Financial crises, Business cycles, and Bankruptcies in the Very - - PowerPoint PPT Presentation

Financial crises, Business cycles, and Bankruptcies in the Very Long Run: France during the 19th Century

Vincent Bignon

Goal

• Compile a bankruptcy rate in France,

1820-1913

• Study its short-term fluctuations
• Link them with the varying policy of the

central bank during crises

Roadmap

• Computing the bankruptcy rate: the stock
• f firms (and what type of legal statute)
• Extracting the short-term evolutions
• Compares them with other business

fluctuations indicators

• Looks at 19th century financial crises in

France

• Links the pattern with the changes in the

LLR policy of the Banque de France

Hunting the primary numbers

• Bankruptcy numbers for each year

– Taken in the Comptes de la justice civile et commerciale (1840-1913) and archives

• Number of firms that may file for

bankruptcy

– Excluded the agriculture and (sometimes) Professions libérales – Use a fiscal source: la Patente tax (revolutionary tax on each business selling something on the market) except farmers and some professions libérales

Adjusting the series # of firms for legal changes

• Generate spurious fluctuations of the BR

– Tax evasion and the 1841 census – Changes in geographic borders: 1860 (Savoie), 1870 (Alsace) – Fiscal reforms: 1844, 1850, 1858, 1862, 1868

• Fiscal reforms modified the population liable to

tax’s payment:

– Commissioned workers – Professions libérales

• Corrections were implemented

What types of firms?

• legal framework: 1807 code de Commerce
• Individual private firms (unlimited liabilities)
• Partnerships

– ordinary partnership (société en nom collectif): at least 2 individuals liable on personal wealth – limited partnership (société en commandite simple): general partner(s) – manage with unlimited liability – and special partner(s) – limited liability – Joint-stock companies, i.e. limited partnership with

• shares. Liability of partners identical to limited

partnerships but shares are tradable – Public company (société anonyme): liabilities limited to capital contribution. Before 1867, required gvt agreement

Share of companies in the total # of firms

0% 2% 4% 6% 8% 10% 12% 20 30 40 50 60 70 80 90 00 10 Partnership Public companies (limited liabilities)

Share of the limited liabilities companies in the total # of firms

0.0% 0.2% 0.4% 0.6% 0.8% 1.0% 1.2% 1.4% 1825 1850 1875 1900

French bankruptcy rate

.1% .2% .3% .4% .5% .6% .7% 20 30 40 50 60 70 80 90 00 10

Long-run vs. short-run fluctuations

• f the bankruptcy rate
• Liquidity dry-ups:

– likely impacted the BR in the short run (chain

• f default)

Long-run vs. short-run fluctuations

• f the bankruptcy rate
• Liquidity dry-ups:

– likely impacted the BR in the short run (chain

• f default)

– But they likely do not have a long-run impact

Long-run vs. short-run fluctuations

• f the bankruptcy rate
• Liquidity dry-ups:

– likely impacted the BR in the short run (chain

• f default)

– But they likely do not have a long-run impact

• Similar to the impact of monetary policy on

GDP

Long-run vs. short-run fluctuations

• f the bankruptcy rate
• Liquidity dry-ups:

– likely impacted the BR in the short run (chains

• f defaults)

– But they likely do not have a long-run impact

• Similar to monetary policy impact on GDP
• Key question is whether of more

refinancing during crises impacts on the moral hazard (and then on long-run evolution):

Long-run vs. short-run fluctuations

• f the bankruptcy rate
• Liquidity dry-ups:

– likely impacted the BR in the short run (chains

• f defaults)

– But they likely do not have a long-run impact

• Similar to the impact of monetary policy
• Key question is whether of more

refinancing during crises impacts on the moral hazard (and then on long-run evolution): will show this is unlikely to have

• ccured

Long-run vs. short-run fluctuations

• f the bankruptcy rate
• Liquidity dry-ups:

– likely impacted the BR in the short run (chain of default) – But they likely do not have a long-run impact

• Similar to the impact of monetary policy
• Key question is whether of more refinancing

during crises impacts on the moral hazard (and then on long-run evolution): will show unlikely ⇒ Use of filtering method to separate short-run from long-run components

Vanishing fluctuations of the BR

• 40
• 30
• 20
• 10

10 20 30 40 20 30 40 50 60 70 80 90 00 10

Cyclical component computed with HP filter, Lambda set to 6.5

Deviation of the cyclical component of the bankruptcy rate to its trend (HP filter)

• .8
• .6
• .4
• .2

.0 .2 .4 .6 20 30 40 50 60 70 80 90 00 10 Deviation from trend of the bankruptcy rate (HP 100) Deviation from trend of the bankruptcy rate (HP 6.5) Growth rate of the bankruptcy rate

Vanishing fluctuations of the BR

.00 .01 .02 .03 .04 .05 .06 .07 20 30 40 50 60 70 80 90 00 10 Moving variance (10 yrs) of the dev iation of the BR (in %) to its trend

Moving variance of the deviation of the bankruptcy rate (10 yrs)

Removing the outliers (+ 2 st. dev)

.00 .01 .02 .03 .04 .05 .06 .07 20 30 40 50 60 70 80 90 00 10 Moving variance of the deviation of the BR (in %) to its trend Mov ing variance of the dev of the BR (in %) to its trend ('31, '33-4, '47 amd 70-1 excl)

• .4
• .3
• .2
• .1

.0 .1 .2 .3 .4 .5 20 30 40 50 60 70 80 90 00 10 Deviation to the log trend of IP (base 1) Deviation to the log trend of the BR (in % plus 1)

IP and BR Deviations compared

0.0 0.2 0.4 0.6 0.8 1.0 1.2 20 30 40 50 60 70 80 90 00 10 Industrial production (base 1 in 1906-12) Bankruptcy rate in %

Evolution IP index and BR

• .4
• .3
• .2
• .1

.0 .1 .2 .3 .4 .5 20 30 40 50 60 70 80 90 00 10 Deviation of the BR (in %) to its trend Deviation of the wheat price to its trend

Deviations of BR vs wheat price

.00 .01 .02 .03 .04 .05 .06 .07 .08 .09 20 30 40 50 60 70 80 90 00 10 Moving variance (10 yrs) of the dev of the BR (in %) to its trend (outliers excl.) Moving variance (10 yrs) of the deviation of the BR (in %) to its trend Moving variance (10 yrs) of the deviation of wheat price to its trend

Comparing it with the wheat price

.00 .01 .02 .03 .04 .05 .06 .07 20 30 40 50 60 70 80 90 00 10 Moving variance (10 yrs) of the dev of the IP (base 1) to its log trend Moving variance (10 yrs) of the deviation of the BR (in %) to its trend Moving variance (10 yrs) of the dev of the BR (in %) to its trend (outliers excl.)

Moving variance IP vs. BR

What about financial crises?

• Extract financial crises using

– Stock prices index (Arbulu, 2006) – 3 months interest rate (offshore –London – before 1870) and onshore after) – CB liquidity ratio: banknote to metallic reserves – CB refinancing ratio: discount to metallic reserves

• Shows that crisis occurred regularly
• Indicators give broadly the same crises’

years

Deviations of the BR and crises

• 40%
• 30%
• 20%
• 10%

0% 10% 20% 30% 40% 20 30 40 50 60 70 80 90 00 10

Deviations (in %) of bankruptcy rate to its trend Deviations of the bankruptcy rate (HP filtered) to its trend and Juglar's criterion of crisis

Central bank refinancing

• Main instrument: bills of exchanges

Central bank refinancing

• Main instrument: bills of exchanges
• 19th c. changes of monetary policy stance

Central bank refinancing

• Main instrument: bills of exchanges
• 19th c. changes of monetary policy stance
• Two constraints were removed

Central bank refinancing

• Main instrument: bills of exchanges
• 19th c. changes of monetary policy stance:

Two constraints were removed

– From a policy of rationing during crises (up to the 1850s) to a policy of rediscounting at will (Bignon, Flandreau and Ugolini, 2010) : increase of metallic reserve

Central bank refinancing

• Main instrument: bills of exchanges
• 19th c. changes of monetary policy stance:

Two constraints were removed

– From a policy of rationing during crises (up to the 1850s) to a policy of rediscounting at will (Bignon, Flandreau and Ugolini, 2010)

• increase of metallic reserves

Central bank refinancing

• Main instrument: bills of exchanges
• 19th c. changes of monetary policy stance:

Two constraints were removed

– From a policy of rationing during crises (up to the 1850s) to a policy of rediscounting at will:

• increase of metallic reserve

– Change in monetary policy implementation with the development of a network of branches

• decentralized refinancing with tight bills’ screening

• 4,000
• 2,000

2,000 4,000 4,000 8,000 12,000 16,000 20,000 24,000 28,000 20 30 40 50 60 70 80 90 00 10 Total of discounts and advances Trend (HP filter, lambda=6.5) Cycle

Evolution CB refinancing (discounts and advances)

1,000 2,000 3,000 4,000 5,000 20 30 40 50 60 70 80 90 00 10 C.B. gold and silver reserves (millions of francs)

Metallic reserves in the vaults of the central bank

• .6
• .4
• .2

.0 .2 .4 .6 .8 20 30 40 50 60 70 80 90 00 10 Deviation from trend of the total of discounts and advances

Deviation of CB refinancing

Central bank’s branching development

50 100 150 200 250 1836 1844 1852 1860 1868 1876 1884 1892 1900 1908

0% 2% 4% 6% 8% 10% 12% 20 30 40 50 60 70 80 90 00 10 Protested bills in the Banque de France's balance sheet Protested bills in the French economy

Moral hazard and protested bills

100 200 300 400 500 1,000 1,500 2,000 2,500 3,000 3,500 20 30 40 50 60 70 80 90 00 10 Bankrupted financial intermediaries # of banks

Financial intermediary bankruptcies

Bankruptcy rate of financial intermediaries

Bankruptcy rate financial intermediaries

19.24% 22.21% 4.55% 2.01% 4.49% 7.15% 7.79% 0% 5% 10% 15% 20% 25%

1833 1838 1843 1848 1853 1858 1863 1868 1873 1878 1883 1888 1893 1898

Conclusion

• Century-long increase of the BR
• Fluctuations became smoother over time
• They peaked during crises
• Financial crises did not disappear
• But the monetary policy of LLR changed

– Reserves increases allowing more refinancing – Expanding network decentralized both screening and refinancing with very limited risk for the central bank

Corrections to patentes series (1)

• No corrections for the changes of

geographic borders

• Fiscal evasion: for any t < 1842

( )⎟

⎟ ⎠ ⎞ ⎜ ⎜ ⎝ ⎛ + =

−1841 1820 1841 1842 '

1 * x x x x y

t t

Corrections to patentes series (2)

⎪ ⎪ ⎩ ⎪ ⎪ ⎨ ⎧ ≤ < ⎟ ⎟ ⎠ ⎞ ⎜ ⎜ ⎝ ⎛ − = ≤ ⎟ ⎟ ⎠ ⎞ ⎜ ⎜ ⎝ ⎛ − = ≤ ⎪ ⎪ ⎩ ⎪ ⎪ ⎨ ⎧ ≤ < ⎟ ⎟ ⎠ ⎞ ⎜ ⎜ ⎝ ⎛ − = ≤ ⎟ ⎟ ⎠ ⎞ ⎜ ⎜ ⎝ ⎛ − = ≤ ⎪ ⎪ ⎩ ⎪ ⎪ ⎨ ⎧ ≤ < ⎟ ⎟ ⎠ ⎞ ⎜ ⎜ ⎝ ⎛ − = ≤ ⎟ ⎟ ⎠ ⎞ ⎜ ⎜ ⎝ ⎛ − = ≤ 1862 1858 if 000 , 100 1 1858 if 000 , 100 1 1862 any t For 1858 1844 if 000 , 120 1 1844 if 000 , 120 1 1858 any t For 1844 1842 if 000 , 68 1 1842 if 000 , 68 1 1844 any t For

1862 " " 1862 ' " " " 1858 ' " 1858 " ' " 1844 " 1844 ' "

t x x y t x y y t x x y t x y y t x x y t x y y

t t t t t t t t t t t t

.000000 .000004 .000008 .000012 .000016 .000020 .000024 .000028 .000032 20 30 40 50 60 70 80 90 00 10

Real GDP

.0000 .0001 .0002 .0003 .0004 .0005 .0006 20 30 40 50 60 70 80 90 00 10

Index of industrial production

.00001 .00002 .00003 .00004 .00005 .00006 .00007 20 30 40 50 60 70 80 90 00 10

Index of industrial prices

.000 .002 .004 .006 .008 .010 20 30 40 50 60 70 80 90 00 10

Wheat prices

Regression Dev. BR on lagged

• Dev. wheat

0.000344 Prob(F-statistic) 1.514417 Durbin-Watson stat 6.843690 F-statistic

• 1.269272

Hannan-Quinn criter. 63.13338 Log likelihood

• 1.202972

Schwarz criterion 1.295593 Sum squared resid

• 1.314075

Akaike info criterion 0.122740 S.E. of regression 0.134285 S.D. dependent var 0.164563

• Adj. R-squared
• 0.001419

Mean dependent var 0.192724 R-squared 0.9472

• 0.066413

0.012943

• 0.000860

C 0.0140

• 2.508997

0.103232

• 0.259010

DEV WHEAT(-3) 0.4862 0.699413 0.109554 0.076624 DEV WHEAT(-2) 0.0074 2.744666 0.103613 0.284384 DEV WHEAT(-1) Prob. t-Statistic

• Std. Error

Coefficient Variable

Included observations: 90 after adjustments Sample (adjusted): 1823 1912 Method: Least Squares Dependent Variable: DEVTXFAILLJPERCEN

Regression Dev. BR on lagged

• Dev. Wheat and Dev BR

0.000229 Prob(F-statistic) 2.043533 Durbin-Watson stat 4.940200 F-statistic

• 1.260281

Hannan-Quinn criter. 67.24087 Log likelihood

• 1.144256

Schwarz criterion 1.182571 Sum squared resid

• 1.338686

Akaike info criterion 0.119364 S.E. of regression 0.134285 S.D. dependent var 0.209881 Adjusted R-squared

• 0.001419

Mean dependent var 0.263147 R-squared

0.9473

• 0.066276

0.012590

• 0.000834

C 0.0334

• 2.163782

0.112353

• 0.243108

DEVWHEAT(-3) 0.8190 0.229577 0.113548 0.026068 DEVWHEAT(-2) 0.1199 1.571571 0.110009 0.172887 DEVWHEAT(-1) 0.4957

• 0.684295

0.110590

• 0.075676

DEVTXFAILLJPERCEN(-3) 0.9110

• 0.112062

0.117761

• 0.013197

DEVTXFAILLJPERCEN(-2) 0.0108 2.608348 0.114174 0.297805 DEVTXFAILLJPERCEN(-1) Prob. t-Statistic

• Std. Error

Coefficient Variable

Included observations: 90 after adjustments Sample (adjusted): 1823 1912 Method: Least Squares Dependent Variable: DEVTXFAILLJPERCEN

Log IP vs. log BR (+1)

.1 .2 .3 .4 .5 .6 .7 .8 20 30 40 50 60 70 80 90 00 10 Log IP base 1 (plus 1) Log BR (in % plus 1)

20% 40% 60% 80% 100% 120% 140% 20 30 40 50 60 70 80 90 00 10 Ratio of banknotes to the metallic reserves of the central bank Vertical lines are crisis' years according to Juglar's criterion

Crisis years according to liquidity ratio of the CB

3 months interest rate in London

2% 3% 4% 5% 6% 7% 8% 9% 10% 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 20 30 40 50 60 70 80 90 00 10 3 months implicit interest rate (maximum monthly value, left axis) Ratio of discounted bills to the metallic holdings (right axis) Vertical lines are crisis' years according to Juglar's criterion

Short-term interest rate peaks

• 1%

0% 1% 2% 3% 4% 5% 20 30 40 50 60 70 80 90 00 10 Mean minus 1 standard deviation Mean plus 1 standard deviation Short term interest rate (filtered for its decreasing trend)

• 40
• 30
• 20
• 10

10 20 30 40 20 30 40 50 60 70 80 90 00 10

Cyclical component computed with HP filter, Lambda set to 6.5

Deviation of the cyclical component of the bankruptcy rate to its trend (HP filter)

CB advances and discounts

75% 80% 85% 90% 95% 100% 20 30 40 50 60 70 80 90 00 10 share of the discounts in the total of advances and discounts