The Cash Nexus: Money and Politics in Modern History, 1700-2000

by Niall Ferguson

  Because of the Cold War, the sharp reductions in military budgets that followed the defeat of the Axis powers were short-lived. Having fallen to just 21 per cent of the central government budget in 1949, British defence spending rose to a post-war peak of 38 per cent in 1954, which was also the peak year for France. The ‘shrinking pains’ of decolonization faded thereafter: the British defence budget was already in relative decline by the time of Suez, while the French fell rapidly after Dien Bien Phu. By 1968 defence accounted for just a fifth of spending in both countries.83 Nor was the downward trend of British defence spending more than slowed by the Thatcher government. As a share of expenditure it rose only slightly from 10 per cent in 1975 to 11.8 per cent in 1986; but in 1990 it was back down to 10.7 per cent.84 In 1997–8 it accounted for less than 7 per cent of the general government spending ‘control total’. This is a figure lower than at any time in British history since the Wars of the Roses.

  None of the above figures, however, tells us the relative economic importance of military expenditure. Indeed, given the profound changes in the nature of total state expenditure, not only at the central but also at the local level, it may be that they tell us hardly anything meaningful at all. For example, in order to make German and British figures comparable after 1870, the spending of the German states of Länder need to be added to the federal government’s expenditure total; or alternatively, the defence budget should be calculated as a proportion of total public sector spending, including all tiers of government. Table 1 gives a rather better indication of the remarkable decline of military expenditure in relation to public spending by all levels of government in the past hundred years. In Britain, France and Germany alike, the share of defence spending in the total public sector budget has declined from around a quarter to barely a twentieth.

  Table 1. Defence expenditure as a percentage of total public spending, 1891–1997

  Sources: 1891–1971: Flora et al., State, Economy and Society, vol. i, pp. 345–449; 1997: SIPRI and OECD.

  More important than calculations of that sort, are those which express the ‘military burden’ of expenditure as a proportion of total economic output. To give a classical example: Goldsmith estimates total Athenian public expenditure at around 20 per cent of national product – necessarily a very approximate calculation – compared with an equivalent figure for Augustan Rome of no more than 5 per cent. In relative economic terms, therefore, the Greek military burden was probably higher than the Roman: perhaps around 7 per cent of national product, compared with a Roman figure of just 2 or 3 per cent. This kind of calculation – the cost of military expenditure in relation to gross national or domestic product – is not without its technical difficulties even in the present day, when estimates of national product are relatively reliable, though still far from perfect. Nevertheless, there is no better way of estimating relative military spending that allows comparisons between countries and over time.

  The proportion of military spending in relation to national product naturally fluctuates quite substantially according to whether or not a state is at war; and this is the crucial point. In the case of Medici Florence in the 1420s, for example, the ratio of military spending to ‘national’ product varied between 3 per cent in peacetime and 20 per cent during wars.85 As a proportion of national income, British defence spending in the eighteenth century varied between 4 and 18 per cent depending on whether or not the country was at war, reaching a peak between 1778 and 1782.86 This was a significantly larger proportion than the French state spent in the same period. According to one calculation, total British war expenditure between 1776 and 1782 was nearly two and a half times the equivalent French figure in absolute terms. However, this differential does not take into account the relative size of the rival states’ economies. In fact, the cost in relation to a year’s GNP was even higher for Britain than the absolute numbers suggest: 75 per cent compared with just 15 per cent for France.87 In relative terms, war was far more burdensome for Britain than for France; or, to put it differently, Britain was able to mobilize a larger share of national product at times of military crisis.

  Figure 4. Defence spending as a percentage of national product, 1850–1998 (log. scale)

  Sources: Defence spending: UK: 1850–1914: Correlates of War database; 1914–88: Butler and Butler, British Political Facts, pp. 393 f.; 1989–98: SIPRI. US: 1870–1913: Hobson, ‘Military-extraction Gap and the Wary Titan’, p. 501; 1914–85: Correlates of War database; 1986–98: SIPRI. Germany: 1872–1913, 1925–32: Andic and Veverka, ‘Growth of Government Expenditure’, p. 262; 1933–8: Overy, War and Economy, p. 203; 1938–44: Petzina, Abelshauser and Faust (eds.), Sozialgeschichtliches Arbeitsbuch, vol. iii, p. 149 (however, 1933–43 percentages are from Abelshauser, ‘Germany’, p. 138); 1950–80: Rytlewski (ed.), Bundesrepublik in Zahlen, pp. 183 f.; 1982–98: SIPRI. France: 1820–70: Flora et al., State, Economy and Society, vol. i, pp. 380–82; 1870–1913: Hobson, ‘Military-extraction Gap and the Wary Titan’, p. 501; 1920–75: Flora et al., loc. cit.; 1981–97: SIPRI. Italy: 1862–1973: Flora et al., op. cit. vol. i, pp. 402 ff.; 1981–97: SIPRI. Russia: 1885–1913; Hobson, ‘Military-extraction Gap and the Wary Titan’, p. 501; 1933–8: Nove, Economic History, p. 230; 1940–1945: Harrison, ‘Overview’, p. 21; 1985–91: IISS, Military Balance; 1992–7: SIPRI. GDP/GNP/NNP: UK: 1850–70: Mitchell, European Historical Statistics, p. 408; 1870–1948: Feinstein, National Income, Expenditure and Output, statistical tables, table 3; 1948–1998: ONS. US: 1850–1958: Mitchell, International Historical Statistics: The Americas, pp. 761–74; 1959–98: Federal Reserve Bank of St Louis. Germany: 1870–1938: Hoffmann, Grumbach and Hesse, Wachstum; 1950–60: Rytlewski (ed.), Bundesrepublik in Zahlen, p. 188; 1960–99: OECD. France: 1820–1913: Lévy-Leboyer and Bourgignon, L’Économie française, pp. 318–22; 1960–99: OECD.

  Notes: UK: GDP figures after 1920 excluding Southern Ireland. Germany: GDP 1950–60: West Germany, excluding Saarland and W. Berlin; 1960–90: West Germany; 1991–9: reunified Germany.

  As Figure 4 shows, such levels were rarely attained in the nineteenth century. Between 1850 and 1914 the highest proportion of GDP consumed by the British armed services was just 11 per cent in the first year of the Crimean War; even during the Boer War the figure did not rise above 6 per cent. None of the other European powers ever spent more than 5 per cent of national output on defence, with the exception of Italy in 1866 (though if GDP figures were available for Prussia before German unification, the military quota would almost certainly exceed 5 per cent in the period 1866–71). Average defence expenditure as a percentage of net national product between 1870 and 1913 amounted to just 3.1 per cent for Britain and Austria, 3.2 per cent for Germany, 3.3 per cent for Italy and 4 per cent for France.

  Considering how much has been written on the subject of the pre-First World War arms race – not to mention the scramble for overseas empires – these numbers are surprisingly low. It is especially striking that Germany, the state most notorious for its ‘militarism’ in this period, was by this measure somewhat less militaristic than her two neighbours and rivals, France and Russia.88 However, the idea of ‘militarism run mad’ as a general European phenomenon seems more intelligible when these figures are compared with those for the United States. On average, Americans spent less than 1 per cent of net national income on the military between 1870 and 1913. Nor was this significantly altered by the First World War. Only in the last year of the Great War did defence spending rise above 5 per cent of GNP and, after peaking at 13 per cent in 1919, it rapidly fell back down below 1 per cent for most of the 1920s. Again, the contrast with the European powers is very marked. At their respective peaks in the First World War, both Britain and Germany spent more than 50 per cent of GDP on the military; Italy was not far behind with 35 per cent.

  The inter-war period saw a vain attempt by Britain to return to the pre-war pattern of expenditure; no other power attempted to do so. From the mid-1920s onwards both Italy and France increased milit
ary expenditure ahead of the growth rate: the French defence burden exceeded 5 per cent of GDP in 1930, the Italian in 1935. Germany, of course, had its military budget slashed almost to American levels by the Versailles Treaty; but after Hitler came to power an immense shift of resources took place, increasing the military quota from less than 2 per cent in 1933 to 23 per cent in 1939.

  To the European powers, the relative cost of the Second World War was in fact not much greater than had been the First. The most striking difference, however, was that from 1943 onwards the United States for the first time began to divert resources to warfare on a scale comparable with the European states. Nor, since this ‘rise to globalism’, has it been possible for Americans to revert to their earlier level of military parsimony. On the contrary: since the time of the Korean War the United States has consistently spent a higher proportion of GDP on defence than her principal allies. Needless to say, this reflected the high level of military expenditure necessitated by the Cold War.

  The greatest difficulties arise in the case of Russia and the Soviet Union: hence the many gaps in the series in Figure 4. This is because of the patchiness of Tsarist data and, more seriously, the idiosyncrasies of Soviet accounting conventions – notably the concept of ‘net material product’, which effectively excluded services from the national accounts – as well as the policy of under-pricing armaments mentioned above. Before the First World War, Tsarist Russia was certainly the most economically militaristic of the great powers, spending more than 5 per cent of net national product on defence between 1885 and 1913 – though this average was undoubtedly inflated by the relatively high cost of the 1904–5 war with Japan. Between 1915 and 1917 the military burden also probably rose slightly higher than those of the other combatants. The picture, however, becomes obscure in the Soviet period. If defence expenditure appears to have been relatively low in the period of the New Economic Policy and Stalin’s collectivization, it rose quite rapidly after 1935: ahead of Britain’s, though behind Germany’s. At the height of the Second World War the relative military burden exceeded 60 per cent, still slightly less than the same figure for Germany. It is much harder to be sure how much of Soviet output went on defence after 1945, however. Official Soviet figures were certainly too low. In 1975 the Central Intelligence Agency doubled its estimate of Soviet military spending from 6–8 per cent of GNP to 11–13 per cent on the basis of new price data.89 Ten years later the International Institute for Strategic Studies put the figure at 16 per cent.90 The equivalent figure for the United States at this time was 6 per cent. Even at the height of the Korean War American defence spending as a proportion of output was below the Soviet level of the 1980s.

  Finally, Figure 4 shows how sharply defence expenditure has fallen in relative terms since the end of the Cold War. The latest estimates from the Stockholm International Peace Research Institute (for 1999) suggest expenditure to GDP ratios of around 4 per cent for Russia, 3.2 per cent for the United States, 2.8 per cent for France, 2.6 per cent for Britain, 2 per cent for Italy and just 1.5 per cent for Germany.91 These are figures reminiscent of the 1920s, if not the nineteenth century. The United States, Russia, Germany and Britain have not spent so little on defence since the 1920s, though in the German case this was under duress. French and Italian defence spending has not been so low in relative terms since the early 1870s.

  THE ‘DEMILITARIZATION’ OF THE WEST

  The demilitarization of the West in the late twentieth century seems remarkable when compared with the era of the world wars. The average Western man now has every chance of avoiding war altogether. Indeed, the most violent experience he is ever likely to have is a Saturday night brawl or a mugging. If he has an appetite for war, he must rest content with electronically generated visions: occasional television bulletins from far-off places or, more often, cinematic re-enactments of past wars or fictional future wars. In the first half of the twentieth century, men saw action: their grandsons and great-grandsons see acting. In 1999 many thousands of American actors feigned death in harrowing but hugely popular war films like Saving Private Ryan. Only a handful of American soldiers died as a result of real military operations, and all were the victims of accidents rather than enemy action.

  Yet it would be wrong to attribute demilitarization to that revulsion against war which characterized both élite and ‘pop’ culture during and after the Vietnam War. Demilitarization has been the norm in times of peace, as Figures 2 and 4 make clear. In addition, there has been a long-run tendency in Britain and the United States to reduce military participation by substituting capital for labour.

  Historically, the two most appealing things about war have been the pleasure of comradeship and the excitement of combat. But with the advance of military technology in the twentieth century, both experiences became more elusive. The nadir of conventional warfare was reached on the Eastern Front in the Second World War. With the death toll averaging nearly one in three, there could be no enduring bonds and no thrill, simply a desperate struggle for survival:

  Man becomes an animal. One must destroy in order to live. There is nothing heroic on this battlefield … The battle returns here to its most primeval, animal-like form; whoever does not see well, fires too slowly, fails to hear the crawling on the ground in front of him as the enemy approaches, he will be sent under … The battle here is no assault with ‘hurrah cries’ over a field of flowers.92

  In this war, female medics used their teeth to amputate smashed limbs.93 Starving prisoners of war were reduced to cannibalism. This was not just total war but totalitarian war, in which the value of human life sank close to zero on the battlefield, and to precisely zero in the slave labour camps which were an integral part of the war effort on both sides.94

  The alternative route, taken by the United States and Britain precisely in order to economize on lives, was to industrialize war – shifting resources into artillery, tanks, warships and, above all, aircraft. In many ways, the turning point was 1940, when Britain evacuated her army from Dunkirk and then relied on a force of just 1,400 fighter pilots to deter a German invasion and keep Britain in the war.95 But it was the bomber rather than the fighter that became the key to subsequent British (and American) strategy. In effect, investment in bombers reduced casualties among Allied servicemen and greatly increased casualties among Axis civilians, a process that culminated at Hiroshima. Once dominance of the skies had been established, ground forces could be used at a far lower cost to life and limb.

  The present ‘Revolution in Military Affairs’ made possible by improvements in electronic communications is therefore part of a prolonged and far from revolutionary process. What does not change over the long run is that money must be found – whether it is for the mass armies of the age of total war, or the ‘smart weapons’ that account for a rising share of modern military budgets. And often, as this chapter has made clear, the money needs to be found at very short notice. The sums involved have varied greatly in relation to economic growth, as well as in relation to the destructive efficiency of weaponry. But this basic need to finance war has been – until the relatively recent past – the prime mover in the process of state formation; the father, indeed, of what follows.

  2

  ‘Hateful Taxes’

  And it came to pass in those days, that there went out a decree from Caesar Augustus, that all the world should be taxed.

  LUKE 2: 1

  ‘In this world,’ as one revolutionary wrote to another in the fateful year 1789, ‘nothing can be said to be certain except death and taxes.’1 Even in the New Testament, tax plays its part: it was to render what was due to Caesar that Mary and Joseph went to Bethlehem. Without tax, Christ would not have been born in a manger.

  The quest for increased revenues – usually, as we have seen, to pay for war or preparation for war – has led in more than one direction. In some systems, including feudal monarchies and socialist republics, a substantial portion of revenue has come from state-owned assets, whether royal dom
ains or ‘nationalized’ monopolies. In theory, then, taxes in the conventional sense are not quite inevitable: a state could notionally rely exclusively on public assets to generate revenue. But the profits from those assets would be generated by taxes of a sort, whether in the form of additional labour by royal serfs or above-cost charges by state industries. In any case, the temptation to sell state assets to meet sudden increases in expenditure has tended to mean that such assets dwindle over time: the sale of crown lands in the medieval period has its modern counterpart in the ‘privatization’ of publicly owned utilities. Taxes are therefore inevitable – though not unavoidable.

  In systems with limited representation confined to wealthy élites, there is a tendency to rely heavily on indirect taxation – principally customs levied on imports and excise duties on consumption – for revenue. The taxation of consumption may, within certain limits, be economically preferable to the alternative, namely taxation of wealth and incomes. But indirect tax rarely suffices for long because, first, in times of crisis trade and consumption tend to be reduced, and with them also tax revenues; secondly, because indirect taxation is usually regressive, and over-reliance on it can lead to political unrest. Sooner or later most states have therefore been obliged to raise direct taxes, such as levies on property or deductions from income.