Kong SAR, Indonesia, Korea, Malaysia, and Thailand.

Identifying Housing Price Booms

An econometric model is constructed to investigate the existence of house price bubbles for the six more financially integrated economies of Latin America. The idea is to determine the extent to which the recent performance of house prices can be explained by economic fundamentals. Much like Tumbarello and Wang (2010) and Tsounta (2009), we determine the existence of cointegration relationships to uncover the long-term relationship between real house prices and their fundamentals with variables expressed in levels.13

House prices and income (as well as other fundamentals) are found to be cointegrated (Cubeddu, Tovar, and Tsounta, 2012), suggesting that the gap between actual prices and estimated prices, estimated using an error correction model, may be a useful indicator of when house prices are above or below their equilibrium values.14 The model we estimate consists of two to four cointegrating I(1) variables, real house prices, Pt, real interest rate, Rt, real per capita GDP, Yt, and population, St, depending on the country considered. All variables are in logarithms except for the real interest rate, which is in levels. The error-correction equation for Pt, in its entirety is:

where 0 < α < 1, is the error correction term, β and λ are estimated parameters, and Δ is the difference operator. Using similar vector error correction models for Brazil, Chile, Colombia, Mexico, Peru, and Uruguay, we find real GDP per capita, population, and the real lending rate—a proxy for the real mortgage rate—to be important determinants of equilibrium prices in all six countries (Table 11.2).15 Moreover, the coefficients are generally statistically significant and of the right sign, although the importance of each variable in explaining price movements differs significantly across countries. Our stylized model has the following implications:

Table 11.2.Determinants of Equilibrium House Prices: Selected Latin American Countries
Sample period2011: Q2-2014:Q12006: Q3-2013:Q41998: Q1-2013:Q42006: Q1-2014:Q12001: Q3-2013:Q42002: Q1-2012:Q4
Total population2.375.21
Real GDP per capita1.621.271.910.120.61
Real mortgage rate−0.91−0.1−0.41−1.54
Error correction−0.19−0.47−0.2−0.52−0.31−0.1
Source: IMF staff calculations.Note: Quarterly data. All variables in the vector error autoregression model are in log levels with the exception of the real mortgage interest rate. T-statistics in brackets.
  • In the long term, a 1 percent increase in population—a proxy for the formation of households—will raise the equilibrium house price by about 2 percent in Chile and 5 percent in Peru.16

  • Real GDP per capita—a proxy for households’ purchasing power and borrowing capacity—has a significant positive effect on house prices with an elasticity ranging from 0.1(Mexico) to around 2.0 (Colombia). These estimates are in line with other findings in the literature as summarized in Iossifov, Čihák, and Shanghavi (2008).

  • The real mortgage rate that affects households’ ability to borrow also has a negative and generally statistically significant impact on house prices in Brazil, Chile, Mexico, and Peru; in the long run, a 1 percentage point increase in the interest rate will lead to a fall in house prices of 0.1 percent in Chile and 1.5 percent in Peru.17 These coefficients are broadly in line with other estimates in the literature, although there is a large dispersion ranging from –0.9 for the Netherlands (Hofman, 2005) to –6 for the United Kingdom (Hunt, 2005).

  • The negative sign of the coefficient of the error correction term for all six countries suggests that, indeed, the system is correcting back to its long-term equilibrium, with a pace of about one-half of the disequilibrium per quarter in Mexico and Chile and a much slower pace in the remaining sample countries.

In addition, the econometric results allow us to conclude that:

  • House price dynamics in Latin America can mostly be explained by the basic stylized model of economic fundamentals (IMF, 2011b).

  • Prices remain aligned with fundamentals and within a one-standard deviation range from the trend in most countries (Figures 11.7 and 11.8). The most dynamic market is Peru, where house prices currently deviate by almost 10 percent from levels dictated by economic fundamentals (and above the one-standard deviation). The misalignment was much greater in 2012–13. Nonetheless, Peru’s price-to-rent ratio is relatively low by regional and international comparison, suggesting that, if any, signs of overvaluation are modest.18 Brazilian house prices are found to be at the upper bound of the deviation distribution. It is also worth noting that the recent run-up in house prices in some countries (notably Uruguay and Colombia) appears to reflect some catching up from undervalued house prices in the mid-2000s. House prices in Chile and Mexico appear to be the most aligned with economic fundamentals, though there was some misalignment in mid-2000s in Chile.

  • Our stylized model captures the large bubble experienced in Colombia in the mid-1990s that led to the mortgage crisis (see Box 11.1). House prices are estimated to have exceeded the trend by as much as 60 percent in the mid-1990s, only returned to levels dictated by fundamentals in the mid-2000s. In Uruguay, house prices undershot economic fundamentals following the crisis in the early 2000s and are now approaching levels dictated by fundamentals.

Figure 11.7Selected Latin America: Actual and Estimated Real House Prices

Source: IMF staff calculations.

Figure 11.8Selected Latin America: House Price Over/(under)valuation

(Percent deviation from actual)
(Percent deviation from actual)

Source: IMF staff calculations.

A drawback to this analysis is the considerable uncertainty about the right technique to model equilibrium house prices, including the possible biases that may arise due to model specification (for example, failing to capture macroeconomic volatility or inward migration) and the ensuing “omitted variable bias” or unstable estimated relationships. However, this is intrinsic to most of techniques found in the literature (Gallin, 2003; Gurkaynak, 2005; Kluyev, 2008; Girouard and others, 2006; Tsounta, 2009; Allen and others, 2006; and IMF, 2004, Box 2.1). As mentioned previously, the relatively short time series for data on prices and its limited coverage also constrain the analysis.

The Household Debt Burden and Financial Balance Sheets

Financial stability concerns related to the fast growth in mortgage credit and real estate prices are counterbalanced by the relatively low exposure of banks in the mortgage market, the small share of nonperforming mortgages, the strength of household balance sheets, and sound prudential practices. In particular recent data suggest that:

  • Mortgages account for less than 20 percent of banks’ total credit in many countries in Latin America, and banks have a sound funding structure that relies little on cross-border funding or complex instruments.

  • In line with the trends for aggregate bank credit quality in the region, the share of nonperforming mortgage loans is relatively low, averaging about 2–3 percent of total mortgage credit for the more financially integrated economies.19

  • The few existing household indebtedness indicators suggest that the average debt burden for most countries remains at manageable levels, although debt burdens have been on the rise in recent years (Figure 11.9).

  • Some countries in the region have a history of implementing prudential measures for the mortgage market. Colombia, for example, introduced loan-to-value (LTVs) and debt-to-income (DTIs) limits for mortgage borrowers in 1999, following the housing crisis (see Box 11.1). As of end-2013, mortgage LTVs averaged about 55 percent and mortgage debt service could not exceed 30 percent of disposable income (IMF, 2014).

Figure 11.9Household Debt in Emerging Market Economies

(Percent of GDP)

Sources: IMF, International Financial Statistics; and IMF staff calculations.

Note: Country groupings are those used in the IMF’s World Economic Outlook.

EMDE = emerging market and developing economies.

1 Including China.

2 Including Russia.

Box 11.1Colombia’s Mortgage Crisis of the Late 1990s: A Cautionary Tale1

The Colombian mortgage crisis of the late 1990s illustrates the possible systemic effects of problems in the housing market. The crisis had its origins in the early 1990s, when a process of financial deregulation set the stage for unsustainable credit growth and asset price overvaluation (Figure 11.1.1) amidst weak regulatory and supervisory frameworks.

Figure 11.1.1Colombia: Selected Economic and Credit Indicators

Sources: Banco de la Republica (Colombia); DANE; FOGAFIN (2009); and IMF staff calculations.

Reforms aimed at increasing competition and efficiency in the Colombian financial system in the early 1990s led to rapid expansion of bank assets along with undesirable changes in their liability structure. A period of easy external financing conditions triggered massive capital inflows that were intermediated by the domestic financial system. Asset prices (including housing prices) rose quickly and credit boomed (bank credit as a share of GDP doubled between 1991 and 1997). At the same time, financial institutions adopted aggressive funding practices in an environment of increased competition.

Weak regulatory and supervisory systems and internal risk management models made the financial system vulnerable. Internal risk models were ill-suited for assessing borrowers’ capacity to pay, and collateral was frequently overvalued. Weak regulation and supervision practices did not prompt an increase in capital requirements or loan-loss provisions to mitigate growing risks as the process unfolded. In addition, there were important informational blind spots that prevented the adequate assessment of risks.

When external conditions became less favorable and the economy began to slow in 1995, housing prices started to fall. This process was compounded by the sudden stop of external financing that followed the Asian and Russian crises and domestic political problems. By 1998 interest rates reached historical highs, and households found themselves unable to continue servicing mortgages. Properties were seized, nonperforming loans skyrocketed, and banks specializing in mortgage lending became illiquid or insolvent.

In 1999, the government was forced to intervene. Financial institutions were nationalized, closed, or recapitalized, and Colombia suffered its first recession since 1933. The crisis also set back the development of the nation’s housing market. In the end, the total fiscal cost of the crisis—including the effects of judicial rulings that further undermined creditors’ rights—exceeded 15 percent of GDP (FOGAFIN, 2009).

1 For a detailed overview of the Colombian mortgage crisis, see FOGAFIN (2009) and Urrutia and Llano (2011).

However, one cannot draw comfort from recent data trends, since the situation could quickly change, particularly should GDP growth slow and financial conditions tighten sharply. Credit quality measures tend to be a lagging indicator of financial distress, while household leverage measures are often understated during periods of strong income growth and record low unemployment. Moreover, many loans are new, and defaults are typically rare early in the life of a loan. It also seems that banks are extending loans to households with unknown credit and payment histories; such households tend to be more vulnerable during downturns.

Conclusions and Policy Recommendations

This chapter documented developments in mortgage credit and the housing sector in Latin America during the past decade and compared them with those in other emerging market economies. Although data limitations hamper a more rigorous analysis of trends, vulnerabilities, and risks, the analysis found few signs of misalignments in the mortgage and real estate sector. Moreover, house prices in most markets are not far from equilibrium levels. Thus, although the analysis does not envision any immediate vulnerability, if left unattended the current credit and house price growth levels could result in dangerous misalignments down the road. Housing price corrections tend to have significant economic and social ramifications, particularly where the share of employment in construction, which employs relatively unskilled workers, has grown and is larger than justified by fundamentals.

Action is needed to close information gaps by developing more comprehensive and timely information on home prices (with national coverage, distinguishing between new and existing homes and between commercial and residential real estate, and measured in terms of repeat sales) and housing construction activity (housing stock and flows, employment by sector, and prices of construction inputs, including land prices). Supervisory authorities should continue to strengthen the infrastructure needed to maintain current information on housing-specific financial soundness indicators and household balance sheet data.20 The latter are critical for assessing credit risks based on leverage ratios and meaningful affordability indicators. Progress on this issue also lags behind that in other regions.

Better data would also facilitate stronger oversight of the overall situation. Internal risk models to gauge misalignments in housing and other credit sectors can help with risk assessments. Further efforts are needed to improve property rights,21 credit registries, and the underwriting standards of mortgage loan originators and brokers, especially given the rising importance of the securitization market that is in its very early stages in some markets. Standards should take into account the value of the underlying property (based on sound independent appraisals) and the borrower’s creditworthiness (via credit registries), with proper verification of the submitted information. These efforts should be complemented by measures to strengthen creditor rights and programs to increase consumer financial literacy, particularly as credit access expands to lower-income households.22 Attention should also be given to the financial implications associated with the rapid expansion of new forms of housing financing (for example, trusts) in some markets, and to mortgage securitization.

Finally, if vigor in the housing sector is sustained, targeted macroprudential measures similar to those recently adopted in Asian economies should be considered. In this context, the use of loan-to-value and debt-to-income limits could be particularly useful to dampen credit and house price growth; the limits may need to be lower for emerging market economies, as they tend to suffer deeper recessions with more severe financial downturns than advanced economies (Claessens, Kose, and Terrones, 2010, 2011). The adoption and implementation of LTV and DTI limits might be particularly challenging in some Latin American countries. LTV limits are less effective where a larger share of mortgage origination is outside the regulated sector, and where household debt and income are difficult to verify.


    Agnello, L., and L.Schuknecht,2009, “Booms and Busts in Housing Markets: Determinants and Implications,ECB Working Paper No. 1071. Frankfurt: European Central Bank.

    Allen, J., R.Amano, D. P.Byrne, and A. W.Gregory,2006, “Canadian City Housing Prices and Urban Market Segmentation,Working Paper 2006–49. Ottawa: Bank of Canada.

    Case, Karl E., and Robert J.Shiller,1989, “The Efficiency of the Market for Single-Family Homes,American Economic Review79(1): 12537.

    Central Bank of Chile, 2009, “Encuesta financiera de hogares: EFH 2007 metodología y principales resultados,Santiago.

    Claessens, Stijn, M. AyhanKose, and Marco E.Terrones,2011, “How Do Business and Financial Cycles Interact?Working Paper No. 11/88, International Monetary Fund, Washington, DC.

    Claessens, Stijn, M. AyhanKose, and Marco E.Terrones,2010, “Financial Cycles: What? How? When?” In NBER International Seminar in Macroeconomics 2010, edited by RichardClarida and FrancescoGiavazzi. Cambridge, Massachusetts: National Bureau of Economic Research.

    Cottarelli, Carlo, GiovanniDell’Ariccia, and IvannaVladkova-Hollar,2005, “Early Birds, Late Risers, and Sleeping Beauties: Bank Credit Growth to the Private Sector in Central and Eastern Europe and in the Balkans,Journal of Banking and Finance29(1): 83104.

    Cubeddu, Luis, Camilo E.Tovar, and EvridikiTsounta,2011, “Latin America: Vulnerabilities Under Construction?Working Paper 12/193, International Monetary Fund, Washington, DC.

    Davis, Morris, AndreasLehnert, and Robert F.Martin,2008, “The Rent-Price Ratio for the Aggregate Stock of Owner-Occupied Housing,Review of Income and Wealth54(2): 279328.

    Detken, Carsten, and LuciaAlessi,2009, “Real Time Early Warning Indicators for Costly Asset Price Boom/Bust Cycles—A Role for Global Liquidity,ECB Working Paper No. 1039. Frankfurt: European Central Bank.

    Égert, Balasz, and DubravkoMihaljek,2012, “Determinants of House Prices in Central and Eastern Europe,BIS Working Paper 236. Basel: Bank for International Settlements.

    Égert, Balasz, PeterBacké, and TinaZumer,2006, “Private Credit in Central and Eastern Europe: New (Over) Shooting Stars?ECB Working Paper 687. Frankfurt: European Central Bank.

    Fondo de Garantías de Instituciones Financieras (FOGAFIN), 2009, “Crisis financiera colombiana en los años noventa: origen, resolución y lecciones institucionales,FOGAFIN and Universidad Externado de Colombia, Bogotá.

    Gallin, Joshua,2003, “The Long-term Relationship between House Prices and Incomes: Evidence from Local Housing Markets,Finance and Economics Discussion Series No. 2003–17, Board of Governors of the Federal Reserve System, Washington, DC.

    Garcia, Pablo,2009, “The Changing Face of Housing Finance in Chile,paper presented at the Latin America Housing Finance Summit sponsored by Euromoney and the International Finance Corporation, 19–20February, Miami.

    Girouard, Nathalie, MikeKennedy, Paulvan den Noord, and ChristopheAndré,2006, “Recent House Price Developments: The Role of Fundamentals,Economics Department Working Paper No. 475. Paris: OECD Publishing.

    Global Property Guide, 2012, “Country

    Gourinchas, Pierre-Olivier, RodrigoValdés, and OscarLanderretche,2001, “Lending Booms: Latin America and the World,CEPR Discussion Paper No. 2811. London: Centre for Economic Policy Research.

    Gurkaynak, Refet S.,2005, “Econometric Tests of Asset Price Bubbles: Taking Stock,Finance and Economics Discussion Series No. 2005–04, Board of Governors of the Federal Reserve System, Washington, DC.

    Hoek-Smit, Marja C., and Douglas B.Diamond,2003, “The Design and Implementation of Subsidies for Housing Finance,paper presented at the World Bank Seminar on Housing Finance, March10–13, Washington, DC.

    Hofman, David,2005, “House Prices in the Netherlands,IMF Country Report No. 05/255, International Monetary Fund, Washington, DC.

    Hunt, Benjamin,2005, “How Should Policymakers Respond to a Decline in House Prices,United Kingdom: Selected Issues, IMF Country Report No. 05/81, International Monetary Fund, Washington, DC.

    Hunt, Benjamin, KhoiNguyen, and PatriziaTumbarello,2009, “Australian Household Vulnerabilities,Australia: Selected Issues, IMF Country Report No. 09/249, International Monetary Fund, Washington, DC.

    International Monetary Fund (IMF), 2004, World Economic Outlook, World Economic and Financial Surveys, Washington, DC, September.

    International Monetary Fund (IMF), 2011a, Global Financial Stability Report: Durable Financial Stability, Getting There from Here, World Economic and Financial Surveys, Washington, DC, April.

    International Monetary Fund (IMF), 2011b, Regional Economic Outlook—Western Hemisphere, Watching Out for Overheating, World Economic and Financial Surveys, Washington, DC, April.

    International Monetary Fund (IMF), 2011c, “Uruguay: 2011 Article IV Consultation,IMF Country Report No. 11/375, Washington, DC, December.

    International Monetary Fund (IMF), 2014, “Colombia: Selected Issues Paper,IMF Country Report No. SM/14/119, Washington, DC, May.

    Iossifov, Plamen, MartinČihák, and AmarShanghavi,2008, “Interest Rate Elasticity of Residential Housing Prices,Working Paper 08/247, International Monetary Fund, Washington, DC.

    Jeanneau, Serge, and Camilo E.Tovar,2008, “Latin America’s Local Currency Bond Markets: An Overview,” in New Financing Trends in Latin America: A Bumpy Road towards Stability, BIS Paper No. 36. Basel: Bank for International Settlements.

    Jha, Abhas K.,2007, “Low-income Housing in Latin America and the Caribbean.En Breve No. 101. Washington, DC: World Bank.

    Kaminsky, Graciela, and CarmenReinhart,1999, “The Twin Crises: The Causes of Banking and Balance-of-Payments Problems,American Economic Review89(3): 473500.

    Klyuev, Vladimir,2008, “What Goes Up Must Come Down? House Price Dynamics in the United States,Working Paper No. 08/187, International Monetary Fund, Washington, DC.

    Mendoza, Enrique G., and Marco E.Terrones,2008, “An Anatomy of Credit Booms: Evidence from Macro Aggregates and Micro Data,NBER Working Paper No. 1404. Cambridge, Massachusetts: National Bureau of Economic Research.

    Morandé, Felipe G., and CarlosGarcia,2004, “Financiamiento de la Vivienda en Chile,Research Department Working Paper No. 4357. Washington, DC: Inter-American Development Bank.

    Persson, Mattias,2009, “Household Indebtedness in Sweden and Implications for Financial Stability—The Use of Household-Level Data,BIS Paper No. 46. Basel: Bank for International Settlements.

    São Paulo State Housing Syndicate (SECOVI), 2012, Industry Data. Available via Haver Analytics and at

    Scatigna, Michela, and Camilo E.Tovar,2007, “Securitisation in Latin America,BIS Quarterly Review (September), Bank for International Settlements, Basel.

    Tsounta, Evridiki,2009, “Is the Canadian Housing Market Overvalued? A Post-Crisis Assessment,Working Paper 09/235, International Monetary Fund, Washington, DC.

    Tumbarello, Patrizia, and ShengzuWang,2010, “What Drives House Prices in Australia? A Cross-Country Approach,Working Paper No. 10/291, International Monetary Fund, Washington, DC.

    UN Habitat, 2011, Affordable Land and Housing in Latin America and the Caribbean, Adequate Housing Series,” Vol. 1, United Nations Human Settlements Program.

    UN Habitat, various years, Housing Finance Mechanisms.

    Urrutia, Miguel, and JorgeLlano,2011, “Los actores de la crisis económica de fin de siglo,Desarrollo y Sociedad, Primer Semestre, Universidad de los Andes, Bogota.

    Warnock, Veronica C., and Francis E.Warnock,2008, “Markets and Housing Finance,Journal of Housing Economics17(3): 23951.

    World Bank, 2012, “Doing Business Indicators,World Bank, Washington, DC.

The Economic Commission for Latin America and the Caribbean estimates the housing deficit between 42 and 51 million units. Meanwhile, the Ministerial Commission on Housing and Urbanization for Latin America and the Caribbean found that only 60 percent of families in the region had adequate housing in 2007. For country-specific data on the housing deficit see UN Habitat (2011, Chapter 3).

Although data limitations hamper a more rigorous analysis of trends and risks, the findings shed some light of the degree of exuberance and potential spillovers of these markets (Agnello and Schuknecht, 2009; Detken and Alessi, 2009).

Since its inception in 2000, the insolvency law in Mexico has been used on very few occasions.

Warnock and Warnock (2008) show that housing finance is positively correlated with the enforceability of legal rights, as well as with the existence of systems to assess credit risk.

See Garcia (2009) for a discussion of housing finance developments in Chile, including the impact of changes in pension fund regulations on mortgage finance.

According to Jha (2007), less than a quarter of all housing in Latin America is financed through formal mechanisms, with the exception of Chile, where mortgage financing represents around half of all house purchases (Morandé and Garcia, 2004; Central Bank of Chile, 2009). Similarly, in Mexico, between 1980 and 2003 more than half of all constructed housing units were built by the households themselves; less than 20 percent of these were built with formal financing (UN Habitat, 2011).

However, housing is not the only factor behind this rising share; commercial real estate and public works also contributed to the rise.

For a discussion of the significance of the house-price-to-rent ratio as a measure of overvaluation, see Davis, Lehnert, and Martin (2008).

Hoek-Smith and Diamond (2003) find that in Brazil and Mexico only households around the 70th income percentile qualified for mortgage financing in early 2000; in Peru the corresponding number was at the 65th percentile. In a series entitled “Housing Finance Mechanisms,” UN Habitat (various years) discusses, among other issues, mortgage access constraints in select emerging market economies.

Housing prices in the region are often published by a wide variety of agencies, and using different methodologies, which makes cross-country comparisons more cumbersome.

A clear example of this is Uruguay, where the assessments may vary depending on the currency in which the analysis is performed. See IMF (2011c).

The Minha Casa, Minha Vida program was introduced to reduce the housing deficit and inequality gap. Under the program, the authorities are planning to build 3.4 million new houses (in partnership with the states, municipalities, and private sector) to be allocated to families on a means-tested basis. As of April 2014, 2.4 million houses had been built.

For a detailed description of the data and sources for house prices, see Cubeddu, Tovar, and Tsounta (2012).

Vector error correction models are preferable to single equation models with variables expressed in percentage changes (as in Hunt and others, 2009; IMF, 2004, Box 2.1) because they allow the estimation of equilibrium values.

The specification varies slightly for each country since a common specification did not yield a cointegration relationship for each country.

Egert and Mihaljecjk (2007) show that the average long-term elasticity of house prices to the working-age population share is close to 4 for 19 countries in the Organisation for Economic Co-operation and Development.

This coefficient expresses the long-term relationship between real house prices and mortgage interest rates and should not be used to gauge the short-term impact of interest rate changes on house prices.

Moreover, the price data for Peru capture only trends in the higher-income areas of Metropolitan Lima, and it is not clear whether the same price dynamics can be replicated in other areas of the capital or the rest of the country.

Although not discussed in this chapter, mortgage credit quality can have important implications for fiscal policy depending on the degree of exposure of public banks.

Brazil, Chile, and Colombia have household financial or expenditure surveys aimed at collecting micro data on households’ real and financial assets. The data provide a picture of credit access and debt concentration among different household income segments, improving the assessment of credit risk and the implications of household debt for financial stability (Persson, 2009).

According to UN Habitat (2011), more than a third of Latin American homeowners may have tenure that falls short of full legal title, with informal housing estimated to constitute anywhere between 25 and 50 percent of the urban housing stock in Latin American countries (concentration of informality ranges significantly across cities ranging from 10 percent in Buenos Aires to 50 percent in Quito and Caracas).

Efforts should be made to reduce the costs, duration, and effectiveness of enforcement and foreclosure processes in the event a borrower defaults. The security of collateral and a relative lack of borrower credit history information have been cited as areas of weakness in the region (UN Habitat, 2011). Warnock and Warnock (2008) find that the size of housing finance systems worldwide—including Latin America—are positively correlated with the enforceability of legal rights relating to foreclosures.

    Other Resources Citing This Publication