No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Are Deep Recessions Followed by Strong Recoveries?
Abstract
We examine the hypothesis that deep recessions are followed by strong recoveries using a monthly data set for industrial production covering the period 1884-1990. There is a statistically significant relationship between growth in the first twelve months of a recovery and the peak-to-trough decline in industrial activity. This effect is still found when we exclude the Great Depression from our sample. We find no evidence that the length of the recession affects the strength of the subsequent recovery.
... The estimation of the effects of recessions on subsequent recoveries is based on extended versions of the models advanced by Wynne and Balke (1992) and Balke and Wynne (1995). Of course, the econometric estimation requires the business cycle regimes to be identified and characterized in terms of depth, steepness and duration (to be defined below), which is carried out by using the approach introduced by Artis et al. (1997); Appendix 1 describes it with more detailed and Appendix 2 shows the characteristics of state recessions and the average growth rates over the subsequent early recoveries. ...
... The econometric models to explain the average growth rate of output over the first k time periods of the recovery, ( ), incorporate two sets of explanatory variables. The first one consists of variables measuring the recession characteristics, as originally proposed by Wynne and Balke (1992), while the second set contains specific shocks that may generate initial impulses that pull greater growth rates of the government expenditure, . 11 Therefore, the estimated models include both the growth rates of the net federal and state government expenditure as regressors. ...
... Finally, models are also extended to consider not only values of k = 3, 4 quarters, as in Wynne and Balke (1992) and Balke and Wynne (1995), but also k equates the number of time periods required to get the value of output at the date of the peak defining the onset of the prior recession. ...
En este trabajo se generan estrategias especulativas en volatilidad para opciones europeas sobre el Índice de Precios y Cotizaciones (IPC) y sobre cuatro de sus componentes (ALFAA, AMXL, BIMBOA y KIMBERA), bajo el supuesto de volatilidad del activo subyacente conducida por un proceso GARCH-M (1,1) calibrado con datos históricos y el precio de la opción se obtiene por simulación Monte Carlo. Con las estrategias de volatilidad construidas con los precios de las opciones simuladas se determinó que la estrategia Cuna es adecuada para inversiones en 45 días para ALFAA y BIMBOA; y para 90 días en el IPC y en ALFAA.
... Third, the growth rates of the first 6 and 12 months of a recovery are greater than those of the subsequent periods and the growth rates of the last 12 months of a recovery are smaller than those of the first 12 months. This relationship is known as the "bounce-back" effect (Wynne and Balke 1992). ...
... In the second stage, the effects of recession characteristics on early expansion growth are estimated based on the model suggested by Wynne and Balke (1992) and Balke and Wynne (1995). In particular, they estimate the following relationship, ...
... The basic idea is that deeper, steeper, and/or longer recessions are followed by stronger early expansions due to a self-corrective response of the system to low economic activity. We have analyzed the effects of prior recessions on subsequent early expansions of total, permanent, and temporary employment of the Mexican states using the models proposed by Wynne and Balke (1992) and Balke and Wynne (1995). These models have been extended to take into account the effects of external and fiscal shocks as well as the role of structural characteristics of the states. ...
This paper estimates conventional linear models to evaluate whether the average growth rate of employment (permanent, temporary, and total) over early expansions depends on the characteristics of prior recessions for the Mexican states during the 2001 and 2008 recessions. After controlling for the initial impulse of external and fiscal shocks as well as for the effects of structural factors, our results suggest that the depth and steepness of prior recessions (measured as the percentage accumulated drop and monthly average growth rate) have negative effects on employment growth over the first 9 and 12 months following the 2001 recession. In contrast, employment growth in the aftermath of the 2008 recession can be explained primarily by external and fiscal shocks. In general, the duration of recessions has no effect on employment growth after recessions.
... In much earlier work, Wynne and Balke (1992) addressed the question of whether deep recessions in the U.S. over the period 1884-1990 had been followed by strong recoveries, and found (p 187) that the relationship between the size of the peak-to-peak decline and growth in the twelve-month period following the trough is statistically significant vi . ...
... With respect to recovery phases relative to expansion phases, it is first important to make clear the definition used for "recovery". Researchers have variously used the number of quarters from trough back to previous peak (Claessens et al., 2010b;Sichel, 1994), and fixed periods such as the initial four quarters (Wynne and Balke, 1992) or initial six quarters (Mussa, 2010). The recovery phases we present are from trough back to previous peak (Table 6, 3 rd panel). ...
... Thus, it would last from when employment falls below its normal level during a contraction phase and continue through to when employment regained its normal level during an expansion phase. vi Wynne and Balke (1992) also assessed whether the length of recession had affected the strength of recovery, and concluded that recession length had not significantly affected the strength of recovery. ...
We compute classical real GDP business cycles and growth cycles, and contrast classical recessions with "technical" recessions. Calling a technical recession after two successive quarters of negative growth can provide conditionally useful information, but can also send false signals. Expansion and contraction phases of classical real GDP and employment cycles have, on average, had an 86% association, but individual cycle circumstances should additionally be assessed. There is prima facie evidence that the severity of New Zealand"s recessions has mattered for a subsequent recovery path, and New Zealand"s average pattern of recovery has differed from that for U.S. NBER cycles.
... Sichel (1994) observes that real GNP tends to grow faster immediately following a trough than during the rest of the expansion phase. Wynne and Balke (1992) and Emery and Koenig (1992) present additional evidence in favor of this 'bounce-back' effect. This suggests the possibility of three business cycle phases: a contraction phase, a high-growth recovery immediately following the trough of the cycle, and a subsequent moderate growth phase. ...
... For example, Sichel (1994) demonstrates that growth in real GDP is larger immediately following a business cycle trough than during later parts of the expansion, suggesting that it might be worthwhile to decompose the expansion phase in a high-growth phase immediately following the trough of a cycle, and a subsequent moderate-growth phase. Wynne and Balke (1992) and Balke and Wynne (1996) document similar 'peak-reverting' behaviour in industrial production. Furthermore, they examine the relationship between growth during the first 12 months following a trough and the severity of the preceding contraction and show that deep recessions generally are followed by strong recoveries. ...
... This confirms the findings of Beaudry and Koop (1993) and Potter (1995b), among others, that contractions are less persistent than expansions. Also note the large constant in the upper regime, which might be taken as an additional indication of a quick recovery following contractions, see Sichel (1994) and Wynne and Balke (1992). ...
The dynamic properties of many economic time series variables can be characterised as state-dependent or regime-switching. A popular model to describe this type of non-linear behaviour is the smooth transition model, which accommodates two regimes facilitating a gradual transition from one regime to the other. The first part of this thesis considers three extensions of the basic smooth transition model. Models are developed which allow for more than two regimes, for time-varying properties in conjunction with regime-switching behaviour, and for modeling several time series jointly. Particular emphasis is placed on the inter-related issues of specification and inference in such models. The second part of the thesis concerns the influence of atypical observations on testing procedures for smooth transition non-linearity and on the estimation of smooth transition models. Traditional methods that are used for these purposes are found to be very sensitive to such outliers. Therefore, outlier robust testing procedures and estimation methods are developed
... Kim and Nelson (1999b) motivate their model as nesting Friedman's (1964Friedman's ( , 1993 plucking model, which assumes output cannot exceed a ceiling level, but is occasionally plucked below full capacity by recessionary shocks resulting from activist monetary policy. In line with Friedman's observations, Kim and Nelson's (1999b) model relates the strength of a recovery to the severity of the preceding recession, with no corresponding link between the strength of an expansion and the severity of a recession (see also Wynne and Balke, 1992, Sichel, 1994, and Balke and Wynne, 1996. Notably, the transitory component for their estimated model achieves the trifecta of business cycle asymmetries in the form of deepness, steepness, and sharpness. ...
... However, beyond the issue of how to define a business cycle, the major question in the literature on reproducing business cycle features is which features to consider. Galvão (2002), Kim, Morley, and Piger (2005), and Morley and Piger (2006) examine the ability of linear and nonlinear models to capture high-growth recoveries that are related to the severity of the preceding recessions, which is the asymmetry emphasized by Friedman (1963Friedman ( , 1993, Wynne and Balke (1992), Sichel (1994), and Balke and Wynne (1996). When considering this feature, there is strong support for Kim and Nelson's (1999b) plucking model and Kim, Morley, and Piger's (2005) bounceback model over linear models. ...
Glossary nonlinear time series in macroeconomics – a field of study in economics pertaining to the use of statistical analysis of data in order to make inferences about nonlinearities in the nature of aggregate phenomena in the economy time series – a collection of data corresponding to the values of a variable at different points of time linear –refers to a class of models for which the dependence between two random variables can be completely described by a fixed correlation parameter nonlinear – refers to the class of models for which the dependence between two random variables has a more general functional form than a linear equation and/or can change over time structural change – a change in the model describing a time series, with no expected reversal of the change level – refers to a definition of the business cycle that links the cycle to alternation between phases of expansion and recession in the level of economic activity deviations – refers to a definition of the business cycle that links the cycle to transitory deviations of economic activity from a trend level 2 fluctuations – refers to a definition of the business cycle that links the cycle to any short-run changes in economic activity deepness – a characteristic of a process with a skewed unconditional distribution steepness – a characteristic of a process with a skewed unconditional distribution for its first-differences sharpness – a characteristic of a process for which the probability of a peak when increasing is different than the probability of a trough when decreasing time reversibility – the ability to substitute -t and t in the equations of motion for a process without changing the process Markov-switching models – models that assume the prevailing regime governing the conditional distribution of a variable or variables being modeled depends on an unobserved discrete Markov process self-exciting threshold models – models that assume the prevailing regime governing the conditional distribution of a variable or variables being modeled is observable and depends on whether realized values of the time series being modeled exceed or fall below certain "threshold" values nuisance parameters – parameters that are not of direct interest in a test, but influence the distribution of a test statistic pivotal – refers to the invariance of the distribution of a test statistic with respect to values of parameters in the data generating process under the null hypothesis size – probability of false rejection of a null hypothesis in repeated experiments power – probability of correct rejection of a null hypothesis in repeated experiments 1. Definition of the Subject and Its Importance Nonlinear time series in macroeconomics is a broad field of study in economics. It refers to the use of statistical analysis of data to make inferences about nonlinearities in the nature of aggregate phenomena in the economy. This analysis is relevant for forecasting, the formulation of economic policy, and the development and testing of macroeconomic theories.
... With the methods that lead to a more rigid trend, the overall contribution increases slightly, the band pass filter leads to a decrease of the overall contribution. 57 Altogether it remains substantial. The following table 7 conveniently summarizes these contributions as well as the cyclical fit; Figure 27 displays the behavior of cyclical aggregate employment in the two recent recoveries under different de-trending methods. ...
... The following table 7 conveniently summarizes these contributions as well as the cyclical fit; Figure 27 displays the behavior of cyclical aggregate employment in the two recent recoveries under different de-trending methods. 58 57 It should be noted, however, that the band pass filter with the moving average is particularly sensitive at the beginning and the end of a time series: excluding the 1949 recovery, which just starts eight quarters after the beginning of the data horizon, leads to considerably higher values: 46% and 32% on average, for 1991 and 2001, Next, I briefly discuss how robust the occurrence of jobless recoveries is with respect to some of the choices for the micro parameters. Thus far, I have calibrated adjustment costs to match the average job turnover rate of continuing establishments in U.S. data. ...
The recent jobless recoveries have puzzled many. Most explanations have focused on structural change. In this paper, I investigate the quantitative contribution of a cyclical mechanism towards generating jobless recoveries. Extending the model by Hansen and Sargent (1988), I calibrate and compute a dynamic stochastic general equilibrium model of heterogeneous establishments that use two margins of labor services: an intensive margin, hours per worker, and an extensive margin, employment. When facing adjustment costs to the latter, aggregate employment at the end of a short and shallow recession is still rela-tively high and the need for new hires is weak. Moreover, establishments increase average hours per worker in the early phase of a weak recovery, before they start hiring anew later on. This pattern is consistent with U.S. data. I find that this mechanism can explain approx-imately half of the differential behavior of aggregate employment in the last two recoveries compared with previous ones.
... 29 The core set of explanatory variables is very similar to the ones we employed in our earlier models, but we now also include the depth of the preceding recession to allow us to test whether economies tend to bounce-back faster from deeper recessions as argued by some earlier studies. 30 Results indicate that the deeper the preceding recession, the strongest the recovery (column 1), consistent with results reported in other studies for the U.S. (Friedman and Kuttner, 1993;Wynne and Balke, 1992;and Mussa, 2009). The nature of this relation does not change when we include other controls (columns 2-8). ...
... This result confirms that recoveries from recessions are characterized by a bounce-back effect and that 29 We do not study the duration of recoveries in the same way since the amplitude of a recovery is measured over a fixed period of four quarters. 30 There could be many factors leading an economy to bounce back faster from a deeper recession, such as rapid productivity growth possibly because of the "cleansing" effects of recessions (see Sichel, 1994;and Wynne and Balke, 1992). recoveries from recessions associated with housing price busts are relatively weak. ...
This paper analyzes the interactions between business and financial cycles using an extensive database of over 200 business and 700 financial cycles in 44 countries for the period 1960:1-2007:4. Our results suggest that there are strong linkages between different phases of business and financial cycles. In particular, recessions associated with financial disruption episodes, notably house price busts, tend to be longer and deeper than other recessions. Conversely, recoveries associated with rapid growth in credit and house prices tend to be stronger. These findings emphasize the importance of developments in credit and housing markets for the real economy.
... They find that although there is weak support for the hypothesis of steepness, there is substantial support for the proposal that the output ceiling plays a major role in business cycle fluctuations in Canada, France, Germany, Switzerland, and the United States. 3 Also, Balke andWynne (1992, 1996) find evidence of depth but not of steepnessd defined as the average monthly growth rate of output over the course of the recessiondor length for G7 countries. Razzak (2001) applies a nonparametric test to a set of six industrial countries and finds that Japan and Australia real GDP series show significant depth, while there is only significant steepness in real GDP for New Zealand. ...
... They find that although there is weak support for the hypothesis of steepness, there is substantial support for the proposal that the output ceiling plays a major role in business cycle fluctuations in Canada, France, Germany, Switzerland, and the United States. 3 Also, Balke andWynne (1992, 1996) find evidence of depth but not of steepnessd defined as the average monthly growth rate of output over the course of the recessiondor length for G7 countries. Razzak (2001) applies a nonparametric test to a set of six industrial countries and finds that Japan and Australia real GDP series show significant depth, while there is only significant steepness in real GDP for New Zealand. ...
Friedman's plucking model of business fluctuations suggests that output cannot exceed a ceiling level, and it is occasionally plucked downward by recession; output has depth and steepness. This study uses a sample of 12 industrial and emerging economies and finds some evidence that negative shocks are largely transitory, while positive shocks are mostly permanent. In a few cases, as implied by the plucking model, output fluctuations tend to be asymmetric: recessions are transitory, and duration dependent, although expansions are not. There is, however, serial correlation in almost all cases. International evidence on Friedman's plucking model is far from robust.
... This is in line with Friedman' s (1964Friedman' s ( , 1993 ' plucking' model of economic fluctuations. Evidence in favor of Friedman' s plucking model, or asymmetry in the transitory component of output, has also been reported by Wynne and Balke (1992), and Goodwin and Sweeney (1993). ...
... Even though the literature discussed in Section 2.1, such as Wynne and Balke (1992), Beaudry and Koop (1993), Sichel (1994), and Kim and Nelson (1999a) provide copious evidence of the transitory nature of recessions, their results are entirely univariate. The purpose of this section is to provide a model with which one can analyze the potential transitory nature of recessions within a multivariate framework, and assess the relative importance of permanent and transitory shocks during recessions. ...
We introduce a new hybrid approach to joint estimation of Value at Risk (VaR) and Expected Shortfall (ES) for high quantiles of return distributions. We investigate the relative performance of VaR and ES models using daily returns for sixteen stock market indices (eight from developed and eight from emerging markets) prior to and during the 2008 financial crisis. In addition to widely used VaR and ES models, we also study the behavior of conditional and unconditional extreme value (EV) models to generate 99 percent confidence level estimates as well as developing a new loss function that relates tail losses to ES forecasts. Backtesting results show that only our proposed new hybrid and Extreme Value (EV)-based VaR models provide adequate protection in both developed and emerging markets, but that the hybrid approach does this at a significantly lower cost in capital reserves. In ES estimation the hybrid model yields the smallest error statistics surpassing even the EV models, especially in the developed markets.
... This is in line with Friedman' s (1964Friedman' s ( , 1993 ' plucking' model of economic fluctuations. Evidence in favor of Friedman' s plucking model, or asymmetry in the transitory component of output, has also been reported by Wynne and Balke (1992), and Goodwin and Sweeney (1993). ...
... Even though the literature discussed in Section 2.1, such as Wynne and Balke (1992), Beaudry and Koop (1993), Sichel (1994), and Kim and Nelson (1999a) provide copious evidence of the transitory nature of recessions, their results are entirely univariate. The purpose of this section is to provide a model with which one can analyze the potential transitory nature of recessions within a multivariate framework, and assess the relative importance of permanent and transitory shocks during recessions. ...
We propose a generalization of existing business cycle models which allows us to decompose recessions into permanent and transitory components. We find that the transitory component of recessions accounts for between 77% and 96% of the observed variance of monthly indicator series. Our results suggest the following three phase characterization of the business cycle: recession, high-growth recovery during which output partially reverts to its previous peak, and normal growth following the recovery. In addition, we find significant timing differences between the permanent and transitory components of recessions; most notably the lack of the usual high-growth recovery phase following the 1990-91 recession. JEL Codes: C32, E32 Kim: Department of Economics, Korea University, Seoul, 136-701, Korea (cjkim@kuccnx.korea.ac.kr); Murray: (Corresponding author) Department of Economics, University of Houston, Houston, TX 77204-5882 (cjmurray@uh.edu), Tel: 713-743-3835, Fax: 713-743-3798 1 1. In...
... 2 Friedman's plucking model (1964,1993) is an asymmetric real business cycle model in which, by considering a ceiling of maximum output, explains that business cycles are the result of a negative shock to the economy during the recession and a self-corrective response known as the "bounce-back" effect during the recovery. The main stylized fact of this model states that deeper recessions breed stronger recoveries and has been empirically confirmed by different methodologies in the U.S. (see, among others, Bordo and Haubrich, 2016;Wynne and Balke, 1992;Beaudry and Koop, 1993). ...
Output in the U.S. and the U.K. recovered slowly following the 2007−09 financial crisis, even though unemployment rates returned to pre-crisis levels. To explain this mismatch, by using Okun's law and a dynamic factor model to estimate the counterfactual recovery, we identify a change in regime in the aftermath of the financial crisis as the main determinant of the slow recovery. Additionally, by applying a trend-cycle decomposition, performed based on the difference version of Okun's law, we distinguish between three driving forces of the slow recovery: declining trend growth began in the 1960s; unprecedented trend deceleration in U.S. potential output started during the 2007−09 financial crisis; and an unusually sluggish cyclical recovery known as hysteresis effects. Further, we develop an open-economy hierarchical dynamic factor model to demonstrate that spillovers of real activity shortfall from the U.S. explain at least half of the productivity puzzle in the U.K.
... 35 Following Kim et al. (2005), we assume N = 6 for the bounce-back models (52) and (53), which allow recovery to persist for up to six quarters following the end of a recession. 36 See Friedman (1993) and Wynne and Balke (1992) for details. 37 This may be due to the fact that Gaussian distribution incorrectly treats positive and negative outliers as switching to 0 and 1 regimes, respectively. ...
This study develops two types of robust test statistics applicable to Markov-switching autoregressive models. The test statistics can be constructed by sum functionals of the “smoothed” probabilities that a given observation came from a particular regime and do not require the estimation of additional parameters. Monte Carlo experiments show that the tests have good finite-sample size and power properties. The tests are applied to investigate the fluctuations in real GNP growth in the U.S.
... Here we focus on the literature surrounding the peak-reverting nature of recessions. Wynne and Balke (1992 Balke ( , 1996) find that the deeper the recession the stronger the ensuing recovery while Sichel (1994) finds evidence of a high growth recovery phase following recessions, both implications of peak reversion. Another implication of peak reversion is that negative shocks are less persistent than positive shocks. ...
This paper investigates the nature of U.S. business cycle asymmetry using a dynamic factor model of output, investment, and consumption. We identify a common stochastic trend and common transitory component by embedding the permanent income hypothesis within a simple growth model. Markov-switching in each component captures two types of asymmetry: Shifts in the growth rate of the common stochastic trend, having permanent effects, and “plucking” deviations from the common stochastic trend, having only transitory effects. Statistical tests suggest both asymmetries were present in post-war recessions, although the shifts in trend are less severe than found in the received literature.
... One approach to modeling the high growth recovery phase is to add a distinct third regime to Hamilton's model, as in Sichel (1994). However, there is some evidence that recoveries are not independent of the preceding recession, as would be implied by a standard three-regime model, but rather the strength of the post-recession recovery is related to the length and severity of the recession (see Friedman, 1964, 1993, and Wynne and Balke, 1992, 1996). Kim and Nelson (1999a) allow for this type of business cycle asymmetry by modeling regime switching in the cyclical component of output only. ...
This paper presents a new nonlinear time series model that captures a post-recession 'bounce-back' in the level of aggregate output. While a number of studies have examined this type of business cycle asymmetry using recession-based dummy variables and threshold models, we relate the 'bounce-back' effect to an endogenously estimated unobservable Markov-switching state variable. When the model is applied to US real GDP, we find that the Markov-switching regimes are closely related to NBER-dated recessions and expansions. Also, the Markov-switching form of nonlinearity is statistically significant and the 'bounce-back' effect is large, implying that the permanent effects of recessions are small. Meanwhile, having accounted for the 'bounce-back' effect, we find little or no remaining serial correlation in the data, suggesting that our model is sufficient to capture the defining features of US business cycle dynamics. When the model is applied to other countries, we find larger permanent effects of recessions. Copyright © 2005 John Wiley & Sons, Ltd.
... An important question is whether the performance of the nonlinear models in reproducing characteristics of the sample data is being influenced by something other than simply capturing nonlinear dynamics related to the business cycle. In particular, regime switching in the mean can generate patterns that look similar to a structural break in volatility or other forms of 5 See also Wynne and Balke (1992 Balke ( , 1996). heteroskedasticity that may be present in the sample data. ...
This paper considers the ability of simulated data from linear and nonlinear time-series models to reproduce features in U.S. real GDP data related to business cycle phases. We focus our analysis on a number of linear ARIMA models and nonlinear Markov-switching models. To determine the timing of business cycle phases for the simulated data, we present a model-free algorithm that is more successful than previous methods at matching NBER dates and associated features in the postwar data. We find that both linear and Markov-switching models are able to reproduce business cycle features such as the average growth rate in recessions, the average length of recessions, and the total number of recessions. However, we find that Markov-switching models are better than linear models at reproducing the variability of growth rates in different business cycle phases. Furthermore, certain Markov-switching specifications are able to reproduce high-growth recoveries following recessions and a strong correlation between the severity of a recession and the strength of the subsequent recovery. Thus, we conclude that nonlinearity is important in reproducing business cycle features.
... 3 Some economists (for example, Wynne and Balke (1992)) have argued that understanding recoveries requires understanding the recessions that preceded them. 4 The expansions that started after the 1948-49, 1953-54, and 1957-58 recessions are excluded from the average because data on most of the labor-market variables examined in this article are not available for them. ...
In January 2005, after more than three years of sluggish employment growth, the U.S. economy finally recovered the jobs lost during the 2001 recession. Baffled by such a delayed rebound in payrolls, many speculated about the cause. Inevitably, observers compared the 2001 and 1991 recoveries, both widely considered to have been jobless. Schreft and Singh showed previously that one common feature of the first year of the jobless recoveries was the greater use of just-in-time employment practices. They also speculated that the greater availability of just-in-time employment practices contributed to the recoveries’ lack of job growth. This explanation of delayed hiring is termed the “wait-and-see hypothesis.” Flexible hiring practices allow firms to more easily adjust output in the short term without hiring full-time, potentially permanent workers. This practice is especially effective around the troughs of business cycles, when there is uncertainty about the strength of the recovery. As a result, firms are willing to wait to hire until they see sufficient improvement in business conditions to justify expanding payrolls. Schreft, Singh, and Hodgson take a longer-term perspective, considering the behavior of employment in the first three years of the jobless recoveries. They also describe how a wait-and-see approach to hiring can contribute to such recoveries
... Here, business cycles have three phases: recession, high-growth recovery phase, and normal expansion phase. Wynne and Balke (1992 Balke ( , 1996) find that the deeper the recession the stronger the ensuing recovery while Sichel (1994) finds evidence of a high growth recovery phase following recessions. Both of these results are implications of Friedman's model. ...
This paper investigates the relationship between permanent and transitory components of U.S. recessions in an empirical model allowing for business cycle asymmetry. Using a common stochastic trend representation for real GNP and consumption, we divide real GNP into permanent and transitory components, the dynamics of which are different in booms vs. recessions. We find evidence of substantial asymmetries in postwar recessions, and that both the permanent and transitory component have contributed to these recessions. We also allow for the timing of switches from boom to recession for the permanent component to be correlated with switches from boom to recession in the transitory component. The parameter estimates suggest a specific pattern of recessions: switches in the permanent component lead switches in the transitory component both when entering and leaving recessions.
... We wish to thank Adrienne C. Slack and Shengyi Guo for excellent research assistance on this project. This article draws on work reported in an Economics Letters article (Wynne and Balke 1992) and in an unpublished paper (Balke and Wynne 1992). expansion " (p. ...
The U.S. recession that began in July 1990 may have ended in April or May 1991. The pace of the subsequent recovery has been so sluggish as to be indistinguishable, in the eyes of many, from continued recession. One explanation for the sluggish pace of the recovery is that the recession itself was not particularly severe, at least when compared with others. ; In this article, Mark Wynne and Nathan Balke use monthly data on industrial production to examine the hypothesis that the severity of a recession determines the pace of the subsequent recovery. They show that, historically, the relationship between growth in the first twelve months of a recovery and the decline in industrial activity from peak to trough is statistically significant. However, there is no relationship between the length of a recession and the strength of the recovery. Consistent with their finding of a bounce-back effect for industrial production, the recovery from the 1990-91 recession is the weakest in the period covered by the Federal Reserve Board's industrial production index, just as the decline in industrial production over the course of that recession is the mildest on record.
... However, one implication of the two-state Markov-switching trend model is that recessions have permanent effects on the level of output, that is the economy never recovers output lost during a recession. Many authors have provided evidence that this implication is not consistent with the data, instead, following steep, short recessions the economy seems to undergo a high-growth recovery phase to gain back what was lost, see for example Friedman (1964 Friedman ( , 1993), Wynne and Balke (1992 Balke ( , 1996), and Sichel (1994). In other words, the business cycle is better characterized with three phases rather than two. ...
: We investigate the power and size performance of unit root tests when the true data generating process undergoes Markov regime-switching. All tests, including those robust to a single break in trend growth rate, have very low power against a process with a Markov-switching trend growth rate as in Lam (1990). However, for the case of business cycle non-linearities, unit root tests are very powerful against models used as alternatives to Lam (1990) that specify regime-switching in the transitory component of output. Under the null hypothesis, the received literature documents size distortions in Dickey-Fuller type tests caused by a single break in trend growth rate or variance. We find these results do not generalize to most parameterizations of Markov-switching in trend or variance. However, Markov-switching in variance can lead to over-rejection in tests robust to a single break in the level of trend. Keywords: unit root, stochastic trends, deterministic trends, Markov-switching, str...
We integrate Friedman’s plucking model and the gap version of Okun’s law by embedding output and the unemployment rate in a bivariate unobserved components model with Markov-switching to examine their asymmetric co-fluctuations in the U.S. economy. The results establish that output and unemployment are synchronously and proportionally characterized by the plucking property. Given that the labour market has been identified as the source of the plucking property, we suggest that the plucking property, through stable Okun’s law, transmits from unemployment to output. Our proposed asymmetric bivariate model provides two additional results regarding two specification aspects of trend-cycle decomposition: First, we identify an unprecedented deceleration in U.S. potential output in the aftermath of the 2007−09 global financial crisis. Second, our model yields a robust estimation of parameters and components with insignificant correlations between shocks.
p>El objetivo de este documento es explicar la recuperación del empleo formal (total, permanente y eventual) de los estados mexicanos durante la fase post-pandemia del COVID-19, para lo que se realiza un análisis econométrico. Entre los resultados principales destacan una relación negativa y robusta entre el crecimiento del empleo estatal con la amplitud de la recesión precedente, lo que da evidencia del efecto “rebote”, así como efectos positivos y significativos del grado de apertura económica y de la capacidad de ajuste de los mercados a los choques, pero muy limitados del gasto público.</p
A key challenge in the analysis of South African business cycles lies with understanding how South African business cycles differ from the business cycles of other countries, whether advanced or emerging. South African cycles, quite different from especially US cycles, do not exhibit a distinct high-growth phase during recovery, which is defined as the first stage of an expansion. Using alternative recovery identification methods, we find that South African cycles share this property with cycles in developed, as well as emerging economies, but are different from cycles in other commodity exporters. The absence of a distinct recovery phase carries over to disaggregated components of expenditure, including consumption and investment. Especially investment only accelerates later in South African expansions.
This paper investigates the nature of business cycle asymmetry using a dynamic factor model of output, investment, and consumption. We first identify a common stochastic trend and a common transitory component by embedding the permanent income hypothesis within a simple growth model. We then investigate two types of asymmetry commonly identified in U.S. business cycle dynamics: (1) Infrequent negative permanent shocks, modeled as shifts in the growth rate of the common stochastic trend and (2) infrequent negative transitory shocks, modeled as "plucking" deviations from the common stochastic trend. Tests of marginal significance suggest both types of asymmetry were present in post-war recessions, although the shifts in trend are less severe than the received literature suggests.
We investigate the power and size performance of unit root tests when the true data generating process undergoes Markov regime-switching. All tests, including those robust to a single break in trend growth rate, have very low power against a process with a Markov-switching trend growth rate as in Lam (1990). However, for the case of business cycle non-linearities, unit root tests are very powerful against models used as alternatives to Lam (1990) that specify regime-switching in the transitory component of output. Under the null hypothesis, the received literature documents size distortions in Dickey-Fuller type tests caused by a single break in trend growth rate or variance. We find these results do not generalize to most parameterizations of Markov-switching in trend or variance. However, Markov-switching in variance can lead to over-rejection in tests robust to a single break in the level of trend.
This paper investigates the relationship between permanent and transitory components of U.S. recessions in an empirical model allowing for business cycle asymmetry. Using a common stochastic trend representation for real GNP and consumption, we divide real GNP into permanent and transitory components, the dynamics of which are different in booms vs. recessions. We find evidence of substantial asymmetries in postwar recessions, and that both the permanent and transitory component have contributed to these recessions. We also allow for the timing of switches from boom to recession for the permanent component to be correlated with switches from boom to recession in the transitory component. The parameter estimates suggest a specific pattern of recessions: switches in the permanent component lead switches in the transitory component both when entering and leaving recessions.
Do steep recoveries follow deep recessions? Does it matter if a credit crunch or banking panic accompanies the recession? We look at the American historical experience in an attempt to answer these questions. The answers depend on the definition of a financial crisis and on how much of the recovery is considered. But in general recessions associated with financial crises are followed by rapid recoveries. We find three exceptions to this pattern: the recovery from the Great Contraction in the 1930s, the recovery after the recession of the early 1990s, and the present recovery. The present recovery is strikingly more tepid than the 1990s. Possible factors to explain the slowness of this recovery include residential investment and policy uncertainty. (JEL E32, N10, G01)
This paper analyzes the causes of the slow recovery of the US economy since the financial crisis and Great Recession of 2008-9. Fallen house values and excessive household debts continue to depress consumer spending, while corporations are failing to invest in spite of record profits. The increasingly unequal distribution of income limits demand, while long-term structural transformations continue to erode employment creation. An expansionary monetary policy has been incapable of sparking a more robust recovery and fiscal policy has been shifted to an austerity stance. In this context, Brazil and other emerging market nations cannot count on the United States to continue to be the leading source of global demand as it was in previous decades.
We investigate the power and size performance of unit-root tests when the data undergo Markov regime switching. All tests, including those robust to a single break in trend growth rate, have low power against a process with a Markov-switching trend. Under the null hypothesis, we find that previously documented size distortions in Dickey–Fuller-type tests caused by a single break in trend growth rate or variance do not generalize to most parameterizations of Markov switching in trend or variance. However, Markov switching in variance can lead to overrejection in tests allowing for a single break the level of trend.
Do steep recoveries follow deep recessions? Does it matter if a credit crunch or banking panic accompanies the recession? Moreover, does it matter if the recession is associated with a housing bust? We look at the American historical experience in an attempt to answer these questions. The answers depend on the definition of a financial crisis and on how much of the recovery is considered. But in general recessions associated with financial crises are generally followed by rapid recoveries. We find three exceptions to this pattern: the recovery from the Great Contraction in the 1930s; the recovery after the recession of the early 1990s and the present recovery. The present recovery is strikingly more tepid than the 1990s. One factor we consider that may explain some of the slowness of this recovery is the moribund nature of residential investment, a variable that is usually a key predictor of recessions and recoveries.
Recoveries from recessions associated with a financial crisis tend to be sluggish. In this paper, we present evidence that stressed credit conditions are an important factor constraining the pace of recovery. In particular, using industry-level data, we find that industries relying more on external finance grow more slowly than other industries during recoveries from recessions associated with financial crises. Additional tests, based on establishment size, on alternative definitions of financial crises, and on corporate-government interest rate spreads, support the findings. Moreover, for subsets of industries where financial frictions are more severe, we find much stronger differential growth effects.
Recoveries vary considerably across countries: our paper compares recoveries in bank-based and market-based economies and finds that market-based economies experience significantly and durably stronger rebounds than the bank-based ones (in particular the more bank-based economies of continental Europe). Further, stronger recoveries also tend to be associated with broader economic flexibility. Our findings suggest that dealing with bank sector vulnerabilities is paramount to support the recovery. In the medium term, structural policies to deepen financial markets are useful, but need to be complemented with structural measures to address rigidities more broadly in the real economy.
The Great Recession of 2008–09 was by far the most severe United States economic downturn since the Great Depression of the 1930s. Real gross domestic product (GDP), the most comprehensive measure of U.S. economic activity, topped out in fourth quarter 2007 and has yet to approach that peak. Employment totaled just below 138 million jobs in January 2008 and, as of July 2011, was still nearly 5 percent below its precrisis level.
The interest in business cycle asymmetry has been steadily increasing over the last fifteen years. Most research has focused on the different behaviour of macroeconomic variables during expansions and contractions, which by now is well documented. Recent evidence suggests that such a two-phase characterization of the business cycle might be too restrictive. In particular, it might be worthwhile to decompose the recovery phase in a high-growth phase (immediately following the trough of a cycle) and a subsequent moderate-growth phase. In this paper, the issue of multiple regimes is addressed using Smooth Transition AutoRegressive [STAR] models. A possible limitation of STAR models as they are currently used is that essentially they deal with only two regimes. We propose a generalization of the STAR model such that more than two regimes can be accommodated. It is demonstrated that the class of Multiple Regime STAR [MRSTAR] models can be obtained from the two-regime model in an elegant way. The main properties of the MRSTAR model and several issues which might be relevant for empirical specification are discussed in detail. In particular, a Lagrange Multiplier-type test is derived which can be used to determine the appropriate number of regimes. Application of the new model class to US real GNP and US unemployment rate provides evidence in favor of the existence of multiple business cycle phases.
How does the management and resolution of the current crisis compare with the response of the Nordic countries in the early 1990s, widely regarded as exemplary? We argue that, while intervention has been prompter, the measures taken so far remain less comprehensive and in-depth. In particular, the cleansing of balance sheets has proceeded more slowly, and less attention has been paid to reducing excess capacity and avoiding competitive distortions. In general, policymakers have given higher priority to sustaining aggregate demand in the short term than to encouraging adjustment in the financial sector and containing moral hazard. We argue that three factors largely explain this outcome: the more international nature of the crisis; the complexity of the instruments involved; and, hardly appreciated so far, the effect of accounting practices on the dynamics of the events, reflecting in particular the prominent role of fair value accounting (and mark to market losses) in relation to amortised cost accounting for loan books. There is a risk that the policies followed so far may delay the establishment of the basis for a sustainably profitable and less risk-prone financial sector.
The hypothesis is examined that the severity of a recession favourably affects the rate of growth of output during the period immediately after the recession. Our empirical analysis is based on the behaviour of industrial output in the G-7 countries during the period 1960 to 1985. The depth of a recession, defined as the cumulative output loss between the peak and trough dates, is shown to be negatively correlated with growth in the first 12 months of the subsequent expansion.
The interest in business-cycle asymmetry has beensteadily increasing over the past 15 years. Most research has focused onthe different behavior of macroeconomic variables during expansions andcontractions, which by now is well documented. Recent evidence suggests thatsuch a two-phase characterization of the business cycle might be toorestrictive. In particular, it might be worthwhile to decompose the recoveryphase in a high-growth phase (immediately following the trough of a cycle)and a subsequent moderate-growth phase. The issue ofmultiple regimes in the business cycle is addressed usingsmooth-transition autoregressive (STAR) models. A possiblelimitation of STAR models as they currently are used is that essentiallythey deal with only two regimes. We propose a generalization of the STARmodel such that more than two regimes can be accommodated.It is demonstratedthat the class of multiple-regime STAR (MRSTAR) models can beobtained from the two-regime model in a simple way.The main properties of the MRSTAR model and several issues that arerelevant for empirical specification are discussed in detail.In particular, a Lagrange multiplier-type test is derived that can beused to determine the appropriate number of regimes. A limitedsimulation study indicates its practical usefulness.Application of the new model class to U.S. real GNP provides evidence infavor of the existence of multiple business-cycle phases.
The authors evaluate the ability of a simple real business cycle model to generate business cycles in the classical NBER sense of the term, where recessions are periods of absolute declines in economic activity. They use the 'phase' classification of Arthur F. Burns and Wesley C. Mitchell (1946) to determine the 'shape' of the business cycle and to look for asymmetries between expansions and contractions. The authors show that such a model can generate business cycles of plausible duration and depth but cannot match the actual 'shape' of the business cycle. Nonlinear models, such as Milton Friedman's (1993) 'plucking' model may more closely match the observed shape. Copyright 1995 by Oxford University Press.
Gaps between output and employment growth are often attributed to transitional phases by which the economy adjusts to shifts in the rate of trend productivity growth. Nevertheless, cyclical factors can also drive a wedge between output and employment growth. This article shows that one measure of cyclical dynamics-the expected output loss associated with a recession-helps predict the gap between output and employment growth in the coming four quarters. This measure of the output loss associated with a recession can take unexpected twists and turns as the recovery unfolds. The empirical results in this paper support the proposition that a weaker-than-expected rebound in the economy can partially mute employment growth for a time relative to output growth.
In many models with imperfect capital markets, credit plays an important role in the propagation of shocks. Furthermore, this propagation mechanism often implies nonlinear dynamics in the form of asymmetry and regime switching. In this paper, we examine empirically whether credit plays a separate role as a propagator of shocks. We model this propagation empirically as a threshold model in which the dynamics of output grouth changes if the commercial paper/heasury bill spread exceeds a critical threshold. We test and estimate both a single equation threshold model for output growth and a threshold vector autoregression that includes output growth, inflation, a monetary variable, and the paper-bill spread and find evidence of a threshold structure. Using nonlinear impulse response firnctions, we evaluate the dynamics implied by the threshold model. These suggest that money and paper-bill shocks have a larger effect on output in the "tight" credit regime than is normally tle case and that negative money shocks typically have a larger effect than positive shocks. Finally, using a nonlinear version of historical decompositions, we examine post-1960 macroeconomic history throueh the lens of our threshold vector autoresression.
In this article, Evan F. Koenig demonstrates that the Federal Reserve Board's initial estimate of manufacturing capacity utilization is helpful in predicting subsequent growth in manufacturing output. Together with lagged real-time output growth and growth in the composite index of leading indicators, capacity utilization explains more than 50 percent of the variation in output growth at a four-quarter horizon. Based on data available at the beginning of the year, the forecasting equation predicts little or no growth in manufacturing output during 1996.
We argue that Ireland experienced a great depression in the 1980s comparable in severity to the better known and more studied depression episodes of the interwar period. Using the business cycle accounting framework of Chari, Kehoe and McGrattan (2005), we examine the factors that led to the depression and the subsequent recovery in the 1990s. We calculate efficiency, labour, investment and government wedges and evaluate the contribution of each to the downturn and subsequent recovery. We find that the efficiency wedge on its own can account for a significant portion of the downturn, but predicts a stronger recovery in output than occurred. The labour wedge also helps account for what happened during the depression episode. We also find that the investment wedge played no role in the depression.
In this paper, we propose a generalised specification of a time varying transition probability Markov switching model for U.S. industrial production index. The model is specifically designed to investigate the presence of asymmetries in the shape of the cycle, given its relevance in the debate about long-run effects of recessions on output level. We can think about asymmetries in the shape of the cycle along two main dimensions. First, we can think about patterns of variation in growth rates over the course of expansions and recessions. Second, we can consider to which extent recessions are simply negative expansions. The model, estimated using Bayesian methods, generates posterior probabilities of being in recessions which correspond to the NBER dated recessions, provides support to the presence of a recovery early in expansions, consistent with what found in the literature, and estimates the shape of recessions to be linear, contrary to some previous parametric studies. When we investigate the ability of our specification to produce plausible business cycle features, where those features are a set of statistics proposed by Harding and Pagan (2002), we find that the model is able to capture all of them. Finally, the effects of recessions on long-run output level implied by our specification lie between those predicted by two important benchmark models of this literature.
In this paper the author analyzes the dynamics of business cycles in the G7 countries. His aim is to answer the question of whether growth varies (in an absolute sense) over the business cycle, and hence to examine whether the simple two-phase characterization (downturn/upturn) is an appropriate description of the business cycle. To study growth over the cycle, the author estimates single-country equations and then a pooled regression that takes into account the important interactions of international business cycles. The results confirm the steepness asymmetry of the growth cycles and show that growth varies systematically within upturns and downturns. These are consistent with the notion of deepness asymmetry which implies that contractions are more pronounced in the later stages of a downturn whereas expansions are stronger immediately after troughs. Copyright 1997 by The London School of Economics and Political Science
Two-state models of the Commerce Department's leading and coincident indexes appear to be misspecified. Cyclical peaks in these indexes are more rounded than are cyclical troughs. Further, the variability of changes in the indexes is unusually high near cyclical troughs.
The goal of this paper is to theoretically account for business cycle asymmetries of deepness and steepness. The former means that recessions are deeper than expansions are tall, and the latter that recessions are steeper than expansions. In this paper I introduce the process of technology diffusion and learning like general purpose technology in the framework of real business cycles. I assume that a positive technology shock diffuses over the economy with some time lag, while a negative one does without any lag. Generally, a positive shock can be literally interpreted as an innovation to technology. Economic agents may take some time to adopt a new technology and learn how to use the technology efficiently. In contrast, a negative shock can immediately decrease the level or growth of productivity. No learning is needed to suffer a loss of productivity induced by a negative shock. A positive shock makes the near-future level of productivity higher than the present level as a result of technology diffusion. Because of intertemporal substitution behavior, it leads to a recession in the present and then the subsequent expansion. In contrast, a negative innovation is assumed to immediately generate a recession. When an S-shaped diffusion is assumed, a positive shock can induce a deeper and steeper recession. This gives a theoretical explanation of deepness and steepness asymmetries.
In recent decades the American economy has experienced the worst peace-time inflation in its history and the highest unemployment rate since the Great Depression. These circumstances have prompted renewed interest in the concept of business cycles, which Joseph Schumpeter suggested are "like the beat of the heart, of the essence of the organism that displays them." In The American Business Cycle, some of the most prominent macroeconomics in the United States focuses on the questions, To what extent are business cycles propelled by external shocks? How have post-1946 cycles differed from earlier cycles? And, what are the major factors that contribute to business cycles? They extend their investigation in some areas as far back as 1875 to afford a deeper understanding of both economic history and the most recent economic fluctuations. Seven papers address specific aspects of economic activity: consumption, investment, inventory change, fiscal policy, monetary behavior, open economy, and the labor market. Five papers focus on aggregate economic activity. In a number of cases, the papers present findings that challenge widely accepted models and assumptions. In addition to its substantive findings, The American Business Cycle includes an appendix containing both the first published history of the NBER business-cycle dating chronology and many previously unpublished historical data series.
The article derives a new monthly index of industrial production for the United States for 1884 to 1940. This index improves upon existing measures of industrial production by excluding indirect proxies for industrial activity, using only component series that are consistent over time, and not making ad hoc adjustments to the data. Analysis of the new index shows that it has more within-year volatility than conventional indexes, has relatively unimportant seasonal fluctuations, and has cyclical turning points that are grossly similar to but subtly different from existing series.
This paper examines the role of aggregate demand stimulus in ending the Great Depression. A simple calculation indicates that nearly all of the observed recovery of the U.S. economy prior to 1942 was due to monetary expansion. Huge gold inflows in the mid- and late-1930s swelled the U.S. money stock and appear to have stimulated the economy by lowering real interest rates and encouraging investment spending and purchases of durable goods. The finding that monetary developments were crucial to the recovery implies that self-correction played little role in the growth of real output between 1933 and 1942.
Industrial production and capacity utilization: G.17(419) historical data and source and description informationBoard of Governors of the Federal Reserve System A new monthly index of industrial production
- Federal
- Reserve
Federal Reserve Board, 1990, Industrial production and capacity utilization: G.17(419) historical data and source and description information, (Board of Governors of the Federal Reserve System, Washington D.C.), Miron, Jeffrey A. and Christina D. Romer, 1990, A new monthly index of industrial production, 1884-1940, The Journal of Economic History, L, 321-337
Appendix B: Historical data The American business cycle: Continuity and change An analysis of the dynamics of recoveries Are deep recessions followed by slrong recolberies? 1884-1940
- N S Balke
- R J Gordon
Balke, N.S. and R.J. Gordon. 1986, Appendix B: Historical data, in: R.J. Gordon, ed.. The American business cycle: Continuity and change (University of Chicago Press, Chicago, IL) 781-850. Balke, N.S. and M.A. Wynne, 1992, An analysis of the dynamics of recoveries, Manuscript. MA. Wynne und N.S. Balke / Are deep recessions followed by slrong recolberies? 1884-1940, Journal of Economic History 50, 321-337.
Peak to trough change in industrial production (percent)
- Research Papers
- The
- Research
- Federal
- Bank
- Dallas
Peak to trough change in industrial production (percent)
RESEARCH PAPERS OF THE RESEARCH DEPARTMENT
FEDERAL RESERVE BANK OF DALLAS
Industrial production and capacity utilization: G.17(419) historical data and source and description information
- Federal Reserve
- Board
Federal Reserve Board, 1990, Industrial production and capacity
utilization: G.17(419) historical data and source and description
information, (Board of Governors of the Federal Reserve System,
Washington D.C.),
Appendix B: Historical data
- Balke