Portal de Periódicos da CAPES

Financial Cycles: What? How? When?

2011; University of Chicago Press; Volume: 7; Issue: 1 Linguagem: Inglês

10.1086/658308

2150-8372

Stijn Claessens, M. Ayhan Köse, Marco E. Terrones,

Monetary Policy and Economic Impact

Previous articleNext article FreeFinancial Cycles: What? How? When?Stijn Claessens, M. Ayhan Kose, and Marco E. TerronesStijn ClaessensInternational Monetary Fund, University of Amsterdam, and CEPR Search for more articles by this author , M. Ayhan KoseInternational Monetary Fund Search for more articles by this author , and Marco E. TerronesInternational Monetary Fund Search for more articles by this author PDFPDF PLUSFull Text Add to favoritesDownload CitationTrack CitationsPermissionsReprints Share onFacebookTwitterLinkedInRedditEmailPrint SectionsMoreI. IntroductionA short history of economic developments over the past two decades vividly shows that gyrations in financial markets have greatly influenced real activity around the world. Following the largest housing bubble of modern history, Japan experienced a massive asset market crash in the early 1990s that marked the start of its "Lost Decade." After prolonged credit booms, many emerging countries in Asia faced major financial crises in the second half of the 1990s. The equity market booms of the late 1990s in a number of advanced economies ended with synchronized busts and cyclical downturns. Many countries enjoyed credit and housing booms over 2003–7 as the global economy registered its best performance of the past four decades. However, these episodes also ended with severe financial disruptions in the form of credit crunches and asset price busts and resulted in the deepest financial crisis since the Great Depression. Not surprisingly, understanding financial cycles has become an important research area.The objective of this paper is to provide a comprehensive empirical overview of financial cycles. We ask three specific questions. First, what are the main features of financial cycles? Second, how synchronized are financial cycles within and across countries? Third, when do the coincidences of cycles lead to magnified financial outcomes? In order to answer these questions, we employ an extensive database of cycles in credit, house prices, and equity prices for a large number of advanced countries over a long period.Although there is a rich literature analyzing various aspects of financial market developments, our understanding of financial cycles is still limited.1 This reflects the simple fact that most of the literature considers only selected aspects of financial cycles. For example, a number of studies examine the implications of only booms in asset prices and credit rather than considering full cycles in these markets. Others focus on financial crises—in many respects only the extreme versions of the downturn phases of cycles.We extend the literature on financial cycles in a number of dimensions. First, we provide the first detailed, cross-country empirical analysis documenting the main features of financial cycles and the interactions across different cyclical phases using a large sample. Second, in parallel with the business cycle literature, we use a well-established and reproducible methodology for the dating of financial downturns and upturns. Furthermore, since we employ quarterly data rather than the annual data typically used in other cross-country studies, we can better identify and document the properties of financial cycles. Third, taking advantage of our large data set and using regression models, we study various factors associated with the duration and amplitude of financial cycles.Section II presents our database and methodology. The data set includes 21 "advanced" OECD countries and covers the period 1960:1–2007:4. It is easy to provide a qualitative characterization of financial cycles.2 However, this is not very useful for our purpose since our objective is to study cycles across a large number of countries over an extended time period. To identify cycles in a systematic way and present a quantitative characterization, we employ a methodology widely used in determining the turning points of business cycles for advanced countries. This allows us to create a chronology of financial cycles following the tradition of Burns and Mitchell (1946), who provided the fundamental approach for the study of U.S. business cycles. Specifically, we rely on their "classical" definition of a cycle since it provides a simple but effective way to identify turning points. Using this methodology, we determine the dates of upturns and downturns and identify more than 470 financial cycles. To study how financial cycles have evolved over time, possibly as a result of ongoing globalization, we divide our sample into two distinct periods: the preglobalization period (1960–85) and the globalization period (1986–2007).To answer the first question, Section III documents the main features of financial cycles. It highlights three facts. First, downturns tend to feature sharp declines in short periods, lasting about 5–8 quarters, whereas upturns are often much longer and slower. Second, equity and house price cycles tend to be longer and more pronounced than credit cycles. Third, there have been changes in the features of cycles over time; in particular, equity price cycles have become shorter.We answer the second question by analyzing the extent of synchronization of cycles within and across countries. Results indicate that financial cycles are closely but not perfectly correlated with each other. Cycles in credit and house prices appear to be the most highly synchronized within countries. The degree of synchronization across countries is the highest for credit and equity cycles and has been increasing over time.In Section IV, we study the implications of the coincidence of financial cycles. We find that there are indeed feedback effects between house price and credit cycles as disruptions in one market aggravate the problems of the other, probably because of collateral constraints and complementarities between credit and housing finance. When housing downturns are accompanied by financial crises, downturns tend to become longer and deeper. Globally synchronized financial downturns also result in longer and deeper episodes, especially for credit and equity cycles.These results set the stage for the more formal empirical analysis in Section V, where we employ various regression models to analyze the roles played by a wide range of factors in explaining the duration and amplitude of cycles. We find positive duration dependence for the downturn phase of financial cycles, implying that the longer a downturn has gone on, the more likely it is to end. The regression results also confirm the presence of feedback effects between financial cycles, even after controlling for other potential factors. Section VI concludes with a brief discussion of results and directions for future research.II. Database and MethodologyDatabase. We construct an extensive data set using quarterly series of financial variables for 21 advanced OECD countries covering the period 1960:1–2007:4.3 To study how cycles have evolved over time, we divide our sample into two distinct periods: the preglobalization period (1960–85) and the globalization period (1986–2007). We use 1985 as the demarcation for three reasons. First, global trade and financial flows have increased markedly since the mid-1980s. Second, the earlier period witnessed a number of common shocks associated with sharp fluctuations in the price of oil in the 1970s and common contractionary monetary policies in major advanced economies in the early 1980s. Third, the beginning of the globalization period coincides with a structural decline in the volatility of business cycles in advanced countries. Until the financial crisis erupted in mid-2008, the second period had come to be known as the period of the "Great Moderation" because of the prolonged decline in the volatility of output accompanied by relatively low and stable levels of inflation.4We concentrate on cycles in three distinct market segments, which together constitute the core of financial intermediation. Specifically, we study cycles in credit, housing, and equity markets. Our measure of credit is aggregate claims on the private sector by deposit money banks. Credit is a natural aggregate to analyze financial cycles since it is the single most important link between savings and investment. Our measure has often been used in earlier cross-country studies on credit dynamics (see Mendoza and Terrones 2008).5 The two other variables we use are house and equity prices. House price series correspond to various measures of indices of house or land prices depending on the source country. Equity prices are share price indices weighted with the market value of outstanding shares. Both of these asset price measures have been employed in earlier studies.6Credit series are collected from the Institute for Fiscal Studies (IFS), house price series are mostly from the OECD, and equity prices are from the IFS and Datastream. All series are seasonally adjusted whenever necessary and are in constant prices. In addition to these variables, we use some other variables in our formal empirical analysis (see Claessens et al. [2011] for details on sources for these variables).Methodology. In order to identify financial cycles, we borrow methods widely employed in the business cycle literature. In particular, we use the "classical" definition of a business cycle that provides a simple but effective procedure to identify turning points. The definition goes back to the pioneering work of Burns and Mitchell (1946), who laid the methodological foundation for the analysis of business cycles in the United States.7Our methodology focuses on changes in levels of variables. An alternative methodology considers how a variable fluctuates around its trend and then identifies a "financial cycle" as a deviation from this trend. Our objective here, however, is to produce a well-defined chronology of financial cycles rather than study the second moments of fluctuations.8 Another advantage of using the classical methodology is that the turning points identified are robust to the inclusion of newly available data: in other methodologies, the addition of new data can affect the estimated trend and thus the identification of a cycle (see Canova 1998). Furthermore, the classical method is easier to apply in a cross-country context.The specific cycle-dating algorithm we use is the one introduced by Harding and Pagan (2002b), which extends the so-called BB algorithm developed by Bry and Boschan (1971) to identify the turning points in the log level of a series.9 It requires a search for maxima and minima over a given period of time. Then it selects pairs of adjacent, locally absolute maxima and minima that meet certain censoring rules. In particular, we require the duration of a complete cycle to be at least 5 quarters and of each phase to be at least 2 quarters. Specifically, a peak in a quarterly financial series ft occurs at time t if Similarly, a cyclical trough occurs at time t ifIt is useful to draw some parallels between the phases of financial cycles and those of business cycles. A complete business cycle comprises two phases: the contraction or recession phase (from peak to trough) and the expansion phase (from trough to the next peak). In addition to these two phases, recoveries from recessions have been widely studied (see Eckstein and Sinai 1986). The recovery phase is the early part of the expansion phase and is usually defined as the time it takes for output to rebound from its trough to the peak level just before the latest decline. Some others associate recovery with the growth achieved within a certain time period, such as 4–6 quarters, following the trough (see Sichel 1994). Given their complementary nature, we use both definitions of recovery in our analysis of financial cycles below.Our characterization of financial cycles closely follows that of business cycles. We call the recovery phase of a financial cycle the "upturn" and the contraction phase the "downturn." These two phases of financial cycles provide rather well-defined time windows. We do not study expansions, which are typically much longer and can be affected by many structural factors (e.g., the level of the country's legal and institutional development greatly affects the scope for financial development) and initial conditions (e.g., the initial depth of the country's financial system has a substantial impact on the scope for long expansions in credit).Compared to the financial crisis literature, our approach has some clear advantages in terms of dating of events. For one, in parallel with the business cycle literature, we use a well-established and reproducible methodology for dating, whereas crisis dating is based on historical records and is often subjective, especially for banking crises (in many cases, ending dates are selected in an ad hoc manner). In a related way, we consider financial events that are not necessarily crises yet did create stress in some markets with potentially severe macroeconomic consequences. In addition, we consider three types of financial events, allowing us to investigate different cycles and evaluate the interactions across them, whereas a financial crisis dummy often lumps them together.Defining the features of financial cycles. The main characteristics of cyclical phases are their duration, amplitude, and slope. The duration of a downturn, Dc, is the number of quarters, k, between a peak and the next trough. Likewise, the duration of an upturn, Du, is the number of quarters it takes for a variable ft to reach its previous peak after the trough. The amplitude of a downturn, Ac, measures the change in ft from a peak (f0) to the next trough (fk), that is, Ac = fk − f0. The amplitude of an upturn, Au, measures the change in ft from a trough (fk) to the level reached in the first 4 quarters of an expansion (fk+4), that is, Au = fk+4 − fk. Finally, the slope of a financial downturn is the ratio of the amplitude to the duration of the downturn. The slope of an upturn is the ratio of the change of a variable from the trough to the quarter at which it attains its last peak divided by the duration. Thus, the slope measures the violence of a given cyclical phase.To examine the extent of synchronization across financial cycles, we use the concordance index developed by Harding and Pagan (2002a). The concordance index for variables x and y, CIxy, over period t = 1, …, T, is defined aswhereIn other words, and change depending on the phase of the cycle. The concordance index provides a measure of the fraction of time the two series are in the same phase of their respective cycles. The series are perfectly procyclical (countercyclical) if the index is equal to unity (zero).We also study the more intense forms of financial cycles, disruptions and booms, and their implications. To identify these, we rank the changes in each variable during downturns and upturns. We then classify an episode as a financial disruption (boom) if the change in the variable during the downturn (upturn) falls into the bottom (top) quartile of all changes. We call disruptions crunches or busts depending on the variable (i.e., credit crunch, house, or equity price bust). Similarly, we have credit, house, and equity price booms.III. What Are the Main Features of Financial Cycles?A. Frequency, Duration, Amplitude, and SlopeFrequency. We identify 473 complete financial cycles over the period 1960:1–2007:4. In particular, our full sample features 114 downturns in credit, 114 in house prices, and 245 equity prices (table 1). Conversely, the full sample includes 115, 114, and 251 upturns in credit, house, and equity prices, respectively. Since equity prices are more volatile than credit and house prices, they naturally feature more downturns and upturns than the others. Financial cycles are more frequent in the preglobalization period (1960–85) than in the globalization period (1986–2007). In the case of credit, for example, the number of downturns (up) in the first period is 69 (67) whereas it is only 45 (48) in the second.10 The sample of equity cycles is roughly equally divided over the two periods.Table 1. Financial Cycles: Frequency DownturnsUpturns NumberTime in DownturnNumberTime in UpturnCredit: Full period114.30115.20 [.30] [.23] 1960–8569.2967.24* [.21] [.23**] 1986–200745.3048.16 [.24] [.10]House price: Full period114.41114.31 [.40] [.32] 1960–8558.54***53.35 [.55***] [.33] 1986–200756.3461.33 [.28] [.33]Equity price: Full period245.45251.38 [.44] [.39] 1960–85128.55***131.40 [.57***] [.40] 1986–2007117.36120.40 [.34] [.40]Note: For the statistics time in downturn and time in upturn, means are shown with medians in brackets. Time in upturn (downturn) refers to the ratio of the number of quarters in which the financial variable is in an upturn (downturn) over the given sample period.*Significant at the 10% level. Significance refers to the difference between the 1960–85 period and the 1986–2007 period.**Significant at the 5% level.***Significant at the 1% level.View Table ImageThe proportion of time spent in upturns or downturns scales the length of cycles by the period studied. This metric varies considerably by financial variable, with equity and house prices in either upturn or downturn phase of the cycle most of the time (data in the table refer to sample means, with medians presented in brackets). While only 30% (20%) of the time credit experiences a downturn (upturn) episode, these fractions are 41% (31%) for downturns (upturns) of house prices and 45% (38%) for equity prices. Across the two subperiods, there are some significant changes in the proportion of time spent in different phases of cycles. For example, the average time spent in downturns for house and equity price cycles becomes significantly shorter in the globalization period whereas the time in upturns is shorter for credit.We can compare the frequency of financial cycles with that of business cycles using earlier work that reported 122 business cycles in advanced countries over the same period (see Claessens, Kose, and Terrones 2009). This indicates that cycles in equity prices are more frequent than business cycles whereas the frequencies of credit and housing cycles are comparable to that of business cycles. The decline in the number of downturns in credit over time is also consistent with the reduction in the number of recessions: while 73 out of 122 business cycles occur before 1985, only 49 take place in the globalization period.Duration. Financial cycles typically feature downturns lasting about 5–8 quarters (table 2). In contrast, the upturns tend to be much longer than downturns. Episodes of equity price upturns, for instance, last on average about 22 quarters whereas house prices take about 14 quarters to recover. Credit upturns are relatively short, on average 8 quarters. Given that a typical recession (or recovery) lasts about 4 quarters, our findings suggest that financial cycles are often more protracted than business cycles are.Table 2. Financial Cycles: Basic Features DownturnsUpturns DurationAmplitudeSlopeDurationAmplitudeSlopeCredit: Full period5.50−4.03−.938.004.361.23 [4.00][−6.68][−1.25][4.00][6.44] 1960–855.07−4.64**−1.31***7.305.56***1.31* [4.00][−6.93][−1.47***][4.00][8.14***][2.26] 1986–20076.16−2.87−.659.052.881.01 [4.00][−6.30][−.92][4.00][3.98][1.63]House price: Full period8.47−5.99−1.0614.253.621.19 [6.00][−10.85][−1.22][6.50][5.64][1.54] 1960–857.93−7.04−1.22**17.31*4.481.10 [6.00][−11.84][−1.40**][8.00][6.74][1.65] 1986–20079.04−5.02−.9311.303.131.36 [5.50][−9.82][−1.03][5.00][4.69][1.44]Equity price: Full period6.64−23.70−4.0721.9320.094.75 [5.00][−27.38][−4.70][7.00][24.08][5.99] 1960–857.84***−25.53−3.68**31.93***19.093.79*** [6.00***][−28.86][−4.12**][11.00***][23.32][5.27**] 1986–20075.32−22.74−4.7210.1421.915.56 [4.00][−25.76][−5.35][5.00][24.92][6.85]Note: The statistics for amplitude and slope refer to sample medians. Means are in brackets. For the statistic duration, means are shown with medians in brackets. Duration for downturns is the number of quarters between peak and trough. Duration for upturns is the time it takes to attain the level at the previous peak after the trough. The amplitude for the downturns is calculated on the basis of the decline in each respective variable during the peak to trough decline in the financial variable. The amplitude for the upturns is calculated on the basis of the 1-year change in each respective variable after the trough. The slope of the downturn is the amplitude from peak to trough divided by the duration. The slope of the upturns is the amplitude from the trough to the quarter at which the financial variable has reached the level at its last peak, divided by the duration.*Significant at the 10% level. Significance refers to the difference between the 1960–85 period and the 1986–2007 period.**Significant at the 5% level.***Significant at the 1% level.View Table ImageWhile the average duration of downturns has been stable over time, financial upturns have become shorter. In particular, asset price upturns are shorter in the globalization period, with equity upturns lasting 10 quarters, compared to almost 32 quarters in the earlier period. The means being (much) larger than the medians suggests though that durations of financial cycles often exhibit rather skewed distributions.11Amplitude and slope. Financial cycles tend to be intense (table 2). A typical credit downturn episode features about a 4% decline in credit and house and equity price downturns typically mean declines of some 6% and 24% in the respective asset price. The strength of upturns generally matches that of downturns. However, the amplitude of downturns and upturns differs across periods. Both phases of credit cycles tend to be deeper in the preglobalization period, but upturns of equity prices are more robust in the globalization period, which coincides with the rapid development of equity markets in many countries.The violence (speed) of cycles varies across markets. Downturns and upturns in credit and housing markets exhibit similar speeds, about 1% per quarter. Equity price cycles, however, tend to be three to four times more violent. These findings suggest that financial cycles, especially those in equity markets, are more pronounced than business cycles since they exhibit larger changes over the cycle and tend to be more violent.12The violence of financial cycles also differs over time. Downturns in credit and house prices are much faster in the preglobalization period, whereas those in equity are faster in the globalization period. These findings are consistent with earlier results in the literature suggesting that equity markets exhibit more rapid adjustment in the globalization period, reflecting more liberalized and expanded sets of arbitrage opportunities (e.g., Bekaert, Harvey, and Lumsdaine [2002] report that equity markets tend to be more volatile following financial liberalization).B. Synchronization of Financial CyclesSynchronization within countries. We next study the extent of synchronization across the three financial cycles within countries (table 3). We first compute the concordance between financial cycles in each country and then calculate both mean and median statistics of the concordance across countries. We also compute these statistics for each subperiod to analyze the evolution of synchronization over time.Table 3. Synchronization of Cycles within Countries CreditHouse PriceEquity Price A. Full Sample (Mean and Median)Credit….68.57House price.68….55Equity price.57.57… B. Subsamples (Median 1986–2007 and Mean 1960–85) Credit….74.63***House price.69….60*Equity price.51.53… C. Subsamples (Mean 1986–2007 and Median 1960–85) Credit….70.62***House price.65….58Equity price.52.53…Note: Each cell represents the mean or the median of the concordance statistics of the respective two cycles within countries. Concordance is calculated as the fraction of time that the two cycles are in the same phase. Panel A presents the means and medians of concordances within countries for the full sample, where the numbers above the diagonal are the means and the numbers below the diagonal are the medians. Panels B and C compare the means and medians of the concordance statistics for the subperiods, where the numbers above the diagonal are the means (medians) for the 1986–2007 subsample and the numbers below the diagonal are the means (medians) for the 1960–85 subsample.*Significant at the 10% level. Significance refers to the difference between the 1960–85 period and the 1986–2007 period.**Significant at the 5% level.***Significant at the 1% level.View Table ImageThe extent of synchronization between financial cycles varies but is not driven by outliers (means and medians are very similar). Cycles in credit and house prices are the most highly synchronized, with a median and mean of