THE ECONOMICS OF WORLD WAR I: AN OVERVIEW

Stephen Broadberry
Department of Economics, University of Warwick, Coventry CV4 7AL, United
Kingdom
S.N.Broadberry@warwick.ac.uk

and

Mark Harrison
Department of Economics, University of Warwick, Coventry CV4 7AL, United
Kingdom
mark.harrison@warwick.ac.uk

31 March 2004

File: WWIintroduction4

Draft chapter for Stephen Broadberry and Mark Harrison, eds, The Economics of
World War I (in preparation).
PREFACE
During the twentieth century the world experienced two deadly global wars followed
by a “cold war” of unparalleled expense and danger. World War I opened this brutal
epoch. To many who took part the experience was little less than apocalyptic; it
seemed like an end, not a beginning. They saw it as putting a stop to history, progress,
and civilisation. They called it the “Great War”. They did not know that it would be
followed twenty years later by World War II and that the second war would be greater
and more dreadful than the first.

This book brings together eight country studies of the economics of World War I:
four Allies, three Central Powers, and a neutral country. Our book is the first, we
believe, to offer such a systematic comparison of economies at war between 1914 and
1918, and it is certainly the first to include the Ottoman Empire in such a collection.
These investigations suggest two themes that link economics with the study of war.

One theme is the contribution of economic factors to the outcome of the war. Our
book suggests that the outcome of global war was primarily a matter of the levels of
economic development of each side and the scale of resources that it wielded; in this
respect our conclusion is similar to that of our previous study of World War II
(Harrison, 1998). How well the resources were organised mattered greatly, but rich
countries could usually organise themselves more efficiently than poor ones. The
human factor mattered too: how well the people were motivated. Generally we find
that, given superior resources, the richer countries could solve the motivation
problems that defeated the poorer ones. Thus, organisation and motivation tended to
be endogenous; to this extent they did not independently influence the outcome.

Another theme of our book concerns the effects of war on long-run economic
development. It is sometimes claimed that war, however dreadful, may have positive
“spin-offs” for the nations that take part, whether they win or lose. In practice these
are not easy to find. War is, in general, a negative-sum activity. If war was followed
by recovery and accelerated development, this was usually no more than a making
good of wartime delays and losses. If wartime activity had promoted new forms of
technology or economic organisation that turned out to have peacetime applications
too, then there would always have been some cheaper way of achieving the same
result. A spin-off of World War I is that it destroyed several monarchies and imperial
elites: the ancien regimes of Germany, Austria-Hungary, Italy, Russia, and the
Ottoman Empire. This sounds as if it might have been for the best, but the destruction
of states associated with the war led to the displacement of populations on a colossal
scale (Mazower, 1998; Gatrell, 1999). Moreover, the war was followed in Russia by
civil war, communism, and dictatorship, in Italy, Austria, and Hungary by fascism
and dictatorship, and in Germany by fascism, dictatorship, war, and genocide.

The main lesson that has emerged from our study of the world wars of the
twentieth century is that peace is better than war. The best that can be said for World
War II is that a positive spin-off was a common understanding of this lesson. Because
of this, the main participants in World War II cooperated after the war to promote
recovery and trade. As a result, global economic growth in the half century after
World War II was much faster than in the half century before it. In contrast, only
some of the participants in World War I came away with this understanding. Others
 2

believed that the lesson of the war was to wage war again, only better. Hence World
War II.

Acknowledgements
The editors thank all the participants in the July 2002 Summer Research Euro-
Workshop in Economics held at the University of Warwick, including several co-
authors of this volume, and CORDIS, the Community Research and Development
Information Service of the European Union, and the Department of Economics at the
University of Warwick for financial support of the workshop. We also thank Jari
Eloranta for comments, advice, and assistance with data.
INTRODUCTION
Globalisation has been under way for 200 years. The first century of globalisation, the
nineteenth of our era, gave rise to world trade, a world capital market, worldwide
migration, great powers that competed for colonial empires on a worldwide scale, and
the first world war. The Great War of 1914 to 1918 interrupted and, for a time, set
into reverse the process of globalisation.

How did globalisation lead to war? At first sight it was the competition for
colonies that ran out of control. Britain and France, the established powers of “old”
Europe, had established a condominium over most of Africa and much of Asia;
Germany, the rising power of “new” Europe, had no colonies to speak of, wanted
some, and expected to get them only at the expense of the French and British. Behind
this lay a perception that world power was a zero-sum game. Since Adam Smith, the
Anglo-Saxon liberals had argued that trade was a game from which all could benefit
at once. But in the late nineteenth century liberalism was being challenged by a new
nationalism that gave more weight to the control of territory and settlement than to
trade and competition. When it came to territory, the supply was fixed and there was
only so much to go round. Therefore, the new nationalists reasoned, it was worth
Germany’s while to break up world trade for a while in order to grab territory from
the older powers.

In fact, the European powers did not fight World War I over colonies. The war
took the form not of a naval struggle to control access to the peripheral territories of
Asia or Africa, but of a struggle on land that was fought in the heart of the continental
homeland. At first, it is true, Germany’s desire for colonies stimulated a naval arms
race, but the battle cruisers that were laid down on each side in the process played
only a minor role in the war. More importantly, the quest for a German empire also
provoked an anti-German coalition, the Entente Cordiale between Britain and France
(1904) to which Russia was also admitted in 1907. Germany was not without friends,
having been allied with Russia since 1872, Austria-Hungary since 1879, and Italy
since 1882, but Russia and Germany had drifted apart (and Italy would renege on its
treaty in 1914). The result of rising polarisation of the continental powers was to shift
attention away from Germany’s original aim of an adjustment of the boundaries of the
British and French empires overseas towards the issue of the balance of power in
Europe itself. As a result, the war was largely fought on European soil for power in
Europe.

The events that led to the war in 1914 were the assassination of the crown-prince
of Austria-Hungary by a Serbian nationalist in Sarajevo on 28 June, an Austro-
Hungarian ultimatum that Serbia rejected, and Russia’s mobilisation in defence of
Serbia which, in its turn, triggered a German attack on France and Belgium; this was
followed by British entry into the war on the side of France. The German attack on
France was motivated by a forward-looking calculation: once the coalitions on each
side were fully engaged Germany risked a war on two fronts, against Russia in the
east and France in the west. Having identified the Russians as the less mobile enemy,
the German plan was designed to avoid a war on two fronts at once by attacking
France with a knock-out blow at the first sign of Russian mobilisation; thus, while the
Russians completed their mobilisation German army would have time to defeat the
 2

French before turning their victorious armies to the east to defeat the Russians in their
turn. Of course, this is not how things turned out.1

This book deals with two issues that then arise. First, what did economic factors
contribute to victory and defeat in World War I? Second, how did the war affect
postwar economic institutions and performance in the economies that took part or
were most affected by the war?

As far as the first question is concerned, it is worth recalling that the German war
plan for 1914 anticipated victory in the west within six weeks. The war was intended
to be won by military, not economic means, and was to be finished off long before
economic factors could be brought into play. It was only after this plan had failed, as
the leaders on each side contemplated the ensuing stalemate, that belts began to be
tightened and sleeves rolled up for the mobilisation of entire economies (Chickering
and Forster, 2000).

Once plans began to be drawn up for a longer haul, a war of attrition developed in
the west where the opposing forces of Germany, France, and Britain, each backed by
large, rich, and successful economies, ground each other down with rising force levels
and rising losses. In battles that were intended to be won by the last man left standing,
resources counted for almost everything. The greater Allied capacity for taking risks,
absorbing the cost of mistakes, replacing losses, and accumulating overwhelming
quantitative superiority eventually turned the balance against Germany.

Eastern Europe, the Balkans, and the Near East formed the theatre of combat for
the economically weaker powers: Russia, Italy, Austria-Hungary, and Turkey. The
British and Germans wished to be more involved but neither could withdraw
significant forces from the western front. In the east, therefore, the immediate
outcomes of battles were less determined by economic factors, at least in the short
run. Over a period of years, however, the battles drained the weakest economy first,
and this led to Russia’s exit from the war in 1917. Then, the Central Powers’ chance
for victory in the east was destroyed by Germany’s defeat in the west. Ultimately,
economics determined the outcome.

I. POPULATION, TERRITORY AND GDP

The economic advantage of the Allies over the Central Powers was substantial at the
outbreak of war and rose steadily as the composition of the belligerents changed on
each side. The most striking change was that during 1917 Russia was defeated and
abandoned the Allies, but was replaced by the United States. Thus the richest great
power stepped into the gap left by the poorest, and this led to a further increase in the
Allied advantage.2

1
We do not tell the story of the war in this book. Those who would like a more
narrative account should note a three-volume history in preparation by Hew Strachan
of which the first volume (Strachan, 2003) is newly published. Herwig (1997) gives a
compact account of the war from the perspective of the Central Powers.
2
Technically speaking, the United States of America never joined Britain and
France in a formal Alliance; therefore, the United States was not strictly an “Ally”.
 3

1. Size and development

What were the resources that were deployed on either side in the war? These are best
measured by adding up the populations, territories, and gross domestic products of the
territories at war. Populations limited the numbers of men and women available in
each country for military service or war work. Territories limited the breadth and
variety of natural resources available for agriculture and mining; the wider the
territory, the more varied were the soil types and the minerals beneath the soil. GDPs
limited the volume of weapons, machinery, fuel, and rations that could be made
available to arm and feed the soldiers and sailors on the fighting front. The larger the
population, territory, and GDP of a country, the easier it would be for that country to
overwhelm the armed forces of an adversary.

In adding up the resources available to each country we also compute the

territories and income available per head of the population. Most important was
average GDP per head, which reflected the country’s development level. A poor
country might have a large population, but if most of the adults were engaged in low-
productivity subsistence farming then there would be little real possibility of
transferring many of them out of agriculture to the armed forces or war industry since
the remaining farmers would be unable to produce enough food to keep everyone
alive. Equally, a poor country might have a large territory but, without a high level of
development of roads and railways, would be unable to exploit it economically or
defend it militarily. Finally, a poor country typically lacked efficient government and
financial services of the kind necessary to account for resources and direct them into
national priorities. Thus, a relatively high level of economic development was
essential if territory and population were to count in war.

Table 1 adds up the resources on the Allied side at the outbreak of war and shows
how the volume of resources changed; in this table and the next, countries are listed as
far as possible in order of their entry into the war. In reality, of course, populations
and outputs changed year by year. To assist with comparability the 1913 figures for
each territory are the ones reported in the table. In the first phase of the war Russia,
France, and the United Kingdom were joined together as the power of the Triple
Entente. They brought with them their dependencies and colonies. Other countries
joined in too: Serbia and the other Yugoslav states, the British Dominions, Liberia,
and Japan with her colonies. During 1915/16 a second wave of countries joined the
Allies: Italy, Portugal, and Roumania. In the third wave of 1917/18 Russia dropped
out but the United States joined in, bringing its own possessions, most of Central
America and Brazil. Greece, Siam, and China also joined. By the end of this process
governments representing 70 per cent of the world’s prewar population and 64 per
cent of its prewar output had declared war on the Allied side.

Insert Table 1.

The bare totals on the Allied side do not give any idea of their heterogeneity. The
British empire will do for illustration since it comprised some of the richest and

This had minor consequences for the co-ordination of military strategy in the west,
and major consequences for postwar diplomacy and the negotiation of a peace treaty
with defeated Germany. For the present chapter it is not an important distinction.
 4

poorest regions in the world. Britain itself had a prewar population of 46 million with
an average income per head of nearly $5,000 (at 1990 prices). Its colonies, excluding
the Dominions, had a prewar population of 380 millions, mostly Indians, with an
average income of less than $700. Thus a colonial population eight times that of
Britain produced a similar volume of output. Moreover this output was far less
available than Britain’s for fighting Germany for three reasons: it was hundreds or
thousands of miles away from the theatre of war, the level of development of colonial
government administration and financial services rendered it hard to track and control,
and most of it was already committed to the subsistence needs of the colonial
populations. In short, the mere possession of low income territories was of little value
to a great power in the war. If India helped Britain in the war it was to enable British
trade and commerce rather than because Britain could mobilise Indian resources in
any meaningful sense. And the trade that really mattered to the British economy in the
war was with rich America and Canada, not with poor India.

Insert Table 2.

Table 2 adds up the resources of the Central Powers. This is a much shorter story
with a smaller bottom line. Austria-Hungary began the war, joined immediately by
Germany and soon by the Ottoman Empire. In 1915 the Central Powers were joined
not by Italy, which reneged on its prewar treaty obligations, but at least by Bulgaria.
At its maximum extent the alliance of the Central Powers comprised little more than
150 million people, but their relative lack of success in accumulating low-income
colonies made them relatively well off with an average income per head of $2,450,
comparable to that of Italy on the Allied side.

Insert Table 3.

2. Allied superiority
Table 3 allows us to compare the resources on each side at three benchmark dates:
November 1914, 1916, and 1918. This table offers comparisons for each alliance as a
whole, and also counting great powers only. The rationale for the latter is very simple:
if low-income colonies did not count much, how do the figures look if we do not
count them at all? There is some imprecision here, of course. For example Russia is
included, as a great power, but much of its territory was little more developed than
that of India, which is excluded; also excluded are the British Dominions, which were
much richer than Russia. Still, counting great powers only has the merit of simplicity.

The table shows something very striking: in terms of the resources on either side
the Central Powers do not seem to have had much hope. If Germany could not win the
war for the Central Powers in the first six weeks, using surprise in the west and an
army with superior military qualities, then the chances of victory could only diminish
over a longer span of time in which economies would be mobilised on each side and
the balance of resources would count for more and more.

Even in the first stage of the war the Allies had access to five times the
population, eleven times the territory, and three times the output of the Central
powers. This access was limited by relatively low average incomes across the colonial
empires of Britain and France, and low incomes in Russia; we see that the average
level of GDP per head on the Allied side in 1914 was not much more than half that of
the Central Powers. If we consider great powers only then the Allied advantages in
 5

population and output shrink to twice; the Allied advantage in territory actually
increases, reflecting the German and Turkish propensities to colonise sandy deserts in
Africa and the Middle East.

As the war continued, the Allied powers’ advantage in output grew. The decisive
year was 1917. When America displaced Russia the Allied population and territory
declined but its output multiplied; the average development level of the Allied powers
rose above that of the Central Powers for the first time. Although it would take time
for America’s presence to be felt on the battlefield, it sealed the Central Powers’ fate.

The force of these changes is felt even more strongly when it is remembered that
the figures in table 3 are based on the assumption that in wartime the real output of a
given territory did not change. While we cannot track the changes for all countries,
the figures available suggest further substantial swings which worked primarily to
favour Britain and America. Figure 1, based on Table 4, shows that in wartime the
British and American economies expanded by over 10 per cent. Italy marked time.
Russia began to collapse in 1916 and France in 1917, and this emphasises still more
forcefully the extent to which the Allies were saved by the American entry into the
war. On the side of the Central Powers, however, the dismal failure of wartime
mobilisation was evident from the outset: for much of the war period the German and
Austrian economies flatlined at 20 to 25 per cent below their prewar benchmarks for
real output.

Figure 1. The Wartime Change in GDP: Seven Countries

120
Real GDP, per cent of 1913

110
UK

100 USA
Italy
90 Germany
Austria
80 Russia
France
70

60
1913 1914 1915 1916 1917 1918

Source: Table 4.

Insert Table 4.

3. The human factor

Where, in all this, is there room for factors other than the economic ones? Reviewing
our previous work on World War II (Harrison, 1998) the historian Richard Overy
(1998) objected that we left no role to “a whole series of contingent factors – moral,
political, technical, and organisational – [that] worked to a greater or lesser degree on
national war efforts.” Such factors were clearly significant in World War I, and
 6

economists have considered why they must matter in principle (Brennan and Tullock,
1982) yet we do not apologise for giving due weight to the quantities of resources.

At first the two sides were unequal in military and civilian organisation,
motivation, and morale. Germany entered the war with first-rate military advantages
associated with the evolution of a highly successful military organisation, past
victories, and the exploitation of initial shock and rapid movement. But the effects of
looming defeat electrified Britain and France, transformed public opinion, and forced
their armies and governments through intensive courses in the new rules of warfare
and mobilisation. This proved to be the pattern through the war: each temporary
setback was followed by strenuous efforts to refine strategy and strengthen morale
and organisation, and these efforts generally succeeded within the limits permitted by
the resources available to support them. In short, the “moral, political, technical, and
organisational” issues of the war on each side were not independently variable factors
but proved to be endogenous to the progress of the war. Other things being held equal,
a deficit of organisation or morale on one side tended to be overcome through a self-
balancing process. The one thing that could not be overcome was a deficit of
resources.

This approach is well illustrated by comparing the two offensives that appeared to
give Germany its best chances of winning the war: August 1914 and March 1918. In
the first of these Germany planned to exploit mass, movement, and surprise to destroy
the French Army before the British could intervene in the West and before the
Russians could mobilise in the East. In practice the German army succeeded in many
of its planned objectives but failed in the ones that were vital. The stalemate of the
trenches resulted. Had the German plan succeeded the economic factors on each side
would never have had time to be felt. Given that it did not, the richer Allies won time
to put right their military and organisational failings, but they could not have done so
without resources on their side.

Its spring offensive in 1918 again seemed to offer Germany the prospect of
winning the war on a purely military advantage. For the first time since 1914 its
soldiers opened up great gaps in the Allied lines and advanced dozens of kilometres
towards the Channel ports. The offensive badly shocked the Allies and forced them
into reorganisation; the Americans had to accept a unified command. Resources
defeated the advancing Germans: their own lack of supply, for they were badly
clothed and undernourished even before they began their advance; the abundance of
supplies they found in the Allied trenches that caused them to eat and drink their
advantages away; and the superabundance of war materials that enabled the Allies to
regroup and go on to inflict a far greater defeat on the exhausted enemy.

II. MOBILISATION AND THE LEVEL OF DEVELOPMENT

1. Military mobilisation
The evidence of the chapters that follow suggests that the comparative success of the
various economies in mobilising their resources depended on three factors that varied
independently: their level of economic development, their proximity to the front line,
and the duration of their engagement. A statistical warning may be in order: poorer
countries had less good government and national accounts, so we have less
confidence in their data and several are missing from our tables and figures. France
and Britain were rich and close to the action; Figure 2, based on Table 5, shows that
 7

in France, government outlays had taken up nearly half of national income at current
prices by 1917, and in Britain and Germany nearly 60 per cent.. Italy, in contrast,
could only manage less than one third. Australia, Canada were rich but distant, so that
the burden of government on their national incomes reached less than one sixth. The
USA, richer, distant, entering the war late, also gave 17 per cent of its GDP to its own
war effort at the peak of mobilisation and lent another 5 per cent to its Allies. Despite
their central involvement and the gambling of their essential state interests Austria-
Hungary, and Russia appear to have been relatively unsuccessful; in the case of
Austria-Hungary the proportion of national income that the government could
command for the war was no more than one third of national income but this
proportion proved to be unsustainable and had declined to one quarter by 1917/18
when the Habsburg empire was heading for defeat (see chapter 00). In the Ottoman
Empire the proportion of GDP under the control of the state was no more than 16-20
per cent at the peak (chapter 00).

Figure 2. The Share of Government Spending in National Income: Seven Countries

80%
Central Government Expenditure,

France
60% Germany
share of GDP

UK
Italy
40%
Australia
Austria-Hungary
20% Canada
USA

0%
1914 1915 1916 1917 1918

Source: Table 5, except Austria-Hungary (military expenditure only) from chapter 00.

Insert Table 5.

The richer countries maintained their advantage despite the fact that, in peacetime,
they tended to spend a lower proportion of their national income on defence (Eloranta,
2003). Thus, their ability to transfer resources rapidly from peacetime to wartime uses
was somewhat greater than even these figures imply.

Men and weapons provide more unambiguous measures of mobilisation than

money. In the mobilisation of young men we find a pattern that again rises with
development and falls with distance. Figure 3 plots the wartime mobilisation rates of
various countries against their prewar incomes per head in three distance bands. The
first band comprises the front-line Eurasian states on whose territory or borders the
war was fought. The second band is for the European countries separated from the
war by land or sea, with only two members: Britain and Portugal. The third band
includes countries that joined the war from continents beyond Europe and the Near
East. Cumulative numbers mobilised are shown as a proportion of young men in the
age group from 15 to 49 years of age. In each distance band, i.e. controlling for
 8

distance, the figures show a consistent positive dependence of the proportion

mobilised in each country on its prewar income level. However, dropping a band
lowered the proportion substantially.3

Figure 3. Military Mobilisation, Development Level, and Distance: Eighteen

Countries and the French Colonies
100% 60%
Mobilisation, per cent of males,

Mobilisation, per cent of males,

80%

40%
60%
15-49

15-49
40%
20%

20%

Front Line Eurasia European Periphery

0% 0%
0 1000 2000 3000 4000 5000 6000 0 1000 2000 3000 4000 5000 6000
GDP per head, $ and 1990 prices GDP per head, $ and 1990 prices

60%
Mobilisation, per cent of males,

40%
15-49

20%

Non-European States
0%
0 1000 2000 3000 4000 5000 6000
GDP per head, $ and 1990 prices

3
The power of this relationship may be confirmed by multiple regression. We
code the three distance bands 0 for front-line Eurasian states, 1 for the European
periphery, and 2 for non-European states. We measure the duration of each state’s
engagement in the war in years rounded to the nearest quarter. Then we regress the
mobilisation rate rate on GDP/head in dollars, distance, and duration. With 19
observations and t-statistics in parentheses we find:
Mobilisation = 0.0692 + 0.0767× 10−3 × GDP / head + 0.1007× Duration − 0.2743× Distance
( 6.02 ) ( 4.61) ( −11.60 )

In words, each additional thousand dollars of GDP per head raised the mobilisation
rate by more than 7 points; each additional year of engagement raised the mobilisation
rate by 10 points; dropping one distance band lowered the mobilisation rate by 27
points. All the slope coefficients are significant at the 0.01% level and the R-squared
has a value of 0.91. In words, the relationships are very significant; by far the greater
part of the variation in mobilisation is explained by them; hardly any room is left for
traditional historical accounts based on the peculiarities of national public and private
institutions and government policies.
 9

Sources: GDPs per head in 1913 from Tables 1 and 2 or, if not listed there, from
Maddison (2001: 185); cumulative mobilisation rates, 1914-1918, from Urlanis
(1971: 209).

Note: Observations, reading from left to right in order of increasing GDP per head are
as follows. Front line Eurasia: Turkey, Serbia, Russia, Bulgaria, Roumania, Greece,
Austria Hungary, Italy, France, and Germany. European periphery: Portugal and UK.
Non-European States: French colonies, India, South Africa, Canada, New Zealand,
USA, Australia.

The richer countries were not only able to mobilise more men. Regardless of
distance, they also supplied them better. Capital-abundant economies were able to
support capital-intensive warfare. Figure 4 plots cumulative war production of rifles,
machine guns, field guns, tanks, and aircraft in units per thousand men mobilised
through the war and per year of the war. In each case we see that supply rose strongly
with the development level of the country.
 10

Figure 4. War Production and Development Level: Six Countries

600 14
Rifles per thousand combatant-
12

Machine guns per thousand

500

10

combatant-years
400
8
years

300
6
200
4

100 2

0 0
0 1000 2000 3000 4000 5000 6000 0 1000 2000 3000 4000 5000 6000
GDP per head, $ and 1990 prices GDP per head, $ and 1990 prices

1.2 0.18
Guns per thousand combatant-years

Tanks per thousand combatant-

0.16
1.0
0.14
0.8 0.12

0.10

years
0.6
0.08
0.4 0.06

0.04
0.2
0.02
0.0 0.00
0 1000 2000 3000 4000 5000 6000 0 1000 2000 3000 4000 5000 6000
GDP per head, $ and 1990 prices GDP per head, $ and 1990 prices

2.5
Aircraft per thousand combatant-

2.0

1.5
years

1.0

0.5

0.0
0 1000 2000 3000 4000 5000 6000
GDP per head, $ and 1990 prices

Sources: GDPs per head in 1913 from Tables 1 and 2; cumulative war production,
1914-1918, from Adelman (1988), 45, except UK from chapter 00 and Austria-
Hungary from chapter 00; cumulative mobilisation as Figure 3. For each country
“combatant years” are numbers mobilised multiplied by years of engagement in the
war rounded to 1.5 years for the USA, 3.5 years for Russia, and 4.25 years for the
others.

Note: Observations, reading from left to right in order of increasing GDP per head are
Russia, Austria-Hungary, France, Germany, the United Kingdom, and the United
States.

2. Mobilisation and agriculture

Countries like Russia and Austria-Hungary were large; why did it make such a
difference that they were also poor? One of the most striking attributes of relative
poverty was the role of subsistence farming. Contemporary observers were aware of
 11

these differences and interpreted them as follows: when war broke out, a country such
as Russia would have an immediate advantage in the fact that most of its population
could feed itself; moreover, the ability to divert food supplies from export to the home
market would actually increase Russia’s advantage. In contrast Britain would quickly
starve (Gatrell and Harrison, 1993).

This diagnosis could not have been more wrong. In practice the presence of a
large peasantry proved to be a great disadvantage when it came to the mobilisation of
resources for war. Peasant agriculture behaved very much like a neutral trading
partner. Why should Netherlands trade with Germany given the latter’s reduced
ability to pay, except under threat of invasion and confiscation? Peasant farmers made
the same calculation. Thus the Russian economy looked large, but if the observers of
the time had first subtracted its peasant population and farming resources they would
have seen how small and weak Russia really was. Meyendorff (cited by Gatrell in his
chapter on Russia) described what happened in Russia as “the Russian peasant’s
secession from the economic fabric of the nation”. And not only from Russia, for
Italy, Austria-Hungary, the Ottoman Empire, and Germany all had large peasant
populations that proved extremely difficult to mobilise for much the same reason.

The common process of the peasant’s secession is clearly visible from a

comparison of the richer and poorer countries’ experience. When war broke out
British and American farmers boosted production because they were offered higher
prices and responded normally to incentives. The fact that British farming had already
contracted to a small part of the economy made its expansion easier: there were
plentiful reserves of land unused or little exploited, and the high productivity of farm
labour meant that substantial increases in farm output could be achieved with
relatively little extra in the way of resources,

In the poorer countries, in contrast, wartime mobilisation began by taking

resources away from farming, particularly young men and horses for the army. Once
in the army these young men and horses still needed to be fed, of course, which
implied a diversion of food supplies from rural households to government purchasers.
But at the same time the motivation for farmers in the countryside to sell food was
greatly reduced. These were subsistence farmers who grew food partly for their own
consumption; what they sold, they took to the market primarily to buy the
manufactured commodities, primarily textiles and metal goods, that they needed for
their families. But war dried up the supply of manufactures to the countryside. The
small industrial sectors of the poorer countries were soon wholly concentrated on
supplying the army with weapons and equipment, uniforms and rations. There was no
capacity left to supply the countryside, which faced a steep decline in supplies.
Consequently, peasant farmers retreated into subsistence activities. As the market
supply of food dried up, in the towns food prices soared.

The economy began literally to disintegrate: there might still be plenty of food,
but it was in the wrong place. The farmers preferred to eat it themselves than sell it for
a low return. The government had to feed the army at all costs for a simple reason:
hungry soldiers will not fight. Between the army and the peasantry the urban workers
were now caught in a double squeeze. Aware of the unequal distribution of food,
public opinion might blame unpatriotic speculators or incompetent officials, but the
truth was that a poor country had few real choices.
 12

The scope for policy to improve the situation was usually more apparent than real,
and government action typically made things worse: for example the Russian,
Austrian, and German governments all began to ration food to the urban population,
while attempting to buy up food from the countryside at purchasing prices that were
fixed low for budgetary reasons. To repeat: in richer countries the government paid
more to the food producers, and this worked, but in poorer countries we will see that
the government wanted to pay less and this had entirely predictable results. The
willingness of farmers to participate in the market was still further undermined.

This process may be illustrated in a couple of diagrams. Figure 5 shows the

prewar food market of two countries, one that we will style as “Russia” and the other
“Germany”; the difference between the two is that before 1914 Russia was a
substantial net exporter of food, Germany a net importer.

Figure 5. The Prewar Food Market: Russia and Germany

price demand price demand

supply

supply

world B A C D
price

RUSSIA quantity GERMANY quantity

Both countries were competitive producers and each faced the same world price;
therefore, both produced at the same marginal cost but, given the differences in their
national resource endowments and demands Russia produced at A, consumed at B,
and exported AB, much of it to Germany; Germany produced at C, consumed at D,
and imported CD, much of it from Russia.
 13

Figure 6. The Wartime Food Market: Russia and Germany

price demand price demand wartime supply

wartime supply K prewar

supply

prewar J
supply

F ration L M
price

ration G E H
price

RUSSIA quantity GERMANY quantity

Figure 6 shows the effects of war on the market equilibrium. When war broke out
the hostilities on land and sea broke up the world market and isolated each country.
Other things being equal, the loss of foreign markets should have reduced food prices
in Russia which could now have produced and consumed at E; Germany, deprived of
foreign supplies, should have produced and consumed with a higher price at J.

At the same time, however, the military mobilisation of young men, horses, and
nitrates raised farm costs. Nitrates proved to be a classic “dual use” commodity of
modern warfare. They were an essential ingredient in both farm fertilisers and high
explosives. Their chemical instability made them very hard to synthesise. Before
World War I the bulk supply of nitrates to Europe came from natural deposits
overseas. The trade disruption associated with the war forced the development of a
German industry to manufacture nitrates artificially, but these were costly and war
needs took up the supply that was created (Lee, 1975). As a result the availability of
nitrates for farming fell sharply in Germany, but the impact was less in Russia where
the initial reliance on nitrates was less widespread. The losses of human, animal, and
chemical power combined to push the supply curve sharply upwards in both countries
in the figure, moving the market equilibrium to a higher price and lower consumption
at F in Russia and K in Germany.

Finally, the government stepped in and tried to hold prices down, creating excess
demand and scope for a black market in each country. To the extent that such controls
were effective, output and consumption tended to fall further in both countries, to G in
Russia with unsatisfied demand equal to GH, and to L in Germany with unsatisfied
demand equal to LM.

To the extent that they failed there was scope for black marketeers to step in and
capture rents; as long as the rents were competed away production and consumption
could both tend back to F and K but popular respect for law and government would
inevitably suffer in the process.
 14

Finally, we see why the outcome was potentially just as bad for German
consumers as for Russians. The Russians did indeed have their prewar export surplus
to fall back on. Although a much richer nation than Russia, urban famine was as acute
in Germany in the closing stages of the war.

Some readers may be surprised to find Germany numbered among the countries
that suffered a decline in agricultural output during the war. Although pre-1914
Germany has entered the economic history textbooks as a developed economic power,
it should be noted that its modernisation was highly unbalanced. High levels of
productivity in heavy industry co-existed with much lower productivity in light
industry, and much of the service sector was also characterised by low productivity,
despite Gerschenkron’s (1962) focus on the modernised railways and the universal
banks (Broadberry, 1998). But perhaps the most obvious sign of Germany’s relative
backwardness was the high share of the labour force engaged in low productivity
agriculture. Germany paid a high price during the two world wars for protecting its
agriculture in peacetime (Olson, 1963).

In summary, to be poor when war broke out was to suffer the consequences of a
peasant agriculture, which was essentially a dead weight on the mobilisation efforts of
the country concerned. For this purpose we include Germany. The process that
resulted had its inexorable conclusion in urban famine, revolutionary insurrection, and
the downfall of emperors.

III. COSTS OF THE WAR

1. Bogart’s study of direct and indirect costs
At the end of World War I, a number of attempts were made to quantify the costs of
the war. As Milward (1984: 9-27) points out, this literature reflected a liberal tradition
that saw warfare as having entirely negative effects, and did not perceive any positive
changes initiated or accelerated by the war. It will be useful to review Bogart’s (1920)
study of the costs of World War I in some detail, since it was carried out as part of the
important series of publications on the Economic and Social History of the Great War
sponsored by the Carnegie Endowment for International Peace, and it has been the
starting point for all subsequent calculations.

Insert Table 6.

Table 6 provides estimates of what Bogart labels “direct costs” of the war.
These costs are calculated as the flow of spending by governments on the prosecution
of the war, i.e. spending over and above normal prewar levels. Inter-allied transfers
are subtracted from gross expenditures to arrive at net costs, which show the heaviest
burden to have been borne by Britain and Germany, with France, Russia and the
United States also bearing a substantial net cost on the Allied side and Austria-
Hungary amongst the Central Powers. On a per capita basis, Britain, France and
Germany stand out as bearing a much higher net cost than the other countries.
Nevertheless there are a number of disadvantages to the way that Bogart presents the
data. First, it is inappropriate simply to add up nominal sums spent at different times,
given the wartime inflation. Second, this problem, as well as the related problem of
the conversion to dollars of all values expressed in national currencies can be avoided
if the war expenditures are expressed as a proportion of national income in each year,
as in Table 5 above.
 15

Insert Table 7.

Table 7 introduces a number of what Bogart labels “indirect costs”, consisting

largely of losses to human and physical capital. The capitalised value of war deaths
shows the biggest losses to have been sustained by Russia and Germany, with other
substantial losses borne by Britain, France and Austria-Hungary. Property losses on
land were heaviest in France and Belgium, which is included here in Other Allies. The
heaviest shipping losses were sustained by Britain, the dominant nation in world
shipping before 1914.

A number of accounting procedures here give cause for concern. Although the
accounting for losses to physical capital is unremarkable (remembering that cargoes
can be seen as inventories), the treatment of human capital requires some attention.
The capitalised value of human life, based simply on lifetime earnings, would
overstate the social loss since people consume as well as produce. One way of
arriving at the social loss is therefore to subtract consumption from lifetime earnings,
as in the work of Clark (1931). Obviously this is not an attempt to capture the loss of
utility arising from war deaths, but merely treats people as human capital to be
replaced like physical capital so as to maintain production. As Edelstein (2000: 349)
points out “It is absurd to think the methods and perspectives of economic history can
come anywhere near to comprehending the meaning of human losses from war. We
are far better served by the speeches and letters of Lincoln or the poetry of Sassoon,
Brooke, Owen, Graves and Seager.” However, for symmetry with the treatment of
physical capital on a replacement cost basis, the simplest procedure is to add up the
cost of rearing and training a worker, since this is the net loss to society by premature
death.

Insert Table 8.

In Table 8, Bogart simply adds the direct and indirect costs to arrive at a grand
total. The justification for this is unclear, since it combines flows of current spending
with changes in the stock of assets needed to generate those flows. To add to the
confusion, lost production (a flow concept) is included as an indirect cost (a stock
concept). Note also that some of the government spending on the war effort, which is
included negatively as a direct cost by Bogart, should actually enter positively in the
national balance sheet, contributing to intangible physical and human capital. To the
extent that the war induced additional spending on health and welfare, this contributed
to the accumulation of intangible human capital, while research expenditure on the
development of weapons may have had spin-off effects on the accumulation of
intangible physical capital. Finally, note that Bogart (1920: 299) makes no attempt to
relate his estimates of the direct and indirect costs of World War I to levels of income
or wealth, but simply concludes that “the figures presented in this summary are both
incomprehensible and appalling”. This is an issue which can be addressed in the
national balance sheet approach.

2. Effects on national balance sheets

Broadberry and Howlett (1998) provide an accounting framework for evaluating the
long run impact of war on wealth, which is based on national balance sheets. The first
important distinction is between stocks and flows in the system of national accounts.
Issues concerned with the scale of mobilisation are best tackled by looking at flows of
 16

income, expenditure and output, and calculating the proportion of these flows that is
devoted to the war effort, as in Table 5. However, the long run impact of the war can
best be assessed by looking at the effects on national wealth, defined here to include
human as well as physical capital, intangible as well as tangible capital and net
overseas assets (Goldsmith et al., 1963; Revell, 1967; Kendrick, 1976).

Tangible physical capital is the conventional form of capital, consisting of

buildings, equipment and inventories. Intangible physical capital is cumulated
expenditure on research and development, which is seen as improving the quality of
the tangible physical capital. Tangible human capital is the spending required to
produce an uneducated, untrained worker, i.e. basic rearing costs. Intangible human
capital is mainly spending on education and training to improve the quality of the
human capital, although it also includes other items such as spending on health and
safety and mobility costs. In an open economy, the impact of the war on net overseas
assets must also be taken into account.

We believe that this accounting framework deals with the main objections
raised by writers such as Hardach (1977: 286) and Milward (1984: 9-27) to previous
attempts to quantify the impact of war on the economy. In particular, note that: (1) a
clear distinction between stock and flow concepts is maintained throughout (2) all
nominal values are converted to a constant price basis so that values for different
years can be added together (3) human capital calculations take account of the fact
that people consume as well as produce (4) the fact that postwar birth rates rise does
not alter the fact that the human capital embodied in those killed by warfare is lost;
this has a negative impact on national wealth as much as any destruction of physical
capital, which is usually followed by increased investment to make good war losses
(5) technological change stimulated by wartime research and development can be
seen as having a positive impact on intangible physical capital (6) social spending
stimulated by the war can be seen as having a positive impact on intangible human
capital.

3. War casualties and human capital losses

One obvious cost of the war was the huge number of deaths resulting from the
“industrialisation” of warfare, which led to the growing use of the term “total war”
(Chickering and Förster, 2000). There are conceptual difficulties with the types of
death to be included in any definition of war deaths, which could be restricted to
battle deaths of military personnel or broadened to include non-battle deaths of
civilians as well as military personnel. We have opted for battle and non-battle deaths
of military personnel, following Urlanis (1971) since this offers a high degree of
uniformity in data across countries while going beyond those killed in battle or who
died from wounds or poison gas. Non-battle deaths includes those who died from
disease, died in captivity or died from accidents and other causes. We exclude most
deaths in the influenza pandemic of 1918, however.

Insert Table 9.

The data in Table 9 show how military deaths were spread across the
combatant countries. Germany suffered the most casualties in absolute numbers,
although a number of countries sustained heavier losses as a percentage of the
population, including France, Serbia-Montenegro and Rumania amongst the Allies
 17

and Turkey amongst the Central Powers. Although Russia sustained the second
highest losses in absolute numbers, this was a lower proportion of the population than
the losses in Britain and Italy amongst the Allies and Austria-Hungary amongst the
Central Powers. Taking the Central Powers and the Allies together, the battle and
non-battle deaths of military personnel represented about 1% of the population of the
combatant nations.

Insert Table 8.

Turning these casualties into estimates of human capital losses in the national
balance sheet framework requires knowledge of the prewar costs of rearing and
educating a child, together with cohort-specific estimates of the education of the
labour force. In the absence of sufficient data for many countries, the human capital
losses in Table 10 are calculated as the ratio of war deaths to the prewar population of
prime working age, taken from Urlanis (1971). This differs from the proportion of
human capital destroyed by the war to the extent that younger cohorts had more
human capital investment, particularly through education. Also, since the human
capital losses are not calculated in monetary units, they cannot be added to physical
capital losses to provide an estimate of the proportion of physical and human capital
destroyed by the war.

Insert Table 10.

4. Physical capital losses and changing national wealth

Turning to physical capital losses in Table 10, we have largely relied for the losses of
domestic assets on Bogart’s (1920) estimates of property losses on land and shipping
and cargo losses from Table 7. However, whereas Bogart (1920) expressed the losses
in terms of US dollars, we have expressed them as percentages of prewar capital.
France’s losses were extremely heavy when expressed as a percentage of prewar
capital in Table 8, as well as in dollar terms in Table 7. Russia’s losses appear rather
heavier in proportionate terms than in absolute dollar values, due to the low level of
Russia’s prewar capital stock. Also in Table 8, for some countries it has been possible
to obtain estimates of the change in overseas assets and national wealth. In the case of
Britain, nearly a quarter of overseas investments were liquidated during the war, so
that the reduction of national wealth was proportionally much greater than the loss of
physical capital. For France, although the loss of overseas assets was proportionally
higher due to heavy exposure to Russian loans, the share of physical capital losses
was also much higher than in Britain (Hardach, 1977: 289-290). Hence the share of
national wealth lost in the war was about the same as the share of physical capital lost.

In principle, some of the government spending on the war effort, which is

included negatively as a direct cost by Bogart (1920) should actually enter positively
in the national balance sheet, contributing to intangible physical capital in the form of
cumulated research and development spending and to intangible human capital in the
form of spending on health and mobility. However, in practice, Broadberry and
Howlett (1998) found that these effects were very small even during World War II.
During World War I, these positive effects were difficult to discern at all in the British
case. Such effects were unlikely to have been of much more significance for other
countries.
 18

5. Reparations and national wealth

Finally in Table 10, we have added in Germany’s reparations bill as a proportion of
prewar capital, since they represented an increase in overseas liabilities and hence a
reduction in national wealth just as much as the liquidation of Britain’s overseas
assets meant a reduction in national wealth. Of course there is a huge debate over the
extent to which Germany actually had to pay these reparations, but that does not alter
the effect on the national balance sheet as it stood immediately after the Treaty of
Versailles (Ritschl, 2003).

IV. THE WIDER IMPACT ON GROWTH AND DEVELOPMENT

Milward (1984: 15-16) is critical of studies that focus on the costs of the war, which
he sees as neglecting the wider impact of the war on growth and development. This
reflects a substantial literature arguing that the two world wars stimulated economic
and social changes which had positive as well as negative effects (Andrzejewski,
1954; Titmuss, 1950). However, there are good grounds to be sceptical here. Milward
(1984: 17-18) cites Bowley (1930) as a pioneer of this view, but Bowley (1930: 21-
23) himself pointed out how difficult it is to show that any of these wider changes
were actually the result of the war and would not have occurred anyway in its
absence. Classifying developments as (a) mainly unconnected with the war, (b)
accelerated or retarded by it or (c) apparently arising out of it, Bowley was himself
reluctant to put anything other than the key elements of the “cost of war” calculations
such as loss of life and destruction of capital into category (c). He did mention the
new economic relation between Europe and the United States in this category, but
with hindsight we can see that the process of US overtaking was already underway
well before World War I (Abramovitz, 1986; Broadberry 1998).

1. The postwar role of government

Whilst holding to this generally sceptical view of the wider impact of the war, it is
nevertheless possible to draw some valuable lessons from a consideration of several
aspects of government. It is clearly true that the twentieth century has seen a
substantial increase in welfare spending. However, it is equally clear that this welfare
spending had its roots in the prewar period. Thus, for example, the German historical
economist Wagner (1890) had already formulated his eponymous law of a growing
share of government expenditure in national income on the basis of pre-World War I
trends, and the modern European welfare state was founded before rather than after
World War I (Lindert, 1994). Similarly, the growing tendency towards combinations
and scale in industry in North America and Europe was already under way well before
World War I and can be seen as a result of technological developments associated
with mass production rather than the outcome of wartime experience with state
regulation and control (Lamoreaux, 1985; Hannah, 1983).

Furthermore, although Peacock and Wiseman (1967: 24-28) formulated their

“displacement hypothesis” on the basis of experience during the two world wars, its
explanatory power is rather limited upon closer examination. Peacock and Wiseman
argued that the war displaced norms concerning acceptable levels of government
spending and taxation and appropriate levels of welfare spending and government
intervention in the economy. As a result, they argued that there was a ratchet effect,
with government spending increasing rapidly during the war through necessity, but
falling back by less after the war. Although this appears to fit aggregate spending
trends, it does not work once debt service payments are excluded. In other words, the
 19

only reason for the ratchet effect was the cost of debt service, with other types of
expenditure merely growing in line with national income. This suggests that
Andrzejewski’s (1954) military participation hypothesis is much overstated. Although
the high military participation ratio may have secured an increase in the absolute level
of welfare spending in the short run, it did not secure any increase in the share of
national income devoted to such spending in the long run.

2. Government and markets during the war

The above analysis suggests that the long run impact of the war on the role of
government may not be as great as suggested by some authors. However, it does not
challenge the traditional belief in the superiority of government intervention and
controls over market forces in the extreme circumstances of war. It is now worth
reflecting, therefore, on an alternative classical view of the war economy. Although
there is no detailed classical analysis of any particular economy during World War I,
there is a study by Ahmed (1986) of the British economy in the twentieth century as a
whole, covering both world wars. From a classical perspective this serves a useful
purpose to remind us that the differences between a war economy and a peacetime
economy may not be as stark as suggested in the traditional analysis. After all, it is
unlikely that the declaration of war suddenly makes governments all-knowing and all-
powerful, or leads to the suspension of all pursuit of selfish interests. There may be
some virtue, then, in analysing how we would expect a perfectly competitive market
economy to react to war. This can then be used as a benchmark against which to
assess the impact of the special measures and controls, rather than simply attributing
all change to such measures.

Barro (1974; 1981) has analysed the effects of government spending in a

closed economy, and the model has been applied to the United States during the major
wars of the nineteenth and twentieth century by Evans (1985). Ahmed (1986) adapts
Barro’s model to the open economy case and provides an econometric application to
the United Kingdom in the twentieth century. There are four key aspects to the model.
First, there is a temporary increase in government spending to fight a war. Although
this displaces some private spending, the “direct crowding out” effect is less than
proportional, since “guns” are not a good substitute for “butter” and people want to go
on consuming butter. Hence the level of aggregate demand increases. Second, there is
an increase in aggregate supply, as real wages increase to bring forth the required
extra labour. In a way, a war acts a bit like a “gold rush”, creating a temporary boom.
Third, if the increase in aggregate demand exceeds the increase in aggregate supply,
there is excess demand, and this can be met by a deterioration in the balance of trade
deficit. Fourth, it makes no difference to the level of economic activity whether the
increased government spending is financed by taxation or borrowing. Under this
“Ricardian equivalence” of taxation and bond finance, private spending decisions are
unaffected by the form of finance of government spending, since bond finance
represents a future tax liability, the present value of which is the same as the taxes
which would otherwise have to be raised now.

The model does capture the crude features of the British economy in both
world wars, and seems qualitatively applicable to other European countries. Overall
activity rose, consumption fell but by less than the increase in government spending,
and excess demand spilled over into an excess of imports over exports. Furthermore,
the issue of taxes versus bonds in a Ricardian framework becomes simply one of
 20

intergenerational transfers and tax smoothing, with a greater reliance on bond

financing spreading the burden onto future generations of taxpayers. Doubtless many
of the strong assumptions of the model do not hold, particularly with regard to the
ubiquity of perfect competition. Nevertheless, it suggests that we should not be too
quick to attribute all changes during wartime to the efficacy of regulations and
controls.

Few historians are likely to be persuaded by Ahmed’s (1986) argument that

the achievements of the British war economy can be put down to the smooth
operation of market forces during the war itself. However, the boom in the US
economy before 1917 was a decidedly market-led affair which seems to fit the
classical model well, with workers increasing labour supply to take advantage of the
high wages on offer in munitions factories. Furthermore, the classical view reminds us
that before the outbreak of war in 1914, Britain had a long history as a market
economy. Clearly, this had to be taken into account by those implementing state
controls during wartime. Also, it meant that Britain had the benefit of capabilities
developed in a market economy context before the war, including high levels of
productivity across all sectors and a high degree of flexibility.

However, the relevance of the classical real business cycle model to other
countries during World War I looks more questionable, at least without serious
modification. For, as noted earlier, in countries at lower levels of development, such
as Austria-Hungary, the Ottoman Empire, Russia, and even Germany, the key feature
of the war economy was a decline rather than an increase in GDP. This seems to have
occurred largely as a result of a retreat into subsistence by peasants working in
agriculture, as governments tried to shift the inter-sectoral terms of trade in favour of
urban areas so as to keep down the price of food for industrial workers producing vital
munitions.

3. Nationalism and economic disintegration

Finally, we cannot leave an evaluation of the wider impact of World War I on growth
and development without considering the effects on national rivalry and the road to
World War II. Although World War I may be seen as the culmination of a period of
existing national rivalry, there can be little doubt that it served to strengthen the forces
of nationalism. This can be seen as having serious economic consequences, giving a
boost to protectionism and autarkic policies during the 1920s and 1930s.

The consequences of this economic disintegration for the growth of per capita
income in Europe and other parts of the world can be seen in Table 11. The first point
to note is that growth of per capita GDP for a weighted average of fifteen European
countries was 1.8 per cent per annum between 1890 and 1994. However, whilst
Europe grew at roughly this secular rate before 1914 and after 1973, there was a
period of slower growth between 1913 and 1950, followed by a period of more rapid
growth between 1950 and 1973. This slower growth during 1913-1950 is interpreted
by Feinstein et al. (1997:8-9) as the destructive impact of World War I, followed by
the economic disintegration of the interwar period and the further destruction of
World War II. The argument is given added weight by the fact that the impact was
much greater in Europe than in the United States, since the war was fought largely on
European soil with unprecedented severity, and Europe’s economies were more
 21

dependent on international economic transactions before 1914. On this interpretation,

the period 1950-1973 is best seen as catching-up in a more integrated world economy.

Insert Table 11.

Insert Table 12.

Turning in Table 12 to variation between European countries in the growth

rate of GDP during the shorter period 1913-1929, we see that the most important
difference is between neutral and combatant countries. The lowest growth rate
amongst the neutrals (Sweden) was equal to the highest growth rate amongst the
combatants (France). This again supports the emphasis on the costs of war in the
traditional literature. Important themes stressed in this literature include the
protectionist environment and the general lack of international co-operation over the
international monetary system as well as the international trading system
(Eichengreen, 1992). One factor which needs to be mentioned here is the proliferation
of independent nation states following the break-up of the Austro-Hungarian and
Ottoman Empires. This was based on one of the founding principles of the League of
Nations, the self-determination of nations. In eastern and central Europe, this led to a
proliferation of states with separate currencies and customs jurisdictions. In a less
protectionist environment, this may not have been of great significance, but in the
context of protectionist interwar Europe, it clearly had serious trade-diverting effects.
Nevertheless, although there was clearly a net effect of economic disintegration in
central and eastern Europe, we should not forget that there were also areas of
increased integration. Probably of most significance here was the increased
integration of the reunited parts of Poland that had previously been partitioned
between Prussia, Austria and Russia (Wolf, 2003).

Moving beyond the narrowly economic effects of nationalism, one of the most
important developments, which cast a shadow over Europe for the next generation,
was the switch in focus of German nationalism away from overseas territories and
towards a “drive to the east”, as noted in Ritschl’s chapter on Germany. This
development can be seen as pointing the way to the horrors of World War II and the
holocaust, with the wrangling over the punitive reparations imposed by the Allies
hastening the journey. World War I also acted as the midwife to the Bolshevik
Revolution in 1917, which introduced a new economic and political schism in Europe
which was to cast another shadow over the world until the end of the 1980s.

V. TOTAL WAR AND ECONOMICS IN THE TWENTIETH CENTURY

After 1939 it became impossible not to see World War I as a dress rehearsal for
World War II. From this viewpoint the first war was rather like the second war, only
not as bad. It was terrible at the time, but this was partly because those involved did
not realise how much worse it could get. This is immediately obvious from any
statistical comparison of the two wars such as one that was published by a Russian
statistician in the last days of the old Soviet Union and is reproduced in Table 13. The
lesson is clear: World War II was simply World War I with more countries, more
soldiers, more time, more money, more guns, more death, and more destruction.

Insert Table 13.

 22

Of course, World War I had some distinct features. One is that economics decided
the outcome of the first war in a direct and straightforward sense, even more so than
in the second. The military decision of World War I was expected on the western
front, where the richest countries engaged most of their forces. Yet the military
decision never came. It is true that there were victories and defeats, and that the front
became considerably less stable during 1918. But the fact remains that the military
struggle ended in ceasefire, not surrender, with the German army still standing on
foreign soil. If Germany’s war effort had become unsustainable it was because of the
failure of its economy, not its army. It was the economic collapse of Austria-Hungary
that ended the military ambitions of the Habsburgs just as urban famine and industrial
collapse signed the death warrant of the Romanovs.

In this limited sense World War II was different: it ended in the crushing military
defeat of the Axis Powers. What remained the same is that the Allied victory of 1945,
like that of 1918, was enabled by an overwhelming predominance of resources.

We conclude by noting the special features of warfare in the first half of the
twentieth century. While there is much debate about the precise definition of “total
war” (Chickering and Förster, 2000). the period between 1914 and 1945 is distinctive
from an economic viewpoint. In both world wars the main combatants were able to
devote more than half of their national income to the war effort. This did not happen
before 1914, or after 1945, and it seems unlikely that it will ever happen again. Before
1914 it was impossible and after 1945 it was no longer necessary. Before the
twentieth century per capita incomes were too low and government services too
inefficient for society to devote such a large share of economic activity to warfare; too
many people were required to labour in the fields and workshops simply to feed and
clothe the population, and government officials were not up to the task of counting
and controlling them. After 1945, the destructive power of nuclear weapons meant
that any rich or large country could acquire devastating military force for a few billion
dollars. Hence the marshalling of economic resources played a much more vital role
in the outcome of the two world wars than in any period before or since. This is why
we maintain that the history of the world wars cannot be written without the
economics.
 23

TABLE 1: The World at War: Allied Populations, Territories, and GDPs of 1913

Gross Domestic
Popul- Territory, Product,
ation, million ha. per $ per
million sq. km head billion head, $
First Wave: Great Powers, 1914
Russian Empire, exc. Finland 173.2 21.7 12.6 257.7 1488
France 39.8 0.5 1.3 138.7 3485
United Kingdom 46.0 0.3 0.7 226.4 4921
Dependencies and Colonies
Finland (Russian Empire) 3.2 0.4 11.7 3.7 1140
French Colonies a 48.3 10.7 22.1 31.5 652
British Colonies b 380.2 13.5 3.6 257.0 676
Other Powers
Yugoslav States c 7.0 0.2 2.2 7.2 1029
British Dominions d 19.9 19.5 97.8 77.8 3909
Liberia 1.5 0.1 6.7 0.9 585
Japan 55.1 0.4 0.7 76.5 1387
Japanese Colonies e 19.1 0.3 1.6 16.3 857

Second Wave: 1915/16

Italy 35.6 0.3 0.8 91.3 2564
Italian Colonies f 2.0 2.0 101.0 1.3 634
Portugal 6.0 0.1 1.5 7.4 1244
Portuguese Colonies g 8.7 2.4 27.9 5.2 603
Roumania 7.7 0.1 1.8 11.7 1527

Third Wave: 1917/18

United States 96.5 7.8 8.1 511.6 5301
US Dependencies and Colonies h 9.8 1.8 18.9 10.6 1088
Central American States i 9.0 0.6 6.4 10.6 1184
Brazil 25.0 8.5 34.0 20.3 811
Greece 4.8 0.1 2.5 7.7 1592
Siam 8.4 0.5 6.2 7.0 835
China 441.5 11.1 2.5 243.7 552

November 1914
Allies, total 793.3 67.5 8.5 1093.6 1379
UK, France, and Russia only 259.0 22.6 8.7 622.8 2405
November 1916
Allies, total 853.3 72.5 8.5 1210.5 1419
UK, France, and Russia only 259.0 22.6 8.7 622.8 2405
November 1918
Allies, total 1271.7 80.9 6.4 1760.6 1384
Per cent of world 70% 61% … 64% …
UK, France, and USA only 182.3 8.7 4.8 876.6 4809
Per cent of world 10% 7% … 32% …

WORLD, 1913 1810.3 133.5 7.4 2733.9 1510

 24

Notes and sources for Table 1.

Sources: Populations and territories are from League of Nations (1927: 10-16). GDPs
per head are from Maddison (2001); where the country or territory is not listed, the
appropriate regional average is used.

Notes:
Figures show populations, territories, and incomes for the year 1913. Currency units
are international dollars at 1990 prices. Countries and territories are listed in
approximate order of their entry into the war.
a) Many countries in Africa, Asia, and Oceania. Algeria, French West Africa,
and Indo-China together accounted for more than 70% of the population and
GDP but less than half of the territory of the French Empire.
b) Many countries in Africa, Asia, and Oceania, including Anglo-French and
Anglo-Egyptian territories. India accounted for more than four fifths of the
population and GDP but only one third of the territory of the British Empire
not counting the Dominions.
c) Serbia, Bosnia-Herzegovina, and Montenegro.
d) Australia, Canada (including Labrador and Newfoundland), New Zealand, and
Union of South Africa.
e) Korea, Formosa, Kwantung, and Sakhalin.
f) Eritrea, Libya, Somalia, the Aegean Islands, and Tientsin.
g) Angola, Cape Verde Islands, Portuguese Guinea, Mozambique, St Thome and
Principe Islands, Portuguese India, Macao, and Timor and Cambing.
h) Alaska, American Samoa, Guam, Hawaii, the Panama Canal Zone, and
Phillipines.
i) Costa Rica, Cuba, Guatemala, Haiti, Honduras, Nicaragua, and Panama.
 25

TABLE 2: The Central Powers’ Populations, Territories, and GDPs of 1913

Gross Domestic
Popul- Territory Product
ation, million ha. per $ per
million sq. km head billion head, $
First Wave: Great Powers, 1914
Austria-Hungary 50.6 0.6 1.2 100.5 1986
Germany 67.0 0.5 0.8 244.3 3648
German Colonies, etc. a 10.7 3.0 27.5 6.4 601
Other Powers
Ottoman Empire b 19.6 1.8 9.0 15.5 794

Second Wave: 1915

Bulgaria 4.8 0.1 2.3 7.4 1527

November 1914
Central Powers, total 147.9 5.9 4.0 366.8 2480
Germany and Austria-
Hungary only 117.6 1.2 1.0 344.8 2933
November 1915
Central Powers, total 152.7 6.0 3.9 374.2 2450

Sources: Populations and territories are from League of Nations (1927: 10-16), except
Austria-Hungary taken from chapter 00. GDPs per head, except Austria-Hungary, are
from Maddison (2001); where the country or territory is not listed, the appropriate
regional average is used.

Notes:
Figures show populations, territories, and incomes for the year 1913. Currency units
are international dollars at 1990 prices. Countries and territories are listed in
approximate order of their entry into the war.
a) Cameroon, Caroline Islands, German East Africa, German South West Africa,
Klau-Chau, New Guinea Samoa, and Togoland.
b) Turkey in Europe and Asia, including Iraq, Palestine, Transjordan, Syria, and
Lebanon.
 26

TABLE 3: Allies Versus Central Powers: Resource and Development Ratios

Gross
Territory Domestic GDP per
Population Territory per head Product head
November 1914
Total 5.4 11.5 2.1 3.0 0.6
Great Powers only 2.2 19.4 8.8 1.8 0.8

November 1916
Total 5.8 12.3 2.1 3.3 0.6
Great Powers only 2.2 19.4 8.8 1.8 0.8

November 1918
Total 8.6 13.7 1.6 4.8 0.6
Great Powers only 1.6 7.5 4.8 2.5 1.6

Source: Calculated from tables 1 and 2. Figures show ratios of Allies (table 1) to
Central Powers (table 2) in populations, territories, and incomes for the year 1913.
Currency units are international dollars at 1990 prices.

TABLE 4: The Wartime Change in Real GDP: 1914-1918, by Country

UK USA Italy Germany Austria Russia France

1913 100 100 100 100 100 100 100
1914 92.3 101.0 97.4 85.2 83.5 94.5 92.9
1915 94.9 109.1 99.1 80.9 77.4 95.5 91.0
1916 108.0 111.5 100.7 81.7 76.5 79.8 95.6
1917 105.3 112.5 96.6 81.8 74.8 67.7 81.0
1918 114.8 113.2 94.3 81.8 73.3 … 63.9

Sources: Maddison (1995: 148-51), except Italy from Rey (1991: 203) and Russia
from Gatrell, this volume, table 2.
 27

TABLE 5: The Share of Government Spending in National Income: 1913-1918,

by Country (per cent of GDP at current prices)

Australia Canada France Germany Italy UK USA

1913 5.5 7.0 10.0 9.8 13.4 8.2 1.8
1914 5.7 10.0 22.3 23.9 12.8 23.5 1.9
1915 9.6 13.1 46.4 43.8 20.0 52.4 1.9
1916 14.0 16.5 47.2 50.3 30.5 63.7 1.5
1917 17.2 15.7 49.9 59.0 38.6 61.3 3.2
1918 17.2 16.9 53.5 50.1 35.4 49.4 16.6

Sources: Obstfeld and Taylor (2003); Mitchell (2003a, 2003b); Germany from
Sommariva and Tullio (1987), and France from chapter 00, Table 8. Thanks to Jari
Eloranta for help with these figures.

TABLE 6: Bogart’s “direct costs” of World War I

($m) ($)
Gross cost Advances to Net cost Net cost per
allies capita
Great Britain 44,029 8,695 35,334 766
Rest of British Empire 4,494 4,494 13
France 25,813 1,547 24,266 613
Russia 22,594 22,594 135
Italy 12,314 12,314 343
United States 32,080 9,455 22,625 229
Other Allies 3,964 3,964 127
Total Allies 145,288 19,697 125,591

Germany 40,150 2,375 37,775 557

Austria-Hungary 20,623 20,623 352
Turkey and Bulgaria 2,245 2,245 85
Total Central Powers 63,018 2,375 60,643

Total 208,306 22,072 186,234

Sources: Cost data from Bogart (1920: 267); Population data from Urlanis (1971:
209).
 28

TABLE 7: Bogart’s “indirect costs” of World War I ($m)

Capitalised Property Shipping

value of losses on and cargo
war deaths land losses
British Empire 3,477 1,750 3,930
France 4,818 10,000 453
Russia 8,104 1,250 933
Italy 2,385 2,710 431
United States 518 365
Other Allies 3,215 11,500 525
Total Allies 22,517 27,210 6,637

Germany 6,751 1,750 121

Austria-Hungary 3,080 1,000 15
Turkey and Bulgaria 1,203 27
Total Central Powers 11,034 2,750 163

Total 33,551 29,960 6,800

Source: Bogart (1920: 269-299).

Notes: For shipping losses, Other Entente Allies includes neutrals.

TABLE 8: Bogart’s “direct and indirect costs” of World War I ($m)

All
countries
Capitalised value of human life:
soldiers 33,551
civilians 33,551
Property losses:
on land 29,960
shipping and cargo 6,800
Loss of production 45,000
War relief 1,000
Loss to neutrals 1,750
Total indirect costs 151,612

Total direct costs, net 186,234

Grand total 337,846

Source: Bogart (1920: 269-299).

 29

TABLE 9: Battle and non-battle deaths of military personnel in World War I

Deaths Population Deaths as %

(1000s) (millions) of population
Great Britain 715 46.1 1.6
British Empire 198 342.2 0.1
France 1,327 39.6 3.4
French colonies 71 52.7 0.1
Russia 1,811 167.0 1.1
Italy 578 35.9 1.6
USA 114 98.8 0.1
Belgium 38 7.6 0.5
Serbia-Montenegro 278 4.9 5.7
Rumania 250 7.6 3.3
Greece 26 4.9 0.5
Portugal 7 6.1 0.1
Total Allies 5,413 813.4 0.7

Germany 2,037 67.8 3.0

Austria-Hungary 1,100 58.6 1.9
Turkey 804 21.7 3.7
Bulgaria 88 4.7 1.9
Total Central Powers 4,029 152.8 2.6

Total 9,442 966.2 1.0

Source: Urlanis (1971: 209).

Notes: Battle deaths includes killed in battle, died from wounds and died from poison
gas. Non-battle deaths includes died from disease, died in captivity and died from
accidents and other causes.
 30

TABLE 10: Destruction of human and physical capital (% of prewar assets)

Physical capital
Human Domestic Overseas Reparations National
capital assets assets bill wealth
Allies
Britain 3.6 9.9 23.9 … 14.9
France 7.2 59.6 49.0 … 54.7
Russia 2.3 14.3 … … ...
Italy 3.8 15.9 … … …
United States 0.3 … … … …
Central Powers
Germany 6.3 3.1 … 51.6 54.7
Austria-Hungary 4.5 6.5 … … …
Turkey and Bulgaria 6.8 … … … …

Sources: Human capital: war deaths as a percentage of population aged 15-49 from
Urlanis (1971: 209). Physical capital: Britain: Broadberry and Howlett, Table 13;
France: Hautcoeur, p.31 and Hardach (1977: 289-290); Russia: Gatrell, pp. 25-26;
Italy: Property and shipping losses from Bogart (1920), capital from Ercolani (1969);
Germany: Property and shipping losses from Bogart (1920), capital from Hoffmann
(1965), with reparations bill from Hardach (1977: 248); Austria-Hungary: Property
losses from Bogart (1920), capital from Fellner (1915).

Notes: Reparations bill expressed as % of prewar physical capital.

TABLE 11: Growth of real GDP, 1890-1994: Europe and the United States (per
cent per year, average)

Europe USA,
GDP population GDP per GDP per
head head
1890-1994 2.4 0.6 1.8 1.8
1890-1913 2.2 0.7 1.4 2.0
1913-1950 1.4 0.5 0.9 1.4
1950-1973 4.8 0.8 4.0 2.9
1973-1994 2.1 0.4 1.7 1.4

Source: Feinstein et al. (1997: 7, 9).

 31

TABLE 12: Growth of real GDP, 1913-1929: selected European countries

World War I World War I

neutrals % p.a. combatants % p.a.
Sweden 1.9 United Kingdom 0.7
Finland 2.4 France 1.9
Denmark 2.7 Italy 1.7
Switzerland 2.8 Belgium 1.4
Norway 2.9 Germany 1.2
Netherlands 3.6 Austria 0.3

Source: Feinstein et al. (1997: 13).

 32

TABLE 13: Two World Wars in Quantitative Comparison

World War I World War II

Length of war, days 1,564 2,194
Belligerent nations, number 33 62
Theatres of military action:
number of nations 14 40
number of continents 1 4
Population of belligerent nations, millions 1,100 1,700
Of which, number mobilised 70 110
number wounded 20 35
number disabled 15 25
Excess deaths, millions 20 55
Of which:
deaths among servicemen 10 32
deaths among civilians 10 23
including:
in concentration camps … 11
of partisans … 2
from hunger and disease 10 10
Munitions produced, units:
thousand guns 150 1,040
thousand aircraft … 700
thousand tanks 9 300
Economic losses, $billion at 1938 prices 692 4,000
Of which, direct losses 416 1,433
including:
budget outlays 354 1,117
destruction, looting 316 316
indirect losses 258 2,567

Source: Nesterov (1990), 6. The valuation of economic losses in both wars is

evidently based on the Bogart methodology that was critically described, found to be
at fault, and revised in Tables 4 through 8 above; Bogart’s own figures for World War
I have been revalued by the 1938 benchmark used for World War II. With this caveat
the comparison is still informative and the estimated sign and slope of the gradient
from the first war to the second are plausible.
 33

REFERENCES
Abramovitz, M. (1986), “Catching Up, Forging Ahead and Falling Behind”, Journal
of Economic History, 46, 385-406.
Adelman, J.R. (1988), Prelude to the Cold War: The Tsarist, Soviet, and U.S. Armies
in Two World Wars, Boulder, CO, and London: Lynne Rienner.
Ahmed, S. (1986), “Temporary and Permanent Government Spending in an Open
Economy: Some Evidence for the United Kingdom”, Journal of Monetary
Economics, 17, 197-224.
Andrzejewski, S. (1954) Military Organisation and Society, London: Routledge &
Kegan Paul.
Barro, R.J. (1974), “Are Government Bonds Net Wealth?”, Journal of Political
Economy, 82, 1095-1117
Barro, R.J. (1981), “Output Effects of Government Purchases”, Journal of Political
Economy, 89, 1086-1121.
Bogart, E.L. (1920), Direct and Indirect Costs of the Great World War, (2nd edition),
New York: Oxford University Press.
Brennan, G., and Tullock, G. (1982), “An Economic Theory of Military Tactics:
Methodological Individualism at War,” Journal of Economic Behavior and
Organization, 3(2-3), 225-42.
Broadberry, S.N. (1998), “How did the United States and Germany Overtake Britain?
A Sectoral Analysis of Comparative Productivity Levels, 1870-1990”, Journal of
Economic History, 58, 375-407.
Broadberry, S.N. and Howlett, P. (1998), “The United Kingdom: ‘Victory at All
Costs’”, in Harrison, M. (ed.), The Economics of World War II: Six Great Powers
in International Comparison, Cambridge: Cambridge University Press, 43-80.
Chickering, R. and Förster, S. (2000) (eds.), Great War, Total War: Combat and
Mobilization on the Western Front, 1914-1918, Cambridge: Cambridge University
Press.
Clark, J.M. (1931), The Cost of the World War to the American People, New Haven:
Yale University Press.
Edelstein, M. (2000), “War and the Amerciasn Economy in the Twentieth century”, in
Engerman, S.L. and Gallman, R.E. (eds.), The Cambridge Economic History of
the United States, Volume III: The Twentieth Century, Cambridge: Cambridge
University Press, 329-406.
Eichengreen, B. (1992), Golden Fetters: The Gold Standard and the Great
Depression, 1919-1939, New York: Oxford University Press.
Eloranta, J. (2003), “Responding to Threats and Opportunities: Military Spending
Behavior of the Great Powers, 1870-1913”, University of Warwick, Department
of Economics.
Ercolani, P. (1969), “Documenti statistica di base”, in Fuà (ed.), Lo sviluppo
economico in Italia, Vol. III, Milan, 380-460.
Evans, P. (1985), “Do Large deficits Produce High Interest Rates?”, American
Economic Review, 75, 68-87.
Feinstein, C.H., Temin, P. and Toniolo, G. (1997), The European Economy Between
the Wars, Oxford: Oxford University Press.
Gatrell, P. (1999), A Whole Empire Walking: Refugees in Russia During World War I,
Bloomington, IN: Indiana University Press.
Gatrell, P. (1999), A Whole Empire Walking: Refugees in Russia During World War I,
Bloomington, IN: Indiana University Press.
 34

Gerschenkron, A. (1962), Economic Backwardness in Historical Perspective,

Cambridge MA: Harvard University Press.
Goldsmith, R.W., Lipsey, R.E. and Mendelson, M. (1963), Studies in the National
Balance Sheet of the United States, (2 volumes), Princeton, NJ: Princeton
University Press.
Hannah, L. (1983), The Rise of the Corporate Economy, (2nd edition), London:
Methuen.
Hardach, G. (1977), The First World War, 1914-1918, Berkeley: University of
California Press.
Harrison, M. (1998), ed., The Economics of World War II: Six Great Powers in
International Comparison, Cambridge: Cambridge University Press.
Herwig, H.H. (1997), The First World War: Germany and Austria-Hungary, 1914-
1918, London: Arnold.
Hoffmann, W.G. (1965), Das Wachstum der deutschen Wirtschaft seit der Mitte des
19. Jahrhunderts, Berlin: Springer-Verlag.
Kendrick, J.W. (1976), The Formation and Stocks of Total Capital, New York:
Columbia University Press.
Lamoreaux, N.R. (1985), The Great Merger Movement in American Business, 1895-
1904, Cambridge: Cambridge University Press.
League of Nations (1927), International Statistical Yearbook, 1926, Geneva: League
of Nations, Economic and Financial Section.
Lee, J. (1975), “Administrators and Agriculture: Aspects of German Agricultural
Policy in the First World War”, in Winter, J.M. (ed.), War and Economic
Development: Essays in Memory of David Joslin, Cambridge: Cambridge
University Press, 229-238.
Lindert, P. (1994), “The Rise of Social Spending, 1880-1930”, Explorations in
Economic History, 31, 1-37.
Maddison, A. (2001), The World Economy: A Millennial Perspective, Paris: OECD.
Mazower, M. (1998), Dark Continent: Europe’s Twentieth Century, London: Allen
Lane
Milward, A.S. (1984), The Economic Effects of the Two World Wars on Britain, (2nd
edition), London: Macmillan.
Mitchell, B. (2003a), International Historical Statistics: Europe, 1750-2000 (5th
edition), Basingstoke: Palgrave.
Mitchell, B. (2003b), International Historical Statistics: The Americas, 1750-2000
(5th edition), Basingstoke: Palgrave.
Nesterov, L. (1990), “Tsena voiny,” Vestnik statistiki, no. 5, 000-000.
Obstfeld, M., and A.M. Taylor (2003). “Sovereign risk, Credibility, and the Gold
Standard: 1870-1913 versus 1925-31,” Economic Journal, 113(487), 241-75.
Olson, M. (1963), The Economics of the Wartime Shortage: A History of British Food
Supplies in the Napoleonic War and in World Wars I and II, Durham, NC: Duke
University Press.
Overy, R.J. (1998), “Who Really Won the Arms Race?”, The Times Literary
Supplement, 13 November, 4-5.
Peacock, A.T. and Wiseman, J. (1967), The Growth of Public Expenditure in the
United Kingdom, (2nd edition), London: Allen and Unwin.
Revell, J. (1967), The Wealth of the Nation: The National Balance Sheet of the United
Kingdom, 1957-1961, Cambridge: Cambridge University Press.
Rey, G.M., ed.(1991), I conti economici dell’Italia, vol. 1. Una sintesi delle fonti
ufficiali. 1890-1970, Roma-Bari: Editori Laterza.
 35

Ritschl, A. (2003), “Dancing on a Volcano: The Economic Recovery and Collapse of

Weimar Germany, 1924-33”, in Balderston, T. (ed.), The World Economy and
National Economies in the Interwar Slump, Basingstoke: Palgrave.
Sommariva, A., and G. Tullio (1987), German Macroeconomic History, 1880-1979: a
Study of the Effects of Economic Policy on Inflation, Currency depreciation, and
Growth, Basingstoke, Macmillan.
Strachan, H. (2003). The First World War, vol. 1, A Call to Arms, Oxford: Oxford
University Press.
Urlanis, B. (1971), Wars and Population, Moscow: Progress.
Wagner, A. (1890), Finanzwissenschaft, Leipzig: Winter.
Wolf, N. (2003), Economic Integration in Historical perspective: The Case of
Interwar Poland, 1918-39, unpublished PhD dissertation, Humboldt Universität zu
Berlin.