Download The French Stock Market in War

Monetary and Financial Structures:
The Impact of Political Unrests and Wars
Université Paris 10 Nanterre
19-20 Juin – June 19th-20th
The French Stock Market in War
David Le Bris1
Abstract :
It is interesting to identify the extreme cases of changes in the stock exchange caused by wars.
Stock return depends on how the war is financed. The Franco-Prussian war was financed only
by regular debt thus stocks reflected only situations of real activity. WWI was partially
financed by short term debt; however, since capital markets kept their freedom, stock prices
can be adjusted. WWII and the beginning of German occupation was also financed by short
term debt from the Banque but in a closed market: stocks prices increased. War affects not
only return but structural characteristics of the market too. The Franco-Prussian
war
caused a durable all high interest rate, WWI changed the amplitude of stock movements and
WWII affected the components of markets. After WWI and WWII, the importance of stock
markets in the economy decreased. WWI is the major event affecting markets during the
twentieth century; a contrefactual hypothesis of a quick German victory can help to
understand the real price of victory.
1
David Le Bris is a PhD candidate in economic history. Dissertation under Prof. Dominique Barjot (Paris IV) :
150 ans de Cac 40 ou les actions françaises sur le long terme. David Le Bris is a graduate of DEA Economie
Appliquée (Sciences-Po Paris), DESS Banques et Finances (Toulouse I), Maîtrise en droit international
(Toulouse I), and Institut d’Etudes Politiques de Toulouse. Many thanks to Kim Phan.
War, followed by defeat, is probably the worst event which can impact the stock
market. Current conflict raises several issues of interest. Fernandez (2008) looks at the
volatility change due to war in Irak. Rigobon and Sack (2005) find an important war risk
premium in US assets just before the beginning of the war. There have been a number of
studies with particular focus on the developments around World War II. Frey and Kucher
(2000) analyse war situations using foreign government bond prices at the Zurich stock
exchange. Oosterlinck (2003) compares prices of Vichy bonds with pre-war French bonds to
measure the legitimacy of Vichy France. Brown and Burdekin (2002) study German bonds
traded in London. Occhino, Oosterlinck and White (2007) show how occupied France
financed its own exploitation. Marseille (2000) finds that several industries, like construction
firms, have a better performance than the market.
Most of these studies use market prices to understand war. This paper focuses on wars
to understand stock returns. Confident stock market data exists since 1871 for the United
States thanks to the precursor work of Alfred Cowles (1939), recently completed by Siegel or
Goetzmann, Ibbostson and Peng (2000) for 1815-1871 period. The very specific economic
history of the United States, a winner one, leads to an investigation into other countries.
Dimson, Marsh and Staunton (2002) compile a large panel of stock indices but without the
quality of US data. They found that in Europe too, despite wars, stocks performed well in the
20th century. France is a good point of comparison with the US data due to three wars, not
only victorious, but on national territory and also because of the major role of the Paris stock
market before 1914.
This study uses a new homogeneous stock index for the French stock market from
1854 to 1998.2 Basically, the “Cac 40” is a monthly index of the 40 most prominent shares
among French firms, ranked (each year) by market capitalisation, thus avoiding survivor’s
bias. The index is weighted by these capitalisations. 40 firms are few compared to the number
of quoted stocks but these 40 represent a major part of the total market capitalisation. Today,
the Euronext’s Cac 40 represents about 70 % of the French market capitalisation. It was the
same or even better during war periods. In reality, the distribution of the firm’s size is not
linear. Focusing on the forty first firms, the first one has a weighting of 12,73 % on average
2
See, Le Bris D. and Hautcoeur P-C., “A challenge to triumphant optimists? A new index for the Paris stock
exchange (1854-2007)”, Working Paper, Paris School of Economics, n. 2008-21, 2008 to have a presentation of
this index. http://www.pse.ens.fr/document/wp200821.pdf
2
from 1854 to 2007. The top ten represent more than 60 % whereas the final ten, less than 8 %.
It has remained very stable over time; today3, the first French capitalisation (Total) has a
weighting of 12,48 % while the last one (Air France) only 0,54 % but it has probably not
remained stable internationally. To achieve the same 70 % of the total market capitalisation,
more than 40 firms in United States or Japan are probably needed but fewer in Belgium,
Switzerland or Finland. Thus, for France, with these 40 firms, a correct measure of the stock
exchange is obtained.
18%
Average weight by rank (1854 to 2007)
16%
Average weignt + one standard deviation
Average weight - one standard deviation
14%
AVERAGE WEIGHT BY RANK from 1854 to 2007
RNK Weight Cumulate RNK Weight Cumulate RNK Weight Cumulate RNK Weight Cumulate
12%
1
2
3
4
5
6
7
8
9
10
Weight
10%
8%
6%
12,73%
9,20%
7,73%
6,75%
5,58%
4,71%
4,15%
3,67%
3,29%
3,05%
12,73%
21,93%
29,67%
36,41%
41,99%
46,70%
50,85%
54,52%
57,80%
60,86%
11
12
13
14
15
16
17
18
19
20
2,77%
2,46%
2,27%
2,08%
1,94%
1,81%
1,69%
1,59%
1,51%
1,42%
63,63%
66,08%
68,35%
70,43%
72,37%
74,18%
75,87%
77,45%
78,97%
80,39%
21
22
23
24
25
26
27
28
29
30
1,35%
1,30%
1,24%
1,20%
1,15%
1,12%
1,07%
1,04%
1,01%
0,97%
81,74%
83,04%
84,28%
85,48%
86,63%
87,75%
88,82%
89,87%
90,87%
91,85%
31
32
33
34
35
36
37
38
39
40
0,94% 92,78%
0,91% 93,69%
0,88% 94,57%
0,85% 95,42%
0,82% 96,25%
0,81% 97,05%
0,78% 97,83%
0,76% 98,59%
0,72% 99,31%
0,69% 100,00%
4%
2%
0%
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40
Rank
Figure 1, Firm’s index weight by rank (average 1854-2007)
(Source : author)
Using this database, index and components of the index, we are focusing on the stock
market during war. This paper seeks to measure the impact of war on stock prices and to
understand this impact looking at the way the war is financed. We observe that war affects not
only stocks prices but the regime of the stock market: a high interest rate after 1871, the end
of monetary stability with WWI and the partial nationalisation of the market after WWII.
Consequently, the role of the stock market in the economy decreased after the two World
Wars. We are going to try to measure the real financial cost of WWI using a hypothetical
model of a French defeat in September 1914.
3
05/02/2008
3
THE PERFORMANCE OF STOCKS DEPENDS ON HOW THE WAR IS
FINANCED
Three wars, three different reactions
France suffers three wars on its own national territory: one clear defeat (1870), one
very difficult victory (WWI) and one defeat with a pseudo-victory (WWII). Stock market
prices present similar movements prior to war declaration. Prices are stable, the market seems
to not expect serious negative consequences from the war. Starting from a year before the
declaration of war, prices decrease about 10 % before WWI and WWII and keep perfectly
stable during the same one year period before the Franco-Prussian war. Once the war starts,
variations differ strongly. After the defeat, stock prices fall 20 % in 1871. At the beginning of
WWI, stock prices fall about 15 %. It is a very different scenario for WWII. Two months after
the declaration of war, during the “drôle de guerre” and after the armistice, stock prices rise
very quickly, with a weak impact from the military defeat. This rise in nominal prices during
WWII is very homogeneous until January 1943. Since the declaration of war, prices are
multiplied by 4. After that date, the trend is reversed but the coefficient is always 2,5 at the
end of the war.
5,0
Franco-prussian (january 1869-december1876)
4,5
January 1943
WWI (january 1913-december 1920)
4,0
June 1944
Paris Liberation
WWII (january 1938-december 1945)
Stock price
3,5
3,0
WAR
2,5
"Drôle de guerre"
End of WWII
French defeat
2,0
1,5
End of WWI
1,0
75
72
69
66
63
60
57
54
51
48
45
42
39
36
33
30
27
24
21
15
9
12
6
3
0
-3
-6
-9
-1
8
-1
5
-1
2
18
End of franco-prussian war
0,5
Month
Figure 2, Price reaction of the Cac 40 during wars
(Source : author)
4
Nominal and real prices
Nominal and real prices are definitively different. These stocks prices are those quoted
in the stock market: they are nominal. With the value of the French franc falling, the real
performance is not the same. It is reasonable to assume that investors are more concerned
with real performances than with nominal ones. Only the capital value is measured, the total
return of a stock holder is different since a dividend exists. Dividends or coupons are not used
since yield impact is very weak when price variations are so large. It is difficult to measure
inflation, even more so during a war with the generalisation of the black market. In order to
measure all the “war effect”, it is necessary to look at the resulting situation a few years after
the war. The years immediately following the end of the war must be included in the analysis
of its effects. To understand stock performances, State bond and gold prices are also presented
here.
Three wars, two near total losses. The Franco-Prussian war causes a 40 % loss for both
stocks and bonds. Nominal and real performances are similar since inflation does not really
exist. One year after, the loss in the value of stocks is only 20 % but 30 % for bonds. At the
end of WWI, the decrease in value is about 65 % for stocks and more than 70 % for bonds.
Two years after, it is worse: 75 % for stocks and 80 % for bonds. WWII is different with a
nominal increase but a terrible real performance. On the top of stock prices, the real
performance is + 75 % but at the end of the war, the real variation is – 40 % and – 70 % for
bonds. A few years after, in January 1949, the loss is more or less complete with – 88 % for
stocks and – 95 % for bonds since in 1946, 1947 and 1948, the inflation rate is above 50 %
each year.
During these two last wars, gold achieves very attractive performances. Two years
after the end of WWI, gold holders have the same real value as prior the war. The highest
value of stocks in January 1943 of + 75 % pales in comparison to the + 500 % achieved by
gold! At the end of WWII, gold keeps a 400 % increase and 20 % in January 1949.
5
January 1870
January 1871
January 1872
january 1914
january 1915
january 1916
january 1917
january 1918
january 1919
january 1920
january 1921
january 1939
january 1940
january 1941
january 1942
january 1943
january 1944
january 1945
january 1946
january 1947
january 1948
january 1949
PRICES IN WAR
Stocks
Rente 3 %
Nominal
Real
Nominal
Real
Franco-Prussian war
100
100
100
100
74,11
63,68
69,84
60,02
88,53
81,56
75,79
70,29
WWI
100
100
100
100
85,61
71,94
84,62
71,11
74,83
55,84
74,82
55,84
83,55
52,22
72,77
45,48
84,62
40,88
68,66
33,17
90,23
34,84
72,24
27,89
94,18
26,38
69,72
19,53
84,36
27,04
68,31
21,89
WWII
100
100
100
100
111,46
93,91
90,44
76,20
160,17
115,07
102,35
73,53
241,36
144,33
113,08
67,62
367,00
176,72
114,96
55,35
299,47
117,98
113,61
44,76
262,32
69,66
117,89
31,31
208,72
36,33
116,07
20,20
263,91
30,79
105,34
12,29
212,39
15,62
78,01
5,74
186,06
12,09
73,31
4,76
Gold
Nominal
Real
100
100
100
100
85,93
92,74
100
100,00
100,00
100,00
100,00
105,00
225,00
304,00
100
84,03
74,63
62,50
48,31
40,54
63,03
97,44
100
112,08
575,10
1 082,45
1 237,14
1 514,69
1 676,73
2 193,88
1 371,43
1 906,12
1 857,14
100
94,43
413,17
647,30
595,72
596,72
445,27
381,89
160,02
140,14
120,65
Figure 3, Nominal and real prices during and after wars
(Source : INSEE, author)
Three different ways to finance wars, three different impacts on stocks prices
These different price reactions during these wars can be explained by the method used
to finance each war.
The main cost of the Franco-Prussian war is an indemnity financed by standard debt
(Emprunt Thiers). 5 billion francs of new debt between 1871 and 1872 is just equal to the
market capitalisation of the Cac 40 in January 1871. The comparison with the stock market
capitalisation is used to represent the size of these amounts. Thus, the “financial cost” of the
war is only about 100 % of the market capitalisation. This new debt causes a fall in bonds
prices but financial stability is maintained.
For WWI, it is very different: it is the birth of inflation. Before 1914, the new public
debt is non-existent. The beginning of the war is financed by short term debt (1/3 from the
6
Banque de France) but from 1916, long term bonds issues are massive. Between 1915 and
1924, new “Rente” is about 45 billion franc-or. It is nearly double of all issues during the
preceding century. Between 1798 and 1914, total new “Rente” is only 26,3 billion francs
(nominal capital).4 The next figure shows, for each year, the growth of debt (short and long
term)5 divided by market capitalisation of the Cac 40 at the beginning of the year.
350%
New short term debt / stock market
capitalization
New long term debt / stock market
capitalization
300%
250%
200%
150%
100%
50%
0%
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
-50%
Figure 4, New public debt on stock market capitalisation at the beginning of the year during WWI
(Source: INSEE, author)
War causes different impacts on securities. The impact on bonds is direct because of
the increase in interest rate. Massive issues of long-term public debt should have a negative
impact on stock prices since it is an alternative choice for savers and a higer cost of debt for
firms. The main effect comes from monetary creations (advances from the Banque de France
and “Bons de la Défense Nationale” during WWI too6) which causes a non-anticipated
inflation whose impact is very strong. Perhaps this kind of inflation is never anticipated since
for savers, it is a period of heavy losses each time. Conventional wisdom suggests that
nominal stock returns should be relatively high when inflation is high and vice-versa, since
stocks represent claims on real assets that should increase value with inflation. From 1926 to
1996, in the United States, with a weak inflation compared to France, there is no discernable
4
Vaslin J-M., « Le siècle d’or de la Rente perpétuelle française », in Gallais-Hamonno G., Le marché financier
français au XIXème siècle, vol. II, Paris, Publications de la Sorbonne, 2007, p. 185
5
INSEE, Annuaire Statistique de la France, 1951
6
Eichengreen shows a strong correlation between inflation and monetary growth (M1 which includes “Bons de
la Défense nationale issued in the early 20’s were sufficiently liquid to serve as close substitute for money”)
among 12 countries over the period 1921-1927. Eichengreen B., Elusive stability, Cambridge, Cambridge
University Press, 1990 p. 27
7
relationship between stock performances and inflation rate7. “Accordingly, stocks are not
good hedges against inflation rate”.8 It is worse in France with high war inflation. Stock
prices never follow inflation.
WWI and WWII differ in the level of capital freedom. During WWI, the majority of
the rules of the market keep safe. Exportation of capital is free and foreign securities can be
bought. The French government only rent or buy foreign securities9 to pay for importations in
dollars. Stock prices adjust without any constraint, real values decrease quickly. It is not the
same during the Second World War. During WWII, all assets are controlled: frontiers are
closed to export money. Money can only buy national assets. It is the famous “circuit”.
Consequently, there is a boom for stock prices due to a lot of new money created in the
country.
War needs money. With the “mobilisation générale”, 25 billion from the Banque de
France is made available for the government.10 Another convention11 sees 20 billion more
being offered by the Banque. This amount is nearly the value of all stocks listed (in January
1939, market capitalisation of the Cac 40 is only 42 billion). In February 1940, 20 billion is
spent by the government; on May 9th, only 21,6 billion is spent but after the defeat (10th June),
this figure rises to 36 billion. This advance rises to 57,7 billion on the 1st August after a new
convention between the government and the Banque de France.
German occupation causes an unbelievable creation of new money. The armistice
convention of the 22nd June 1940 sets an occupation indemnity of 400 million francs a day, to
be paid by the French government. In May 194112, this amount decreases to 300 million a day
but rises to 500 million in winter 194213 when the south of France becomes occupied too. As
a result, together with some additional various expenditure (Italian indemnity, maintenance of
7
Correlation coefficient between stock returns and inflation rate is -0,01. Stocks, Bonds, Bills and Inflation 1997
Yearbook, Chicago, Ibbotson Associates, 1997. Various studies find the same absence of relation in the shortterm at least.
8
Sharpe W., Alexander G., Bailey J., Investments, New Jersey, Prentice Hall, 1999, p. 331
9
Pennsylvania 3 % and Chicago Milwaukee 4 % are bought at issue price by French government in June 1915.
After, it is Central Pacific 4 % and New-York New-Haven 4 % . After 5th May 1916, securities are rented by the
State, compensated by a dividend/coupon majoration of 25 %. A certificate of rent can be exchanged on the
market but few transactions are listed.
10
convention of 29th September, 1938
11
29th February 1940
12
May, 11, 1941
13
November, 11, 1942
8
military bases), the French government transfers 80 billion in 1940, 130 billion in 1941, 124
billion in 1942, 220 billion in 1943 et 142 billion in 1944.14 The total amount is about 700
billion.15 A new kind of advance from the Banque de France is used to pay German troops:
“frais d’entretien des troupes d’occupation”. The total of this new money is about 426 billion
francs compared to 40 billion which is the market capitalisation of the Cac 40 in January
1939. With only the 1940 transfer amount, Germany can buy all the stocks in the Cac 40 two
times over, and 17 times over (700/40) with all the francs transferred during the war.
Avances pour "frais d'entretien des troupes d'occupation"
par la Banque de France (en milliards)
Convention
Approuvé
Publié au
relevant le plafond
date
par loi du
J.O le
plafond de porté à
25 août 1940
25 août 1940
13 septembre 1940
50
50
29 octobre 1940
29 octobre 1940
1 novembre 1940
15
65
12 décembre 1940
16 décembre 1940
17 décembre 1940
8
73
30 décembre 1940
31 décembre 1940
20 janvier 1941
12
85
20 février 1941
22 février 1941
22 mars 1941
15
100
30 avril 1941
3 mai 1941
6 juillet 1941
4
104
10 mai 1941
30 mai 1941
6 juin 1941
4
108
11 juin 1941
11 juin 1941
24 juin 1941
10
118
11 septembre 1941
21 septembre 1941
23 septembre 1941
12
130
27 novembre 1941
9 décembre 1941
12 décembre 1941
12
142
26 décembre 1941
31 décembre 1941
8 février 1942
8
150
5 mars 1942
26 mars 1942
5 avril 1942
10
160
30 avril 1942
4 mai 1942
5 mai 1942
9
169
11 juin 1942
12 juin 1942
13 juin 1942
12
181
17 septembre 1942
28 septembre 1942
1 octobre 1942
15
196
19 novembre 1942
4 décembre 1942
6 décembre 1942
15
211
21 janvier 1943
12 février 1943
17 février 1943
20
231
31 mars 1943
8 avril 1943
10 avril 1943
30
261
8 juillet 1943
10 juillet 1943
8 août 1943
30
291
30 septembre 1943
30 septembre 1943
14 octobre 1943
30
321
16 décembre 1943
20 décembre 1943
5 janvier 1944
30
351
23 mars 1944
30 mars 1944
6 avril 1944
30
381
17 mai 1944
8 juin 1944
16 juin 1944
30
411
20 juillet 1944
non publiée
15
426
Figure 5, Advances from Banque de France to finance German occupation
(Source: Sédillot, 1945)
This new money, without any new wealth being generated in the French economy,
causes price adjustment (inflation) but most of the prices are controlled thus it is a flight for
real assets and a boom for stocks. All free prices rise quickly, carpets or stamps, gold (black
market) or stocks. Stocks with real value (real estate or retail stores with many real estate
properties, like Au Printemps) have the best performances. In January 1943, stocks prices are
4 times more than in 1939. Stock prices start to fall in 1943 when the end of this exceptional
14
Sédillot R., Le franc enchaîné, histoire de la monnaie française pendant la guerre et l’Occupation, Paris,
Sirey, 1945, p. 139
15
Pollard A.A., War, Economy and Society 1939-1945, London, 1977 p. 132 said 600 billion.
9
situation seems to be on the horizon. In January, the first big German defeat in Stalingrad
takes place, and also the fusion between the Giraud and Leclerc armies in Tunisia. We notice
that stock increase is very weak compare to gold prices. It is probably a liquidity premium
(gold can be used to pay anything anytime, unlike stocks) and a security premium (stocks and
firms can support political restriction, not gold). War cause spectacular price variations but
changes the nature of the market too.
STRUCTURAL CHANGES AFTER WAR
Franco-Prussian war increases interest rates for years
In order to pay the indemnity to the Prussians, two big issues (2 and 3 billion) of debt
take place. This 5 billion franc of debt is about 20 % of the French GDP. After this major
transaction, the interest rate decreases more slowly than after previous interest rate peaks in
1840, 1848 or 1852. A new progressive wave of saving can probably explain this gradual
decrease in interest rate. The first "Emprunt Thiers", 2 billion of “Rente 5 %” the 21st June
1871 is sold with an attractive prime: the actuarial rate is 6,5 % when the bond market
indicates 5,8 %. The second one of three billion is the most significant of the whole century, it
is sold with a 6,32 % rate.16 Only ten years after, this rate is under 3,5 %. It is the only one
that experiences a slow decrease during this century.
This slow decrease is accompanied by a symmetric increase for stock prices. This
increase is very regular. While a financial crisis affects international markets in 1873, the
French one is not affected. A decrease in interest rate causes a mechanic growth in stock price
since bonds investment is less attractive, the discount rate decreases and an indirect effect
with the fall in the cost of debt for firms. So a relationship exists probably between this
progressive decrease in interest rate and the growth of stock prices.
16
Vaslin J-M., « Le siècle d’or de la rente perpétuelle française », in Gallais-Hamonno G., Le marché financier
français au XIXème siècle, vol. II, Paris, Publications de la Sorbonne, 2007, p. 182
10
250
6,50%
June 1871 & July 1872
"Emprunt Thiers" for prussian tribute
Top of interest rate
230
6,00%
CAC 40
210
Long terme interest rate (Rente 3 %)
5,50%
5,00%
170
+ 138 %
+ 5,35 %/year
1873
"International crisis"
150
4,50%
Interest rate
Stock price
190
130
4,00%
110
3,50%
90
1871
1885
3,00%
18
71
18
71
18
71
18
72
18
72
18
73
18
73
18
73
18
74
18
74
18
75
18
75
18
76
18
76
18
76
18
77
18
77
18
78
18
78
18
78
18
79
18
79
18
80
18
80
18
81
18
81
18
81
18
82
18
82
18
83
18
83
18
83
18
84
18
84
70
Figure 6, Stock price and interest rate (Rente 3 %) (1871-1885)
(Source: author)
WWI affects the distribution of stocks variations
From WWI, the amplitude of price movements increases. Mean and standard deviation
are 0,10 %/month and 2,20 %/month before 1914 but 0,32 % and 5,74 % after WWI. Of
course the real economic situation is more volatile after 1914 but this probably cannot explain
all this variation in price movement. The end of the franc-or provides a new source of risk:
monetary value. This regime change is clear on the next graph. All monthly price variations of
2007
2003
1999
1995
1992
1988
1984
1980
1976
1972
1969
1965
1961
1957
1953
1949
1946
1942
1938
1934
1930
1926
1923
1919
1915
1911
1907
1903
1900
1896
1892
1888
1884
1880
1877
1873
1869
1865
1861
1857
1854
the Cac 40 since 1854 are shown.
30%
Monthly price variation of the Cac 40
1 écart-type
1914
20%
Monthly price variation
10%
0%
-10%
-20%
-30%
1 8 5 4 -1 9 1 3
M ean
S ta n d a rd D e v ia tio n
K u rt o s is
S kew ness
R ange
M in im u m
M a x im u m
C ount
C o n fid e n c e L e v e l(9 5 ,0 % )
J a rq u e - B e rra
0 ,1 0 %
2 ,2 0 %
1 0 ,2 6
- 0 ,5 8
2 5 ,6 9 %
-1 4 ,4 0 %
1 1 ,3 0 %
719
0 ,0 0 1 6
1619
1 9 1 4 -1 9 7 3
M ean
S ta n d a rd D e v ia tio n
K u rto s is
Skew ness
R ange
M in im u m
M a x im u m
C ount
C o n f id e n c e L e v e l(9 5 ,0 % )
J a rq u e -B e rra
0 ,3 2 %
5 ,7 4 %
4 ,6 3
0 ,4 4
4 6 ,6 4 %
-2 1 ,4 0 %
2 5 ,2 4 %
720
0 ,0 0 4 2
103
-40%
Date
Figure 7, Monthly price variation of the Cac 40 (1854-2007)
(Source: author)
11
One may imagine it is only the magnitude of price variation which increases. Actually
the distribution rules change too. It is not a Gaussian one before or after 1914 since the
“Jarque-Berra” is too strong. A Khi² test makes a comparison possible between the nature of
price variation distribution before and after 1914. Price variations are ranked by a quarter of
standard deviation before and after 1914. This bin range makes a comparison possible
between the distribution rule before and after the war. An identical distribution can be rejected
by this khi² test. Thus the distribution is not Gaussian and is different before and after 1914.
The distribution rule is probably never stable during any long period.
Monthly pricevariation
1854-1913 1914-1973
Mean
0,10%
0,32%
Standard-deviation
2,20%
5,74%
Maximal Bin
-10,13%
34,64%
Minimal Bin
-16,35%
-33,96%
133
khi 2
>5%
43,8
WWII affects the components of the French stock market
WWII has an impact on the component of the stock market. Nationalisations of the
Libération can be considered as a result of the war. This policy was in the program of the
Conseil National de la Résistance. At the very least, the war accelerates this movement.
Different industries are nationalised: banking (4 big companies only), energy (electricity,
coal, gas), insurance (not present in the Cac 40 since this industry is not concentrated). From
the 1939 Cac 40, 28 % is nationalised. It is the first time that a big part of the market
artificially disappears; the nationalisation of the railways in 1936 represents only 5,88 % of
the 1936 Cac 40.
12
RANK
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
CAC 40 IN 1939 WITH FIRMS NATIONIONALIZED (in bold)
Firm
Weight RANK
Firm
21 Mines de Lens SC
Canal maritime de Suez
18,65%
Brasseries Argentine Quilmès
5,78%
22 Electro-Chimie d'Ugine
Crédit Foncier de France
5,61%
23 Compteurs à gaz
24 Union d'électricité
Rhône Poulenc
4,05%
Air Liquide
3,65%
25 Française des Pétroles
Banque de France
3,65%
26 Hts fourneauxde Pont à Mousson
Crédit Lyonnais
3,23%
27 Charbonnages du Tonkin
Le Nickel
2,89%
28 Terres Rouges
29 Mines de Courrières
Saint Gobain
2,86%
Sté Lyonnaise des Eaux et d'Eclairage*
2,64%
30 Orléans
Banque de l'Indo-Chine
2,48%
31 Forges et Aciéries du Nord et de l'Est
Société Générale
2,46%
32 Mines de Marles
Cie Parisienne de distibution d'électricité
3,09%
33 Nord
34 Force motrice La Truyère
Produits chimiques d'Alais et Camargue
2,23%
Etb Kuhlmann
2,02%
35 Cie de Béthune
Kali Ste Thérèse
2,00%
36 Energie electrique du littoral medit
Banque de Paris et des Pays-Bas
1,87%
37 Cie Gale d'Electricité
38 Houilles de Blanzy
Raffinerie Say
1,78%
Paris Lyon Méditérannée
1,72%
39 Sarre et Moselle
Comptoir Nationale d'Escompte
1,69%
40 Port du Rosario
* water part is not nationalized
TOTAL NATIONALIZED
Weight
1,65%
1,61%
1,59%
1,58%
1,45%
1,37%
1,35%
1,33%
1,30%
1,19%
1,17%
1,16%
1,15%
1,15%
1,14%
1,13%
1,12%
1,09%
1,08%
1,05%
28,05%
Figure 8, Cac 40 in 1939 with firms nationalised after WWII
(Source: author)
WWII changes the nature of the stock market. The diversification of the index
decreases since specific industries disappear. For example, the banking industry which had
played a major role in the stock market since 1854 became a minor player. Before WWII, the
weighting of the banking industry (without the Banque de France) in the index is always
above 15 % but after nationalisation it is only about 5 %. A new leading role is played after
WWII by oil firms. After a few decades a new quoted bank industry appears but privatisations
during the 1980s are necessary to restore its place in the market.
50%
45%
Total weight of bank (and insurance since 1980s) in the Cac 40
(without Banque de France)
40%
35%
30%
WWII
25%
20%
15%
10%
5%
20
04
19
99
19
94
19
89
19
84
19
79
19
74
19
69
19
64
19
59
19
54
19
49
19
44
19
39
19
34
19
29
19
24
19
19
19
14
19
09
19
04
18
99
18
94
18
89
18
84
18
79
18
74
18
69
18
64
18
59
18
54
0%
Figure 9, Weight of bank industry in the Cac 40 (1854-1990)
(Source: author)
13
LONG TERM CONSEQUENCES OF WORLD WAR
The role of the French stock market decreases after the World Wars
The World Wars cause a fall in the role of the stock exchange in the economy. WWI
changes the importance of the stock market in the economic system. The ratio of the
capitalisation of the 40 biggest stocks to French GDP was 21 % in January 1914. This ratio is
only 7,71 % in January 1919. The 1914 level was only reached in 1995. Market Capitalisation
on GDP falls again after the WWII from 8,16 % to less than 2 %. For each war, the changes
are the consequences of political choices: nationalisation after WWII and fixed selling prices
with high taxation after WWI.
40%
Market Capitalization of the Cac 40
on GDP
35%
1914
21,57 %
30%
%of GDP
25%
20%
15%
1919
7,72 %
10%
1939
8,16 %
5%
1951
1,82 %
18
53
18
56
18
59
18
62
18
65
18
68
18
71
18
74
18
77
18
80
18
83
18
86
18
89
18
92
18
95
18
98
19
01
19
04
19
07
19
10
19
13
19
16
19
19
19
22
19
25
19
28
19
31
19
34
19
37
19
40
19
43
19
46
19
49
0%
Figure 10, Market capitalisation of the Cac 40 on french GDP (1854-1951)
(Sources: INSEE, author)
The market capitalisation is the value of firms. The value of one firm is the present
value of all future profits of the firm, the value of market capitalisation is the present value of
all future profits of all quoted firms. It is always difficult to appreciate the discounted rate but
dividends are known.
14
500 000 000
WWI
450 000 000
WWI
400 000 000
350 000 000
300 000 000
250 000 000
200 000 000
150 000 000
100 000 000
50 000 000
Dividends payed by Cac 40 firms in francs 1914
18
55
18
58
18
61
18
64
18
67
18
70
18
73
18
76
18
79
18
82
18
85
18
88
18
91
18
94
18
97
19
00
19
03
19
06
19
09
19
12
19
15
19
18
19
21
19
24
19
27
19
30
19
33
19
36
19
39
19
42
19
45
19
48
19
51
-
Figure 12, Dividends payed by Cac 40 firms (1855-1951)
(Source: author)
After WWII, it is normal to have fewer dividends because a part of the stock market is
nationalised, so there is a decrease in the market capitalisation on GDP ratio. It is different
after WWI because all firms are present before and after the war. The production level of
1924 is about 115% of the 1913 one17 but dividends are only 44 % of those during 1913.
Why is this market capitalisation on GDP ratio divided by three after WWI? Some of the
damage for a few firms is due to WWI but real activity of most of them does not change.
This difference between production and dividends can be explained (partially) by
inflation. Stocks are not only a real asset. A well-known explanation is the high taxation level
after war whose impact is multiplied by inflation since the tax rate is applied on nominal
profits.18 Another explanation for the fall in dividends is the weighting of firms with fixed
sale prices. In 1914, 37 % of the market capitalisation have activities with fixed income. All
concession firms (electricity, water, railways) have contracts with political authorities.
Inflation increases costs for these firms but sale prices are fixed by contracts. These firms
cannot perform in the same manner. Ten years after, less than 15 % of the market
capitalisation is firms with "fixed sale prices".
17
18
Sauvy A., Histoire économique de la France entre les deux guerres, Paris, Economica, 1984, Tome 1, p. 39
See Hautcoeur (2001)
15
RANK
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
CAC 40 IN 1914 WITH FIXED SALE PRICES (in bold)
Firm
Weight RANK
Firm
Canal maritime de Suez
9,22%
21 Banque de l'Indo-Chine
Paris Lyon Méditérannée
8,50%
22 Crédit Industriel et Commercial
Nord
7,46%
23 Mines de Vicoigne et Noeux SC
24 Cie Parisienne de distibution d'électricité
Banque de France
7,05%
Crédit Lyonnais
6,96%
25 Banque de l'Union Parisienne
Société Générale
6,79%
26 Penarroya
Orléans
6,60%
27 Mines de Béthune SC
Est
4,43%
28 Banque hypothécaire franco-argentine
Mines de Lens SC
3,75%
29 Forges et Aciéries de la marined'Homécourt
Comptoir Nationale d'Escompte
3,50%
30 Nouvelles Galeries
Crédit Foncier de France
2,93%
31 sté Centrale des banques de Province N
Mines de Courrières
2,45%
32 Mokta-el-hadid
Midi
2,34%
33 Boléo
Ouest
2,24%
34 Procédé Thomson-Houston
Banque de Paris et des Pays-Bas
2,05%
35 Mines de Dourges SC
Schneider et cie
1,69%
36 Sté d'Electricité de Paris
Metropolitain Paris
1,46%
37 Cie Gale des Omnibus de Paris
Cie Gale des Eaux
1,40%
38 Sté Marseillaise de Crédit Industriel
Phosphates de Gafsa
1,38%
39 Crédit Mobilier
Banque de l'Algérie
1,26%
40 Paris-France
TOTAL FIXED SALE PRICES
RANK
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
36,70%
CAC 40 IN 1924 WITH FIXED SALE PRICES (in bold)
Firm
Weight RANK
Firm
Canal maritime de Suez
9,74%
21 Raffinerie Say
Banque de France
6,10%
22 Mines d'Aniche
Saint Gobain
4,56%
23 Mines de Marles
Mines de Lens SC
4,08%
24 Banque Nationale de Crédit
Paris Lyon Méditérannée
3,89%
25 Soie artificielle d'Izieux
Crédit Foncier de France
3,81%
26 Procédés Thomson-Houston
Crédit Lyonnais
3,60%
27 Penarroya
Brasseries Argentine Quilmès
3,58%
28 Phosphates de Gafsa
Mines de Courrières
3,23%
29 Comptoir Lyon-Allemand
Société Générale
3,20%
30 Au Bon Marché
Ouenza
3,04%
31 Cie Parisienne de distibution d'électricité
Nord
3,03%
32 Produits chimiques d'Alais et Camargue
Banque de Paris et des Pays-Bas
2,88%
33 Etb Kuhlmann
Orléans
2,63%
34 Union d'électricité
Banque de l'Indo-Chine
2,44%
35 Banque de l'Union Parisienne
Charbonnages du Tonkin
2,43%
36 Banque hypothécaire franco-argentine
Mines d'Anzin
2,25%
37 Compagnie Algérienne
Est
2,25%
38 Forges et Aciéries de Huta-Bankowa
Comptoir Nationale d'Escompte
2,16%
39 Banque de l'Algérie
Mines de Vicoigne et Noeux SC
1,82%
40 Houilles de Blanzy
TOTAL FIXED SALE PRICES
Weight
1,24%
1,18%
1,12%
1,02%
1,00%
0,92%
0,88%
0,82%
0,81%
0,81%
0,75%
0,74%
0,73%
0,71%
0,70%
0,63%
0,63%
0,62%
0,62%
0,62%
Weight
1,79%
1,78%
1,77%
1,73%
1,67%
1,64%
1,58%
1,55%
1,53%
1,52%
1,47%
1,40%
1,31%
1,30%
1,26%
1,21%
1,21%
1,20%
1,19%
1,15%
14,56%
Figure 13, Cac 40 in 1914 and 1924 with fixed sale prices firms
(Source: author)
The price of victory or what happened if the Germans had won the Marne battle ?
The counterfactual hypothesis is a useful method to understand the impact of the
WWI on financial markets. It has already been shown how this war represents a period of
change on the stock markets. This complete upheaval that occurred in WWI can be
underestimated by the appearance of graphs or returns in the long run that seem to tone down
16
the consequences of this single event. Bordo and Hautcoeur (2007) show why a stabilisation
of the Franc was impossible just after WWI thanks to the public debt amounts but possible in
1924. Alfred Sauvy (1984) estimates losses, caused by post-war inflation, for state
bondholders, of 1 000 billion franc-1929 (172 billion franc-or), close to 333 % of GDP.19 In
the same vein as this estimate, a counterfactual hypothesis allows a greater understanding of
the impact of the war, looking at what would have happened if the conflict had not taken
place or, to put it more precisely, if the Germans had achieved a quick victory.
The First World War could have turned out very differently to how it did. Indeed, at
the battle of the Marne (6th/9th 1914), the Germans very nearly claimed victory due to the
Schlieffen manoeuvre. This battle marks a turning point in the German advance; British
troops were ready to go back to the ports of the English Channel. It required the emergency
dispatch of 10 000 soldiers from the military base of Paris, 6 000 of which were transported
by the famous Marne’s Taxis, as well as a little luck to save the capital from the Germans.
What would have happened if the taxis had not existed or more precisely if the Marne battle
had been lost or perhaps more generally if the WWI had ended on the 10th September 1914,
after only one month of combat, with the victorious arrival of the Germans in Paris and a
humiliating armistice for France? Undoubtedly, the course of history would have been
different on many fronts. This study only seeks to measure the financial cost of the actual
longer war by looking at the financial market values if the Germans had been quickly
victorious.
The model used here is the simpliest possible. It consists of comparing what became
of invested savings in stocks and bonds between 1914 and 1928 with what they would have
become if the 19th century returns had been maintained. The difference represents the
“financial cost” of the war for French savers. 1928 is used here since this year constitutes the
legal end of the Germinal franc and it is frequently assumed that the WWI was financially
settled during this year. The Franc has recovered a kind of stability at this point; however, the
time period starting from 1929 is another story. This model is based upon three hypotheses
which are to be discussed.
19
Sauvy A., Histoire économique de la France entre les deux guerres, Paris, Economica, 1984, Tome 1, p. 290
17
Hypotheses discussion:
-
H1, hypothetical financial returns between January 1914 and December 1928 are
identical to that recorded between January 1854 and December 1913
This is the most important hypothesis of this study. It is impossible to consider any other
situation because the best estimate of what would have been is what has already happened in
the past. In addition, the 59 years between 1854 and 1913 were not particularly successful.
They were affected by significant financial crashes (1857 and 1882), by a major military
defeat (1871) and also by a major economic depression at the end of the century.
The 1928 value of stock market capitalisation in 1914 depends on the rate of revenue
reinvestment. This rate is impossible to calculate and difficult to explain in a theoretical
manner. On one hand, the rate of revenue is the only way to estimate the opportunity cost of
choosing to spend rather than to save revenue. On the other hand, a complete reinvestment
appears not to be very realistic. Siegel20 points out that using the rate obtained by a permanent
reinvestment of dividends in US stocks between 1802 and 1992, a rich man at the beginning
of the 19th century who invested one million dollars (equivalent to 12 millions dollars in
1992) would have found himself with 3 000 billion dollars in 1992. This sum of money would
have allowed him to buy the entire stock market capitalisation of North America! The reality
however is somewhat different with investors sometimes spending their own savings. A factor
α, is therefore included, the rate of revenue reinvested with 0<α<1. The two extreme cases are
to be considered : α=0 et α=1.
-
H2, the currency system remains the franc-or, there is no inflation,
Without a pressing need to finance military spending, there is no reason to believe that a
currency system who had already survived five changes of government, two military defeats
and three revolutions, could not have withstood another military defeat.
-
H3, The French market capitalisations are held by the country’s savers
This is a simplified hypothesis because it is impossible to know precisely the proportion of
French assets that were being held by foreigners in 1913. This does not affect the nature of
this study but maybe its scale. The loss caused by WWI can easily be reduced by the
20
Siegel J., Stocks for the long run, Irvin, New-York, 1994, p. 7
18
percentage corresponding to the loss suffered by foreign savers in possession of French stocks
and bonds in 1913. However, taking into account that France was an exporter of capital, it can
equally be claimed that French savers had more capital at their disposal abroad than foreign
savers in France, and that these foreign investments could be subject to the same reasoning.
Would Russian debt have been paid back if the Germans had achieved a quick victory?
Losses to assets abroad caused by the war probably compensated largely the proportion of the
loss in French assets affecting foreign owners.
The model :
World War One cost for Stocks =
(Hypothetical value in 1928 of 1914 stock market capitalisation)
- (Observed value in 1928 of 1914 stock market capitalisation)
stocks
Cost
= MC1914 * (1 + R hyp)
s
s
(15 )
[
− MC
s
1914
* (1 + Robs ) (15)
s
φ
]
s

(1 + Robs ) (15) 
s
(15 )

= MC1914 * (1 + R hyp) −
φ


s
R
with
MC
s
hyp
s
1914
= ∆p
s
hyp
+ α rdt hyp
s
: Market capitalisation for stocks in 1914
obs : for observed, and hyp : for hypothetical (approximate by observed from 1854 to 1913)
∆p
s
hyp
rdt
: Geometric mean of price variations, for stocks (hypothetical)
s
hyp
: Yield (dividends/price), for stocks (hypothetical)
φ : number to switch 1928 francs into 1914 francs
1 franc 1928 = 0,1829267 francs 1914 (INSEE), thus φ = 5,4666703
19
Results for stocks :
Stock market capitalisation of the Cac 40 in January 1914 was 12 000 103 600 francs,
and total capitalisation of the most liquid French stocks was 17 882 613 250 francs. This
figure is that of French savings invested in stocks in 1914.
What really happened between 1914 and 1928: In January 1914, the Cac 40 was at 189
points and it rose to 453 in December 1928. Stock holders benefit, over this 15 years period,
from an annual geometric mean of 6,16 %. 1914 stock market capitalisation, about 18
billions, has a value of 43 billion in December 1928. Yet if expressed in the franc of 1914, it
is no more than 8 billion. Therefore, the loss is about half the 1914 value. The complete stock
market capitalisation observed at the end of 1928 is 61 billion of 1928 francs. Thus, 33 %
((61-43)/61) of this capitalisation comes from new investment. In other words, a new wave of
savings which has occurred since the war allows a gradual reconstitution of the stock market
capitalisation.
What could have happened if 19th century returns were maintained: From 1854 to
1913, the Cac 40 increases by 0,82 % each year (geometric mean). The 18 billion of stock
capitalisation in 1914 which hypothetically could have increased at this rate during the 15
years between 1914 and 1928 would have been therefore 20 billion in francs of 1914 while
only being 8 billion in reality. The loss is thus about 12 billion franc-or. These figures
obviously increase when assuming the hypothesis of a partial dividend reinvestment (α>0).
Assuming a complete dividend reinvestment (α=1), the loss is about 27 billion franc-or. To
help understand the scale of these figures, these amounts are compared to the French GDP in
1914 (54 billion). Therefore, the cost of war for French stock holders, in shortfall, is between
23 and 51 % of 1914 GDP.
Value in 1928 of the 1914 market capitalization
for stocks
in francs 1914
α=0
Hypothetical (I)
Observed (II)
Cost WWI (I-II)
α=1
20 691 656 884
41 403 000 882
8 019 414 351
13 794 281 925
12 672 242 533
27 608 718 957
20
Generalisation for all French savings :
A major part of French wealth was not however invested in stocks. An identical
argument can also be made with the investments in public and private bonds in order to find
the total loss. In 1914, the market capitalisation of public bonds is 25,52 billion francs21.
According to Michalet22, the value of the investment in French private bonds is 30 billion.
Thus, the model becomes:
Cost total = Cost stocks + Cost bonds + Cost statebonds
Cost
total
s
bo


(1 + Robs ) (15) 
(1 + Robs ) (15) 
bo
s
bo (15 )
(15 )
 + MC1914 * (1 + R hyp) −

= MC1914 * (1 + R hyp) −
φ
φ




s
rbo

(1 + Robs ) (15) 
rbo (15 )

+ MC1914 * (1 + Rhyp) −
φ


rbo
with
s : for stocks
bo : for private bonds
rbo : for public bonds (Rente bonds)
Returns of stocks and public bonds are known (Le Bris & Hautcoeur, 2008).
Nevertheless, the financial returns of privates bonds are not available. An average
performance between that in stocks and that in public bonds is used as an estimate for private
bonds. The performance of private bonds (observed and hypothetical) are calculating using :
R
bo
= ∆p + α rdt
bo
bo
s
rbo


+
∆
p
∆
p
bo

 and
with ∆p =
2
(rdt + rdt )
=
s
rdt
bo
rbo
2
All these returns used are on this table
21
22
INSEE, Annuaire Statistique de la France, 1951 p. 296
Michalet C-A, Les placements des épargnants français de 1815 à nos jours, Paris, PUF, 1968 p. 158
21
Hypothetical performances (proxy observed performances january 1854 - december 1913)
Cac 40
Rdt
?p
Dividende
Annual
Yield
AVERAGE
GEOMETRIC MEAN
4,52%
STANDARD DEVIATION
0,95%
State Bonds
Rdt
?p
Total
Coupon
Annual
Yield
Private Bonds
Inflation
Rdt
?p
Total
Coupon
Annual
Total
Yield
Price
Nominal
rate
Price
Nominal
Price
Nominal
Variation
Return
Variation
Return
Variation
Return
if α=0
if α=1
if α=0
if α=1
if α=0
if α=1
1,24%
0,98%
5,76%
5,50%
3,78%
7,25%
7,37%
0,72%
0,96%
0,77%
4,75%
4,55%
6,12%
6,23%
INSEE
0,22%
4,15%
0,87%
5,03%
3,78%
Observed Performances january 1914 - december 1928
Cac 40
State Bonds
?p
Rdt
Dividende
Annual
Total
Rdt
Coupon
Yield
Yield
AVERAGE
GEOMETRIC MEAN
3,90%
STANDARD DEVIATION
0,52%
Private Bonds
?p
Annual
Total
Rdt
Coupon
Yield
Inflation
?p
rate
Annual
Total
Price
Nominal
Price
Nominal
Price
Nominal
Variation
Return
Variation
Return
Variation
Return
if α=0
if α=1
if α=0
if α=1
if α=0
if α=1
7,53%
6,16%
11,43%
10,07%
5,04%
17,20%
17,47%
0,79%
-0,82%
-1,32%
4,33%
3,82%
10,20%
10,61%
INSEE
9,07%
4,47%
2,42%
6,94%
11,81%
Figure 14, Hypothetical and real performances of Cac 40, State Bonds and private bonds
(Sources: INSEE, author)
Therefore, using the same method as for stocks, the observed and hypothetical values
of private and public bonds are mentioned below:
Value in 1928 of the 1914 market capitalization
for private bonds
in francs 1914
α=0
Hypothetical (I)
Observed (II)
Cost WWI (I-II)
α=1
34 177 538 275
62 613 044 789
7 858 134 463
15 021 118 782
26 319 403 812
47 591 926 007
Value in 1928 of the 1914 market capitalization
for state bonds
in francs 1914
α=0
Hypothetical (I)
Observed (II)
Cost WWI (I-II)
α=1
28 621 002 448
49 763 643 121
3 826 612 495
8 187 268 474
24 794 389 953
41 576 374 647
After calculating the total sum of losses for stock holders and private and public bond
holders, war appears very expensive since the minimum cost (revenues consumed, thus α=0)
is about 64 billion of francs 1914, or 118 % of 1914 GDP. If the contradictory hypothesis is
22
assumed, that is to say a complete and permanent reinvestment (α=1), the loss is more than
117 billion of franc-or. In order to comprehend the significance of this amount note that in
January 1929, the first industrial stock capitalisation is Saint Gobain (third after the Canal de
Suez and the Banque de France) valued at 3,35 billion of francs 1929, or 613 million of francs
1914. Therefore, the total cost of war (with α=1) is equal to 190 Saint Gobain. This money, if
it had been made available for the French and the French economy from 1914 would have
probably facilitated higher gains in productivity thanks to the extra investments made
possible.
1928 OBSERVED AND HYPOTHETICAL VALUE OF 1914 MARKET CAPITALISATION (billion Franc 1914)
1914 VALUE
1928 VALUE
COST OF WWI
OBSERVED
HYPOTHETICAL
HYPOTHETICAL - OBSERVED
Stocks
Bonds
18
30
α=0
8
8
α=1
14
15
α=0
21
34
α=1
41
63
α=0
13
α=1
28
26
48
State Bonds
26
4
8
29
50
25
42
TOTAL
73
20
37
83
154
% GDP 1914
64
117
118%
216%
What about an indemnity to Germany?
This counterfactual study suffers from one weakness: assuming a German victory, an
indemnity would probably have been imposed upon the French government. Thus two more
hypothesis are used:
-
H4, the end of the war needs an indemnity equal to 30 % of GDP
The military defeat during the preceding century finished with an indemnity payed by the
defeated party. In the case of a German victory, their interest would have held in imposing the
maximum amount of compensation. A high financial compensation would have been the most
probable, accompanied by the transfer of authority of overseas territories.23 In 1815-1819,
1,65 to 1,95 billion francs were paid to the Allies, that is, 18 to 21 % of the GDP and the 5
billion paid in 1871 represent 25 % of the GDP (Occhino, Oosterlinck, White, 2007).
Therefore, the figure of 30 % of the GDP (16,25 billion francs) can be retained.
23
According to Jacques Marseille, Empire colonial et capitalisme français : histoire d’un divorce, this transfer
would probably have been more a saving than a cost.
23
-
H5, the indemnity payment reduces the public bonds capitalisation
This hypothesis implies that it would have been the “Rente” holders who paid the
indemnity. Assuming a French defeat and a large issue of state bonds like in 1871, the interest
rate demanded by the money lenders would have increased. This increase would have reduced
the value of the existing bonds. Finally, total market capitalisation of state bonds would have
increased but this rise would have come from additional savings (possibly foreign). In March
1870, when French state debt was 12,45 billion francs24, the price of the “Rente 3 %” which
serves as reference was 72,29 (or an actuarial rate of 4,19 %)25. One year later, in March
1871, after the specification of a 5 billion indemnity financed by new Rente (Emprunts
Thiers), the same “Rente 3 %” quoted 51,05 (or a rate of 6,01 %). This was a fall of 29,89 %.
Bonds owners lost close to 30 % of their savings. This 30 % represents a market capitalisation
of 4,32 billion which is very close to the famous “5 milliards”.
Model with indemnity to Germany :
Cost
total
s
bo


(1 + Robs ) (15) 
(1 + Robs ) (15) 
bo
s
bo (15 )
(15 )
 + MC1914 * (1 + R hyp) −

= MC1914 * (1 + R hyp) −
φ
φ




s
[
+ ( MC1914 − Ind ) * (1 + R )
rbo
rbo (15 )
hyp
]
rbo

(1 + Robs ) (15) 
rbo

−  MC1914 *
φ


with Ind : for indemnity
The cost of WWI for stocks and private bonds holders does not change. Nevertheless,
the loss on public bonds is reduced since the 1928 value of these investments is weaker thanks
to the deduction of the indemnity from the 1914 market capitalisation. Therefore, the total
loss is weaker. An average between α=0 and α=1 indicates a lost opportunity of having 65
billion of francs 1914, or « 107 Saint-Gobain ».
24
INSEE, Annuaire Statistique de la France, 1951 p. 296
Vaslin J-M, « Le siècle d’or de la rente française » in Gallais-Hamono G. (dir), Le marché financier français
au XIXème siècle, Publications de la Sorbonne, Paris, 2006
25
24
1928 OBSERVED AND HYPOTHETICAL VALUE OF 1914 MARKET CAPITALISATION
AFTER AN INDEMNITY (billion Franc 1914)
1914 VALUE
1928 VALUE
COST OF WWI
OBSERVED
HYPOTHETICAL
α=0
α=1
α=0
α=1
HYPOTHETICAL - OBSERVED
α=0
α=1
Stocks
18
8
14
21
41
13
28
Bonds
30
8
15
34
63
26
48
State Bonds
26
4
8
10
18
7
10
TOTAL
73
20
37
65
122
% GDP 1914
46
85
84%
157%
Maximum acceptable amount of the hypothetical indemnity:
Indemnity fixed at 30 % of the GDP is arbitrary. The model allows this constraint to
be minimised and even permits the calculation of what would have been the maximum
preferable amount to pay Germany in order to avoid the real returns of 1914-1928. This
estimation is obviously weak since the hypothesis of stability of state bonds returns is not
credible: a massive issue of Rente would have probably raised the interest rate to a higher
level. In this model an indemnity higher than the market capitalisation of the existing public
bonds implies a negative valorisation of
(CB
rbo
1913
)
− Ind . This means that 1914 French savers,
instead of having savings invested in bonds, are collectively in debt following the issue of
new public bonds. The value of this hypothetical debt is obtained using the same method than
the value of a saving. In this particular case, the hypothesis of a complete reinvestment of
revenue α=1, implies the issue of new bonds just to pay interest on the initial amount of debt.
This extreme figure is not totally impossible in the real world. In the case of debt too, a value
of α between 0 and 1 rather than either of two extremities is the more realistic scenario. The
1928 value of this debt rises as the part of the indemnity to borrow increases. The 1928 value
of this debt reduces the advantage of maintained 19th century returns on stocks and private
bonds. An indemnity is preferable until the total loss equal zero. The maximum preferable
indemnity is given by the equation below which comes from a rearrangement of that from the
model with indemnity:
0 = Cost stocks + Cost privatebonds + Cost publicbonds
0 = Cost
stocks
+ Cost
privatebonds
+
(MC
rbo
1914
)
(1 + Robs ) (15)
rbo
− Ind * (1 + R )
rbo (15 )
hyp
− MC1914 *
rbo
φ
25
Therefore:
Cost
stocks
+ Coût
Ind =
privatebonds
rbo

(1 + Robs ) (15) 
rbo (15 )
+ MC1914 * (1 + Rhyp ) −

φ


rbo
(1 + Rhyp ) (15)
rbo
Up to an indemnity of 56 billion francs 1914, it was financially preferable for French
savers to lose the war quickly. Amounts obtained with or without reinvestment of revenues
are close. 56 billion represents near to 100 % of GDP of 1914 and more than all securities
quoted in France in 1914 (73,4 billion). With the obvious small probability of such an
indemnity, this model shows therefore that it would have been distinctly preferable that
Germany won the war.
Maximum amount of preferable indemnity
in francs 1914
% GDP
α=0
56 875 004 615
105%
α=1
59 886 080 552
111%
26
REFERENCES
- Bordo M., Hautcoeur P-C, "Why didn't France follow the British Stabilization after World War One
?", European Review for Economic History, 2007.1
- Brown W., Burdekin R., "German Debt Traded in London During World War II: A British
Perspective on Hitler," Claremont Colleges Working Papers, 1999-19, Claremont Colleges
- Cowles Alfred and Associates, Common-stocks Indexes, Bloomington, Principia Press, 1939
- Destrem P. et D., La bourse sous l’occupation, Paris, Mare & Martin , 2005
- Dimson E., Marsh P. et Staunton M., Triumph of the Optimists: 101 Years of Global Investment
Returns, Princeton University Press, 2002
- Eichengreen B., Elusive stability, Cambridge, Cambridge University Press, 1990
- Fernandez V., “The war on terror and its impact on the long-term volatility of financial markets”,
International Review of Financial Analysis, vol. 17.1, 2008
- Frey B. et Kucher M., “Wars and markets. How bonds value reflect the Second World War ?”,
Economica, 68, 2001
- Frey B. et Kucher M., « History as reflected in Capital markets : the case of World War II », The
Journal of economic history, 1, 2000
- Goetzmann, William, Roger G. Ibbotson and Liang Peng,, “A New Historical Database for the
NYSE 1815 to 1925: Performance and Predictability”, The Journal of Financial Markets 4(1), 2000
- Hautcoeur, P.-C. & S. Grottard, “Taxation of corporate profits, inflation and income distribution in
France, 1914-1926”, WP, DELTA, 2001
- Hautcoeur P-C, "Was the Great War a Watershed ? The Economics of World War One in France", in
Broadberry S. & Harrisson M. (eds), The Economics of World War One, Cambridge, CUP, 2005
- INSEE, Annuaire Statistique de la France, 1951 p. 296
- Le Bris D. & Hautcoeur P-C., “A challenge to triumphant optimists? A new index for the Paris stock
exchange (1854-2007)”, Working Paper, Paris School of Economics, n. 2008-21, 2008
- Marseille J., « La Bourse sous l’occupation » in Dard O. et al (dir.) L’occupation, l’Etat français et
les entreprises, ADHE, Paris, 2000,
- Marseille J., Empire colonial et capitalisme français : histoire d’un divorce, Paris, Albin Michel,
2005
- Michalet C-A, Les placements des épargnants français de 1815 à nos jours, Paris, PUF, 1968
- Occhino F., Oosterlinck K., White E., “How occupied France financed its own exploitation during
WW2”, The American Economic Review, 97(2), 2007
- Oosterlinck K., "Market microstructure and Nazi influence on the Paris stock exchange during
WWII", WP-CEB: N° 04/26, 2004
- Oosterlinck K., "The bond market and the legitimacy of Vichy France", Explorations in Economic
History, 40, 2003
- Pollard A.A., War, Economy and Society 1939-1945, London, 1977
- Rigobon R., Sack B., "The effects of war risk on US financial markets," Journal of Banking &
Finance, vol. 29(7), 2005
- Sauvy A., Histoire économique de la France entre les deux guerres, Paris, Economica, 1984
- Sédillot R., Le franc enchaîné, histoire de la monnaie française pendant la guerre et l’Occupation,
Paris, Sirey, 1945, p. 139
- Sharpe W., Alexander G., Bailey J., Investments, New Jersey, Prentice Hall, 1999
- Siegel J., Stock for the long run : a guide to selecting markets for long-term growth, New York,
Irvin, 1994
- Stocks, Bonds, Bills and Inflation 1997 Yearbook, Chicago, Ibbotson Associates, 1997
- Vaslin J-M, « Le siècle d’or de la rente française » in Gallais-Hamono G. (ed.), Le marché financier
français au XIXème siècle, Publications de la Sorbonne, Paris, 2006
- Villa P., Séries macro-économiques historiques, INSEE Méthodes, N°62-63, Paris, 1997,
(or http://www.cepii.fr/francgraph/bdd/villa/mode.htm).
27