《公共经济学》课程教学资源（书籍文献）The Growth in Government Expenditure and National Income.pdf_大学文库

CHAPTER 3 The Growth in Government Expenditure and National Income OuR case-study of British government expenditure since 1890 begins with presentation of the actual statistics of total annual expenditures and com- parison of the changes indicated by these statistics with the changes in national income.We then examine the effects of the permanent influences (population,price,and level of employment changes)on government expenditure.First,however,we present a brief history of government expenditures in the earlier part of the nineteenth century,intended to serve as a useful general background for the later argument. British Government Expenditure Before 18go In his famous chapter on "The Limits of the Province of Government," John Stuart Mill stated the maxim that "the business of life is better performed when those who hav immediate interest init are left to take their own course,uncontrolled either by the mandate of the law or by the meddling of any public functionary."1 This maxim,shorn of Mill's own detailed qualifications of it,dominated the actions of the statesmen who followed the "old"liberalism.Translated into practical financial policy,it meant that the level of goverment expenditure was to be kept at th tection against the Crown's enemies and of the maintenance of law and order;a wide interpretation of the latter included,to the Victorian mind, the relief of certain forms of social distress. So universal was the acceptance of this attitude that it is difficult to detect in the hist ablic fin ance in the nineteenth century any pronounced disagreement.The main sources of disagreement were about methods rather than aims and were concerned with how best to hold expenditure in check or reduce it.On the expenditure side of the budget,the debate was thus for the most part about the ways in which the payide nts of national debt interest might be reduced and about how to prov 。th cepted govern ent service es as econ in o aer- Gladstone's graphic phrase to bring about"the savingo candle ends.The real truge in Parliament was conerednot with expenditure but with methods of raising revenue,with the relative merits of direct and indirect taxation,the object of minimizing the level of government expenditure being taken for granted.2 P (Ahley tio)ondon,Book V Chapter quiofide toward publicndingand theiresdiscusdin greater detail in Chapter 5. 35

CHAPTER 3 The Growth in Government Expenditure and National Income OUR case-study of British government expenditure since 1890 begins with presentation of the actual statistics of total annual expenditures and comparison of the changes indicated by these statistics with the changes in national income. We then examine the effects of the permanent influences (population, price, and level of employment changes) on government expenditure. First, however, we present a brief history of government expenditures in the earlier part of the nineteenth century, intended to serve as a useful general background for the later argument. British Government Expenditure Before 1890 In his famous chapter on "The Limits of the Province of Government," John Stuart Mill stated the maxim that "the business of life is better performed when those who have an immediate interest in it are left to take their own course, uncontrolled either by the mandate of the law or by the meddling of any public functionary." This maxim, shorn of Mill's own detailed qualifications of it, dominated the actions of the statesmen who followed the "old" liberalism. Translated into practical financial policy, it meant that the level of government expenditure was to be kept at the minimum consistent with the provision of adequate protection against the Crown's enemies and of the maintenance of law and order; a wide interpretation of the latter included, to the Victorian mind, the relief of certain forms of social distress. So universal was the acceptance of this attitude that it is difficult to detect in the history of British public finance in the nineteenth century any pronounced disagreement. The main sources of disagreement were about methods rather than aims and were concerned with how best to hold expenditure in check or reduce it. On the expenditure side of the budget, the debate was thus for the most part about the ways in which the payments of national debt interest might be reduced and about how to provide the accepted government services as economically as possible, in order—in Gladstone's graphic phrase—to bring about "the saving of' candle ends." The real struggle in Parliament was concerned not with expenditure but with methods of raising revenue, with the relative merits of direct and indirect taxation, the object of minimizing the level of government expenditure being taken for granted.2 1 Principles of Political Economy (Ashley edition), London, 1909, Book V, Chapter XI, p. 952. 2 The question of attitudes toward public spending and their effects is discussed in greater detail in Chapter 5. 35

GROWTH IN EXPENDITURE AND NATIONAL INCOME Accurate and relevant statistics of government expenditure and its difficult to come by.Table 1 derives from provisional statistics collected for a study of this period by Jindrich Veverka,now in preparation (see Appendix, "Notes on Tables Not Derived from Appendix Tables").The figures are coparable with the data fo th later period,bcentl some general conclusions ab out how relatively narrowly the functions of government were conceived in the nineteenth century, and about the way the position of the government in the economy has changed since 1800. The table shows that,after declining from a peak of 23 million during the Napoleo nic Wars (1814),goverr t expen diture a current prices remained fairly constant in size until the period of the Crimean War (1854-56).Current money expenditures were one-third higher in 1860 than in 1850,and then continued to rise slowly through the 1870's.In 1880(the period of the Zulu Wars)expenditures were about 25 per cent greater than in 1860.Thereafte ther was a new period of gradu increase,until the rough expenditure plateau was broken by the expendi ture peak in 1900-the period of the South African War,the end of the period under survey,and ten years after the beginning of our period Droper. These statistics of total ex nditures at cur ent prices may be thought toprovidea misleadi ng and unrea listic picture of the e growth i n importanc of government,and from some points of view this is certainly true.Such statistics are given special interest for the period being considered,how- ever,by the fact that the absolute level of government expenditure at curent prices was uppermost in the minds of the nineteenth century Writing it ust after th of the the statistician Robe t Giffen complai ned that in this perio there was"no real discussion of what the expenditure of the state should be and for what purposes,and of what could really be borne by the community,any more than there is now or ever has been at any time in my recollection."3 Taken together with the attitude to expenditures just described,such thinking in money terms meant that the share of government in aps likely to be a passive later generations. However,for us as for Giffen,changes in real government expenditures and in the share of such expenditures in real GNP are also matters of great interest,whatever their causes.As Table I also shows,the pro- 36

GROWTH IX EXPENDITURE AND NATIONAL INCOME Accurate and relevant statistics of government expenditure and its relation to national income for the period before 1890 are difficult to come by. Table 1 derives from provisional statistics collected for a study of this period by Jindrich Veverka, now in preparation (see Appendix, "Notes on Tables Not Derived from Appendix Tables"). The figures are not fully comparable with the data for the later period, but are sufficiently so for us to draw some general conclusions about how relatively narrowly the functions of government were conceived in the nineteenth century, and about the way the position of the government in the economy has changed since 1800. The table shows that, after declining from a peak ofL 123 million during the Napoleonic Wars (1814), government expenditure at current prices remained fairly constant in size until the period of the Crimean War (1854—56). Current money expenditures were one-third higher in 1860 than in 1850, and then continued to rise slowly through the 1870's. In 1880 (the period of the Zulu Wars) expenditures were about 25 per cent greater than in 1860. Thereafter, there was a new period of gradual increase, until the rough expenditure plateau was broken by the expenditure peak in 1900—the period of the South African War, the end of the period under survey, and ten years after the beginning of our period proper. These statistics of total expenditures at current prices may be thought to provide a misleading and unrealistic picture of the growth in importance of government, and from some points of view this is certainly true. Such statistics are given special interest for the period being considered, however, by the fact that the absolute level of government expenditure at current prices was uppermost in the minds of the nineteenth century Chancellors of the Exchequer. Writing just after the turn of the century, the statistician Robert Giffen justifiably complained that in this period there was "no real discussion of what the expenditure of the state should be and for what purposes, and of what could really be borne by the community, any more than there is now or ever has been at any time in my recollection."3 Taken together with the attitude to expenditures just described, such thinking in money terms meant that the share of government in community output was perhaps more likely to be a passive consequence of changes in the level of prices than has been the case in later generations. However, for us as for Giffen, changes in real government expenditures and in the share of such expenditures in real GNP are also matters of great interest, whatever their causes.4 As Table 1 also shows, the pro- 3 See Sir Robert Giffen, Economic Inquiries and Studies, London, 1904, II, 310—311, 325. 4 But see our discussion of the problems of interpretation of these indicators of expenditure growth in Chapter 1. 36

GROWTH IN EXPENDITURE AND NATIONAL INCOME iof tota govement exenditure to GNwas aroundperc nic Wars.After wars,a highe sh for some time,but the old proportion had become established again by the decade 1830-40.Such a return was not unexpected,given the pre- vailing attitude to public expenditures,and given the fact that the goverments of the tim were aware of the potential displacement effec ars and were actively concerned to prevent a permanent upwarc shift in the share of government as a consequence of such wars.From 1841 until 1890 the share of government changes little,declining very slowly from 11 per cent of GNP in 1841 to 9 per cent in 1890.This com- pares with a share of 12 per cent,24 per cent in 1923 and 37p cent in 1955(Appendix Table A-6).The gradu al decline over the secon half of the nineteenth century is the more interesting for being associated after 1875 with a declining price trend which(given the interest of the Chancellors of the day in money expenditures)might have been expected fNapoleoni Era the thr site of expenditur were central government expenditure on debt interest,military expendi- ture.and local government expenditure on poor relief.The size of deb interest was being restricted,if not actively reduced,from that time until the beginning of the South African War,and was becoming a smaller proportion of national pro equ Thus debt current prices wa million in 1860,and 2 million in 1900.Until the Crimean War,debt interest was almost a half of all government expenditure,in 1880 it was only a quarter and in 1890 a fifth.A hundred y ears after the Crimean War,as we shall see,debt expenditure ted as little as Il per cent of total re,but it ad been up to nearly 30 per ent tw tw e mc e in an opposi n from tha debt interest payments.Drastically reduced after the Napoleonic Era they gradually rose after the Crimean War until they overtook national debt interest in the 1880's.From that date onward,defense expenditure has always represented a major portion of central government expendi- ture If we remove these war-related and military expenditures from con sideration,we find that the gradual decline in the share of total expendi- tures after 1840 is not reflected in the residual group.Thus,total expendi- tures less national debt were 6 per cent of GNP in 1841,and had risen to 7 per ent in 1890.If def ens s also excluded the residual ys in mporta from 3 pe cent o f GNP in 1841 to 5 per cer nt in 1890.Tha is,nonmilitary expenditures shared in the displacement that appears to ee0-1090 38

GROWTH IN EXPENDITURE AND NATIONAL INCOME portion of total government expenditure to GNP was around 11 per cent before the Napoleonic Wars. After those wars, a higher share persisted for some time, but the old proportion had become established again by the decade 1830—40. Such a return was not unexpected, given the prevailing attitude to public expenditures, and given the fact that the governments of the time were aware of the potential displacement effect of wars and were actively concerned to prevent a permanent upward shift in the share of government as a consequence of such wars. From 1841 until 1890 the share of government changes little, declining very slowly from 11 per cent of GNP in 1841 to 9 per cent in 1890. This compares with a share of 12 per cent in 1905, 24 per cent in 1923 and 37 per cent in 1955 (Appendix Table A-6). The gradual decline over the second half of the nineteenth century is the more interesting for being associated after 1875 with a declining price trend which (given the interest of the Chancellors of the day in money expenditures) might have been expected to encourage an opposite movement. At the end of the Napoleonic Era the three largest items of expenditure were central government expenditure on debt interest, military expenditure, and local government expenditure on poor relief. The size of debt interest was being restricted, if not actively reduced, from that time until the beginning of the South African War, and was becoming a smaller proportion of national product in consequence. Thus, debt interest at current prices was £32 million in 1822, £26 million in 1860, and £20 million in 1900. Until the Crimean War, debt interest was almost a half of all government expenditure, in 1880 it was only a quarter and in 1890 a fifth. A hundred years after the Crimean War, as we shall see, debt expenditure represented as little as 11 per cent of total government expenditure, but it had been up to nearly 30 per cent between the two wars.5 The costs of defense move in an opposite direction from that of debt interest payments. Drastically reduced after the Napoleonic Era, they gradually rose after the Crimean War until they overtook national debt interest in the 1880's. From that date onward, defense expenditure has always represented a major portion of central government expenditure. If we remove these war-related and military expenditures from consideration, we find that the gradual decline in the share of total expenditures after 1840 is not reflected in the residual group. Thus, total expenditures less national debt were 6 per cent of GNP in 1841, and had risen to 7 per cent in 1890. If defense is also excluded, the residual grows in importance from 3 per cent of GNP in 1841 to 5 per cent in 1890. That is, nonmilitary expenditures shared in the displacement that appears to 5 The 1800—1890 debt statistics iziclude intragovernrnental debt payments, and so slightly exaggerate the significance of the debt burden (see first section of Chapter 4). 38

growth in ExPENDITURE AND NATIONAL INCOME have occurred over the period of the Napoleonic Wars,and also increased their share of GNP during the long period of relative stability thereafter Turning now to responsibility for expenditures,we find that perhaps the most striking development of the period was in the size and character of local government expenditure.At the beginning of the nineteenth century,the major function of local government was the organization and provision of poor relief.By 1890,expenditure had increased to five es the 1820 level as mor and m Mrs Dashwood in Jane Austen's Sense and Sensibilily (published in 1797)sent her furniture by sea from Sussex to Exeter.road haulage being costly and dangerous.One hundred years later the position was quite different; the Turnpike Trusts had been wound up,a much improved road system had bee create and such a mode of transp would have b ought i sual.In the process,County Counci 6 come th main road builders.After the cholera epidemics in 1831 and 1848 much greater attention was paid to public health,and the Public Health Acts of the sixties and seventies fostered local authority enterprise in housing and sanitation.While comprehensive elementary education had to wait until the founding of the Bo ard of Edu ation in 1899,the local authorities had earlie to receive grants from th central government to encour age individual schools and were authorized to raise rates not exceeding one penny in the pound to finance technical education and manual instruction.In sum,poor relief,which had been by far the greatest single item of expenditure after the Napoleonic Era,was only about 12 per cent of local exp re in 1890, change had taker e b growth in the other functions way of a of local government rather than by a decline in the importance of relief functions. One final point needs to be made about these statistics,which we shall find of interest when we come to discuss the later period in Chapter 5.A remarkably large of total nditure during this repre nted parr g to privat as nationa debt interest,poor relief)and to private institutions (such as grants to private schools).National debt interest payments were much the most important of these.The direct share of government in the national product,therefore,was considerably smaller than govement expendi pe age of r nal income. It is not possible to give very detailed curate statistics of government reso urce use,bu we can say that in 1822 it was something like 8 per cent,in 1850 about 6 per cent,in 1870 about 5.5 per cent and in 1890 about 7 per cent of national product. It is usually assumed that rapid industrialization must bring with it a considerable expansion in"s cial overhead capital.Certainly,capital expenditure by local government on roads and other economic services

GROWTH IN EXPENDITURE AND NATiONAL INCOME have occurred over the period of the Napoleonic Wars, and also increased their share of GNP during the long period of relative stability thereafter. Turning now to responsibility for expenditures, we find that perhaps the most striking development of the period was in the size and character of local government expenditure. At the beginning of the nineteenth century, the major function of local government was the organization and provision of poor relief. By 1890, expenditure had increased to five times the 1820 level, as more and more services were undertaken. Mrs. Dashwood in Jane Austen's Sense and Sensibility (published in 1797) sent her furniture by sea from Sussex to Exeter, road haulage being costly and dangerous. One hundred years later the position was quite different; the Turnpike Trusts had been wound up, a much improved road system had been created, and such a mode of transportation would have been thought must unusual. In the process, County Councils had become the main road builders. After the cholera epidemics in 1831 and 1848 much greater attention was paid to public health, and the Public Health Acts of the sixties and seventies fostered local authority enterprise in housing and sanitation. While comprehensive elementary education had to wait until the founding of the Board of Education in 1899, the local authorities had earlier begun to receive grants from the central government to encourage individual schools and were authorized to raise rates not exceeding one penny in the pound to finance technical education and manual instruction. In sum, poor relief, which had been by far the greatest single item of expenditure after the Napoleonic Era, was only about 12 per cent of local expenditure in 1890, and the change had taken place by way of a growth in the other functions of local government rather than by a decline in the importance of relief functions. One final point needs to be made about these statistics, which we shall find of interest when we come to discuss the later period in Chapter 5. A remarkably large part of total government expenditure during this period represented transfers to private individuals (such as national debt interest, poor relief) and to private institutions (such as grants to private schools). National debt interest payments were much the most important of these. The direct share of government in the national product, therefore, was considerably smaller than government expenditure expressed as a percentage of national income. It is not possible to give very detailed or accurate statistics of government resource use, but we can say that in 1822 it was something like 8 per cent, in 1850 about 6 per cent, in 1870 about 5.5 per cent and in 1890 about 7 per cent of national product. It is usually assumed that rapid industrialization must bring with it a considerable expansion in "social overhead capital." Certainly, capital expenditure by local government on roads and other economic services 39

GROWTH IN EXPENDITURE AND NATIONAL INCOME began to increase at a rate faster than the rate of growth of real national ond half of the ninete nth century,but the absolute prdcirilremarkablyall and w have already seen)real government expenditure as a whole did not constitute an increasing share of national income over the same period.Neverthe- less,there is reason to believe that,before the period of this study,the s ds ofinc ing govern nent eco mic activity had already been so The old liberalism was giving way to the new.In a publi c speech in September 1885,Joseph Chamberlain said,"The greater part of municipal work is Socialism,and every kindly act of legislation by which the community has sought to discharge its responsibilities and its obligations to the poor is Socialis sm,but it is e the worse for that."6A younge contemporary was to add to this the Wagnerian argument: "The wh tendency of civilization is,however,towards the multiplication of the collective functions of society.The ever-growing complications of civil- ization create for us new services which have to be undertaken by the state, n。 ate for us a of the existing services."Fifty yea later the sel same speaker presided over a government responsibl directly or indirectly for spending a sum equal to no less than 37 per cent of the peacetime national income. Government Expenditure and Gross National Product Since 18go We now examine the annual statistics of total government expenditure since 1890,its secula r growth secular changes in the proportion government expenditure to GNP,and the shorter-term changes in expenditure that make up the time pattern of growth and (perhaps) show a displacement effect in some periods that encourages further study We also consider how these things are affected by changes in populatior ces,and the level of e t The analysis leads o th nclusion cor rable e growth in goverment expenditurei real terms per head of population,that the rate of growth over the period as a whole was considerably faster than the rate of growth of gross national product in real terms per head of population,and (what is in some ways more important for our general thesis)that when we have taken account of population growth,price change and changes in the level of emp ment,we are left with an important ph omenon to explain time pattern of expenditure growth.The examination and explanation of this displacement effect provides the theme of later chapters. ullock and Shock,cP.210. 40

GROWTH IX EXPEXDITURE AXD NATIOXAL INCOME began to increase at a rate faster than the rate of growth of real national product in the second half of the nineteenth century, but the absolute size of such expenditure was still remarkably small, and (as we have already seen) real government expenditure as a whole did not constitute an increasing share of national income over the same period. Nevertheless, there is reason to believe that, before the period of this study, the seeds of increasing government economic activity had already been sown. The old liberalism was giving way to the new. In a public speech in September 1885,Joseph Chamberlain said, "The greater part of municipal work is Socialism, and every kindly act of legislation by which the community has sought to discharge its responsibilities and its obligations to the poor is Socialism, but it is none the worse for that."6 A younger contemporary was to add to this the Wagnerian argument: "The whole tendency of civilization is, however, towards the multiplication of the collective functions of society. The ever-growing complications of civilization create for us new services which have to be undertaken by the state, and create for us an expansion of the existing services."7 Fifty years later the self-same speaker presided over a government responsible directly or indirectly for spending a sum equal to no less than 37 per cent of the peacetime national income. Government Expenditure and Gross Yational Product Since 1890 We now examine the annual statistics of total government expenditure since 1890, its secular growth, secular changes in the proportion of government expenditure to GNP, and the shorter-term changes in expenditure that make up the time pattern of growth and (perhaps) show a displacement effect in some periods that encourages further study. We also consider how these things are affected by changes in population, prices, and the level of employment. The analysis leads to the conclusions that there has been a considerable growth in government expenditure in real terms per head of population, that the rate of growth over the period as a whole was considerably faster than the rate of growth of gross national product in real terms per head of population, and (what is in some ways more important for our general thesis) that when we have taken account of population growth, price changes and changes in the level of employment, we are left with an important phenomenon to explain—the irregular time pattern of expenditure growth. The examination and explanation of this displacement effect provides the theme of later chapters. 6 Cited in A. Bullock and N. Shock, eds., The Liberal Tradition from Fox to Keynes, London, 1956, p. 207. Extract from a speech delivered in Glasgow in 1906 by Winston S. Churchill, subsequently republished in his book Liberalism and the Social Problem, London, 1909. See Bullock and Shock, eds., cit., p. 210. 40

GROWTH IN EXPENDITURE AND NATIONAL INCOME GOVERNMENT EXPENDITURE AND GROSS NATIONAL PRODUCT AT CURRENT PRICE Table 2 and Charts 1 and 2,derived from Appendix Tables A-2,A-5 and A-6,demonstrate both the growth in government money expendi tures and the characteristic time pattern of that growth. The great absolute increase in the size of money expenditures during the period is striking.From 130.6 million in 1890,spending rose to ,592.1 million in 1920 and to more million n5.The lex of total e enditur 1900 as bas year,was 47 in 1890,567 in 1920,and 2,188 in 1955 (cf.Chart 2).The change in gross nationa product,however,was not nearly so rapid.GNP rose from C1,472 million in 1890 to C6,070 million in 1920 and then to over 16,700 million in 1955.Taking 1900 as base year,the index of GNP at current prices rose from 76 in 1890 to 312 in 1920 and to 863 in 1955.Expressed in ar the way the proportion o total government expenditure to gros national product at current prices rose from around 9 per cent in 1890 to 26 per cent in 1920 and to 37 per cent in 1955.This increasing propor- tion appears in Chart I as a narrowing of the gap between the GNP and government expenditure curves. The rgrowth did not take place inaregular fashion through time;thereis no growth trend of governn ent money exper ure refect. ing the year-by-year changes in national income.Instead,the curve of government expenditures has the outline earlier referred to-plateaus of ascending height separated by expenditure peaks.The pattern shows very clearly in the two charts.The expenditure peaks coincide with years of war (1900,1918,1943,1952),and,as we should exp ct.the selves and the th. ht of th eaus are ess ob in the case of the or War-192)and the Korcan W ar(195l-52) than in the case of the two world wars (1914-19 and 1939-45).This obviously refects the much greater disruption of the life of the com- munity during the world wars.Also the vears 1900 and 1952 are a little too ear the beginni ng and the end of ou riod for satisfactory inter pretation,ar in any case the statistics for the Boer War peric were collected at five-year intervals only.8 There are consequently strong arguments for concentrating our subsequent analysis of the displacement effect of war upon the periods of the two world wars However.we have not vet demonstrated that there is a phenomenon of displacement of sufficient practical interest tom rit detailed e we must first see how the pattern of money expenditures has been affected 8 It appears from these five-yearly figures,howe r,that the South African War did e5 rnment expenditures,and other historical evidence 花

CR0 WTH LV EXPENDITURE AND NATIONAL INCOME GOVERNMENT EXPENDITURE AND GROSS NATIONAL PRODUCT AT CURRENT PRICES Table 2 and Charts 1 and 2, derived from Appendix Tables A-2, A-5, and A-6, demonstrate both the growth in government money expenditures and the characteristic time pattern of that growth. The great absolute increase in the size of money expenditures during the period is striking. From £130.6 million in 1890, spending rose to £1,592.1 million in 1920 and to more than £6,000 million in 1955. The index of total expenditure, taking 1900 as base year, was 47 in 1890, 567 in 1920, and 2,188 in 1955 (cf. Chart 2). The change in gross national product, however, was not nearly so rapid. GNP rose from £1,472 million in 1890 to £6,070 million in 1920 and then to over £16,700 million in 1955. Taking 1900 as base year, the index of GNP at current prices rose from 76 in 1890 to 312 in 1920 and to 863 in 1955. Expressed in another way, the proportion of total government expenditure to gross national product at current prices rose from around 9 per cent in 1890 to 26 per cent in 1920 and to 37 per cent in 1955. This increasing proportion appears in Chart 1 as a narrowing of the gap between the GNP and government expenditure curves. The secular growth did not take place in a regular fashion through time; there is no growth trend of government money expenditure reflecting the year-by-year changes in national income. Instead, the curve of government expenditures has the outline earlier referred to—plateaus of ascending height separated by expenditure peaks. The pattern shows very clearly in the two charts. The expenditure peaks coincide with years of war (1900, 1918, 1943, 1952), and, as we should expect, the peaks themselves and the change in the height of the plateaus are less obtrusive in the case of the Boer War (1899—1902) and the Korean War (1951—52) than in the case of the two world wars (1914—19 and 1939—45). This obviously reflects the much greater disruption of the life of the community during the world wars. Also the years 1900 and 1952 are a little too near the beginning and the end of our period for satisfactory interpretation, and in any case the statistics for the Boer War period were collected at five-year intervals only.8 There are consequently strong arguments for concentrating our subsequent analysis of the displacement effect of war upon the periods of the two world wars. However, we have not yet demonstrated that there is a phenomenon of displacement of sufficient practical interest to merit detailed examination; we must first see how the pattern of money expenditures has been affected 8 It appears from these live-yearly figures, however, that the South African War did produce some displacement of government expenditures, and other historical evidence tends to confirm this (see Chapter 5). 41