Learning from History

Extra tables


Table A.1: Robustness–Different Fixed Effects

This table presents the estimated coefficients of (2) for different fixed effects. The time fixed effects used in the regressions are presented at the column header. Decade0 represents decade dummies that start with years ending in 0: 1800-1809, 1810-1819, etc. Decade1 represents decade dummies that start with years ending in 1: 1801-1810, 1811-1820, etc. Similarly, we consider up to Decade9 (i.e. 1809-1818, 1819-1828 etc.). In Column XII, we repeat the analysis without any fixed effects.  Finally in Column XIII, we report the results where 5 year fixed effects are employed. The panel covers 60 countries and spans 1800–2010. The standard errors, reported in parentheses, are robust and dually clustered at the year and country level.
Dep. Var.: $$C^\text{Banking}_{i,t}$$ Decade0 Decade1 Decade2 Decade3 Decade4 Decade5 Decade6 Decade7 Decade8 Decade9 NoFEs 5YearFEs
I II III IV V VI VII VIII IX X XII XIII
$$\delta^\text{high}_{i,t-1 \text{ to } t-5}$$ 0.20 0.19 0.20* 0.17 0.14 0.15 0.16 0.14 0.14 0.20 0.15 0.19
(0.128) (0.126) (0.118) (0.113) (0.110) (0.110) (0.107) (0.109) (0.115) (0.150) (0.105) (0.124)
$$\delta^\text{low}_{i,t-1 \text{ to } t-5}$$ -0.31*** -0.30*** -0.30*** -0.31*** -0.30*** -0.31*** -0.32** -0.31** -0.33** -0.33*** -0.24** -0.29**
(0.115) (0.113) (0.107) (0.111) (0.113) (0.120) (0.130) (0.135) (0.135) (0.119) (0.094) (0.117)
$$C_{i,t-1 \text{ to } t-5}$$ -7.86*** -7.60*** -7.33*** -7.42*** -7.48*** -7.35*** -7.33*** -7.73*** -7.65*** -7.67*** -3.79** -8.87***
(2.039) (2.027) (1.932) (1.960) (2.025) (2.148) (2.199) (2.311) (2.208) (2.162) (1.931) (2.536)
$$\log GDP_{i,t-1 \text{ to } t-5}$$ 0.07 0.01 0.07 0.03 0.01 0.04 0.10 0.05 0.15 0.13 0.05 0.07
(0.229) (0.219) (0.214) (0.197) (0.196) (0.187) (0.191) (0.189) (0.211) (0.223) (0.257) (0.231)
$$\Delta PD/GDP_{i,t-1 \text{ to } t-5}$$ -0.07*** -0.07*** -0.07*** -0.07*** -0.07*** -0.07*** -0.08*** -0.08*** -0.07*** -0.07** -0.03** -0.07***
(0.026) (0.023) (0.023) (0.023) (0.023) (0.023) (0.025) (0.024) (0.025) (0.028) (0.013) (0.022)
$$POLCOMP_{i,t-1 \text{ to } t-5}$$ -0.09* -0.09* -0.08** -0.09** -0.08* -0.07 -0.07* -0.07* -0.07 -0.08* -0.06 -0.09*
(0.048) (0.044) (0.039) (0.042) (0.044) (0.045) (0.040) (0.042) (0.045) (0.046) (0.049) (0.053)
$$INFLATION_{i,t-1 \text{ to } t-5}$$ 0.02 0.02* 0.02* 0.02* 0.02* 0.02* 0.02 0.02 0.02 0.02 0.01* 0.02
(0.011) (0.010) (0.010) (0.010) (0.011) (0.011) (0.011) (0.012) (0.011) (0.010) (0.005) (0.014)
Num of Obs. 2,134 2,124 2,113 2,103 2,071 2,383 2,352 2,360 2,348 2,233 2,886 2,015
$$\text{Pse.} R^2$$ 0.106 0.0968 0.104 0.0965 0.0964 0.119 0.135 0.135 0.124 0.106 0.0292 0.143

Table A.2: Volatility and risk-taking: censored dependent variable

This table provides the estimated coefficients of (4) considering censored dependent variables. The dependent variable used is listed at the column header. As the dependent variables (high credit or high leverage) are left-censored by definition, a Tobit-type estimator is appropriate. Hence, in Columns I and II, we report the estimated coefficients for the Honore’s (1992) panel regression model in the presence of cross-section fixed effects with and without time dummies, respectively. In Column III, Tobit regressions with random effects are reported. Columns IV through VI present the similar models, where high leverage is the dependent variable. Credit-to-GDP data is obtained from BIS and leverage data is obtained from Lee et al (2017). Region and decade fixed effects are included in all of the specifications. For the sake of brevity, the estimated coefficients of fixed effects are omitted. The standard errors are reported in parentheses.
Honore (2002) model Honore (2002) model Random effect Tobit Honore (2002) model Honore (2002) model Random effect Tobit
$$Y_{i,t}$$ $$\delta_{CR}^{\text{high}_{i,t}}$$ $$\delta_{CR}^{\text{high}_{i,t}}$$ $$\delta_{CR}^{\text{high}_{i,t}}$$ $$\delta_{LR}^{\text{high}_{i,t}}$$ $$\delta_{LR}^{\text{high}_{i,t}}$$ $$\delta_{LR}^{\text{high}_{i,t}}$$
I II III IV V VI
$$\delta^\text{high}_{i,t-1 \text{ to } t-5}$$ -0.52 0.53 -1.27 0.26** 0.50*** 0.78*
(1.580) (1.621) (1.296) (0.132) (0.128) (0.405)
$$\delta^\text{low}_{i,t-1 \text{ to } t-5}$$ -4.57*** -5.27** -5.14*** -0.60*** -0.72*** -1.03***
(1.731) (2.410) (1.099) (0.146) (0.135) (0.354)
$$Y_{i,t-1 \text{ to } t-5}$$ 0.98*** 0.96*** 0.75*** 0.80*** 0.74*** 0.73***
(0.112) (0.125) (0.082) (0.163) (0.235) (0.223)
$$\log GDP_{i,t-1 \text{ to } t-5}$$ 2.45 1.54 0.30 2.08*** 2.01*** 1.80
(1.906) (2.241) (3.357) (0.568) (0.771) (1.252)
$$\Delta PD/GDP_{i,t-1 \text{ to } t-5}$$ -1.41*** -1.57*** -1.69*** 0.11 0.12 0.09
(0.189) (0.236) (0.270) (0.090) (0.095) (0.081)
$$POLCOMP_{i,t-1 \text{ to } t-5}$$ 0.20 0.34 0.41 0.25*** 0.19* 0.16
(0.227) (0.252) (0.609) (0.081) (0.111) (0.112)
$$INFLATION_{i,t-1 \text{ to } t-5}$$ -0.42* -0.53** -0.78*** 0.12 0.06 0.00
(0.224) (0.215) (0.277) (0.226) (0.232) (0.208)
$$INTRATE_{i,t-1 \text{ to } t-5}$$ 0.07 0.09 0.04 0.12*** 0.01 0.04
(0.186) (0.202) (0.217) (0.020) (0.042) (0.072)
Num of Obs. 875 875 875 118 118 118
Cross-sectional FE Yes Yes Yes Yes Yes Yes
Time-series FE/dummy No Yes Yes No Yes Yes
$$\chi^2$$ 9163.4 59.49 263.8 24.04 27.37 35.43
p-val. 0 0 0.0023 0.0012 0.0007

Table A.3: Volatility and risk-taking: different data and definitions

This table provides the estimated coefficients of (4) to explore other datasets and different definitions of excessive lending and leverage. The dependent variable and the data used is listed at the column header. BIS data covers 37 countries from the 1960s. Schularick and Taylor’s (2012) dataset of annual aggregate bank loans as a ratio to GDP is from 1870 for 14 developed countries. Finally, Lee et al. (2017) data covers 31 countries from 1980s. First, following the Basel Committee on Banking Supervision, we proxy credit expansion with credit-to-GDP ratio gap, defined as the difference between credit to GDP ratio and its trend. Second, similar to Schularick and Taylor (2012), we use the log-growth of the credit-to-GDP ratio. Leverage expansion is proxied analogously. Region and decade fixed effects are included in all of the specifications. For the sake of brevity, the estimated coefficients of fixed effects are omitted. The standard errors, reported in parentheses, are robust and dually clustered at the year and country level.
BIS data Schularick and Taylor (2012) data Schularick and Taylor (2012) data Lee et al. (2017) data
$$Y_{i,t}$$ $$CR\_GAP_{i,t}$$ $$CR\_GAP_{i,t}$$ $$\delta_{CR}^{\text{high}_{i,t}}$$ $$LR\_GAP_{i,t}$$
I III IV I
$$\delta^\text{high}_{i,t-1 \text{ to } t-5}$$ -1.66 -37.65* -0.38* 1.28***
(2.210) (21.966) (0.219) (0.471)
$$\delta^\text{low}_{i,t-1 \text{ to } t-5}$$ -4.53*** -39.53** -0.40** -1.51***
(1.328) (16.842) (0.168) (0.370)
$$Y_{i,t-1 \text{ to } t-5}$$ 0.44*** 0.46** 0.46*** 0.27
(0.083) (0.181) (0.146) (0.229)
$$\log GDP_{i,t-1 \text{ to } t-5}$$ 0.33 85.54* 0.86* 1.85
(3.173) (44.421) (0.444) (1.164)
$$\Delta PD/GDP_{i,t-1 \text{ to } t-5}$$ -1.88*** -4.64 -0.05 0.16*
(0.540) (3.356) (0.033) (0.090)
$$POLCOMP_{i,t-1 \text{ to } t-5}$$ 0.60 5.48* 0.05* 0.08
(0.637) (3.075) (0.031) (0.068)
$$INFLATION_{i,t-1 \text{ to } t-5}$$ -0.56 17.41*** 0.17*** -0.01
(0.462) (5.067) (0.049) (0.145)
$$INTRATE_{i,t-1 \text{ to } t-5}$$ -0.16 13.79** 0.14** 0.01
(0.356) (6.804) (0.068) (0.057)
Num of Obs. 875 877 877 118
$$\text{Adj.} R^2$$ 0.274 0.207 0.207 0.0598

Table A.4: Robustness: Volatility channels and financial crises

This table presents the unabridged version of Table 8 in the paper. In Column I, high and low volatility are defined as the deviation of volatility level from its historical mean calculated as the average volatility during the past ten years. In Column II, high and low volatility are defined as the deviation of volatility level from a one standard deviation band. In Column III, we include high and low counterparts of the control variables, all defined analogously to high and low volatility. Similarly in Column IV, we include high and low dividend yields as regressors. In Column V, volatility is calculated by employing monthly returns up to December (end year) instead of mid-year returns. In Column VI, we measure volatility as the sum of absolute monthly returns and in Column VII, we calculate annual volatility using a GARCH(1,1) framework. In Column VIII, we repeat the analysis without any fixed effects. In Column IX, we report the results from OLS regressions. In Column X, we present the results when the lag of the dependent variable is excluded. In Column XI, we include the trend estimated through an HP filter in the regression along with high and low volatility variables. In Columns XII and XIII, we include the interest rate and credit-to-GDP gap as control variables, respectively. In Column XIV, we report the results when the smoothing parameter of the HP filter is set to 100 instead of 5000. Finally, in Column XV, we report the results when we merge the crisis database of Reinhart and Rogoff (2009) with that of Bordo et al (2001), Laeven and Valencia (2008), Gourinchas and Obstfeld (2012), and Schularick and Taylor (2012). The panel covers 60 countries and spans 1800–2010. The standard errors, reported in parentheses, are robust and dually clustered at the year and country level.
Dep. Var.: $$C^\text{Banking}_{i,t}$$ Historical mean Band High/low macro vars. High/low dividends 12M ABS GARCH No FEs OLS No lagged dep. var. Trend included Int. rates included Credit included $$\lambda=100$$ Merged data
I II III IV V VI VII VIII IX X XI XII XIII XIV XV
$$\delta^\text{high}_{i,t-1 \text{ to } t-5}$$ 0.17* 0.23* 0.35** 0.13 0.08 0.21 0.32** 0.15 0.01 0.14 0.23 0.35 0.31 0.04 0.19
(0.088) (0.124) (0.160) (0.171) (0.129) (0.162) (0.159) (0.105) (0.006) (0.131) (0.151) (0.226) (0.261) (0.178) (0.131)
$$\delta^\text{low}_{i,t-1 \text{ to } t-5}$$ -0.29** -0.49** -0.42** -0.44** -0.21** -0.35** -0.61*** -0.24** -0.01** -0.22** -0.41** -0.48*** -0.51*** -0.37** -0.30**
(0.118) (0.217) (0.193) (0.174) (0.103) (0.144) (0.192) (0.094) (0.006) (0.099) (0.206) (0.154) (0.137) (0.184) (0.118)
$$C_{i,t-1 \text{ to } t-5}$$ -8.34*** -8.29*** -8.58*** -9.92*** -7.90*** -7.85*** -10.55*** -3.79** -0.31*** -7.86*** -7.48** -7.98** -7.62*** -8.23***
(1.969) (1.953) (2.573) (2.889) (1.881) (2.020) (1.352) (1.931) (0.067) (2.062) (3.583) (4.061) (1.877) (2.020)
$$\log GDP_{i,t-1 \text{ to } t-5}$$ 0.18 0.16 0.23 0.06 0.06 0.64** 0.05 0.01 0.09 0.07 0.08 0.16 0.03 0.05
(0.195) (0.178) (0.406) (0.221) (0.230) (0.312) (0.257) (0.007) (0.176) (0.229) (0.377) (0.516) (0.213) (0.235)
$$\Delta PD/GDP_{i,t-1 \text{ to } t-5}$$ -0.07*** -0.07*** -0.05* -0.06** -0.07** -0.10*** -0.03** -0.00** -0.08*** -0.07*** -0.07** -0.04 -0.07** -0.07**
(0.025) (0.026) (0.029) (0.025) (0.027) (0.032) (0.013) (0.001) (0.020) (0.024) (0.034) (0.059) (0.027) (0.028)
$$POLCOMP_{i,t-1 \text{ to } t-5}$$ -0.08* -0.09** -0.11 -0.09** -0.09* -0.06 -0.06 -0.00* -0.07* -0.08* -0.19** -0.10 -0.09* -0.09*
(0.047) (0.043) (0.075) (0.043) (0.047) (0.067) (0.049) (0.002) (0.041) (0.048) (0.088) (0.108) (0.047) (0.047)
$$INFLATION_{i,t-1 \text{ to } t-5}$$ 1.80* 1.79* 0.03** 0.02* 0.02* 0.03** 0.01* 0.00* 0.01 0.02* 0.00 0.04** 0.02 0.02
(0.952) (0.939) (0.016) (0.011) (0.011) (0.013) (0.005) (0.001) (0.009) (0.014) (0.015) (0.015) (0.011) (0.011)
$$\log GDP^\text{high}_{i,t-1 \text{ to } t-5}$$ 0.15
(2.383)
$$\Delta PD/GDP^\text{high}_{i,t-1 \text{ to } t-5}$$ 0.02
(0.028)
$$POLCOMP^\text{high}_{i,t-1 \text{ to } t-5}$$ -0.22
(0.209)
$$INFLATION^\text{high}_{i,t-1 \text{ to } t-5}$$ 3.25
(2.354)
$$\log GDP^\text{low}_{i,t-1 \text{ to } t-5}$$ 13.15***
(4.072)
$$\Delta PD/GDP^\text{low}_{i,t-1 \text{ to } t-5}$$ 0.02
(0.042)
$$POLCOMP^\text{low}_{i,t-1 \text{ to } t-5}$$ -0.03
(0.204)
$$INFLATION^\text{low}_{i,t-1 \text{ to } t-5}$$ 1.42
(2.006)
$$dy^\text{high}_{i,t-1 \text{ to } t-5}$$ -0.02
(0.016)
$$dy^\text{low}_{i,t-1 \text{ to } t-5}$$ -0.00
(0.008)
$$\tau_{i,t-1 \text{ to } t-5}$$ -0.06
(0.094)
$$INTRATE_{i,t-1 \text{ to } t-5}$$ -0.02
(0.020)
$$CR\_GAP_{i,t-1 \text{ to } t-5}$$ 0.04**
(0.018)
Num of Obs. 2,168 2,164 1,976 1,248 2,181 2,134 1,618 2,886 2,886 2,134 2,134 1,205 1,047 2,134 2,134
$$\text{Pse.} R^2 - \text{adj} R^2$$ 0.102 0.103 0.108 0.101 0.102 0.105 0.123 0.029 0.041 0.067 0.107 0.123 0.114 0.103 0.111