Hostname: page-component-cd9895bd7-mkpzs Total loading time: 0 Render date: 2024-12-25T20:19:14.709Z Has data issue: false hasContentIssue false
  • English
  • Français

Risk-Sharing and Social Justice: The Motivational Foundations of the Post-War Welfare State

Published online by Cambridge University Press:  27 January 2009

John Dryzek  and
Robert E. Goodin
Get access Rights & Permissions [Opens in a new window]

Extract

Much political behaviour can be interpreted as the pursuit of more or less naked self-interest. Occasionally, though, individuals do apparently exhibit some concern for their fellow human beings. The result is a less cold and dismal world – and one in which moral philosophers can find a role. Our focus here, however, is more on practical problems than philosophical ones. We shall be less concerned with questions of what moralists should demand of people than with questions of how such demands could be enforced upon people. Specifically, we will be asking how it is possible to evoke from people support for policies aiding those less fortunate than themselves. We propose to address this question by exploring the sources of support for the most broadly-based institution presently available for promoting social justice, the welfare state.

Type
Articles
Information
British Journal of Political Science , Volume 16 , Issue 1 , January 1986 , pp. 1 - 34
Copyright
Copyright © Cambridge University Press 1986

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 This is not to imply that the welfare state is necessarily a redistributive success; cf. Grand, Julian Le, The Strategy of Equality (London: Allen & Unwin, 1982)Google Scholar. We only mean to imply that public support for it is predicated on the presumption that it is or could be made one. Neither should ethical concern necessarily stop at national borders. But in order to see what it might take to get a world-wide welfare state, we need to see how we have managed to secure even a nationwide one.

2 For arguments that earlier developments might be explained along lines similar to those we here propose, see Beales, H. L., The Making of Social Policy, L. T. Hobhouse Memorial Trust Lecture No. 15 (London: Oxford University Press, 1946), pp. 6, 14Google Scholar; and Goldthorpe, John H., ‘The Development of Social Policy in England, 1800–1914’, Transactions of the Fifth World Congress of Sociology, Washington, D. C., September 1962 (Paris: International Sociological Association, 1964), Vol. 4, pp. 41–56 at pp. 48 ff.Google Scholar

3 Wildavsky, Aaron, ‘A Cultural Theory of Governmental Growth and (Un)balanced Budgets’, paper presented to the Nobel Symposium on the Growth of Government, Stockholm, Sweden, 08 1984Google Scholar; Wilensky, Harold L., The Welfare State and Equality (Berkeley: University of California Press, 1975)Google Scholar; Flora, Peter and Heidenheimer, Arnold J., eds, The Development of Welfare States in Europe and America (New Brunswick, N.J.: Transaction Books, 1981).Google Scholar

4 For an explicit justification of such a focus, see Wilensky, , Welfare State and Equality, Chap, 1Google Scholar. A similar focus is found in Rimlinger, Gaston V., Welfare Policy and Industrialization in Europe, America and Russia (New York: Wiley, 1971)Google Scholar and Jackman, Robert W., Politics and Social Equality (New York: Wiley, 1975).Google Scholar

5 Hume, David, A Treatise of Human Nature (London: J. Noon, 1739)Google Scholar, Bk. 3 and An Enquiry Concerning the Principles of Morals (London: T. Cadell, 1777)Google Scholar; Smith, Adam, The Theory of the Moral Sentiments, 6th edn (London: A. Strahan and Cadell, T., 1790), Pt. 1, Sec. 1, Chap. 5, Para. 5.Google Scholar

6 Hochman, Harold M. and Rogers, John D., ‘Pareto Optimal Redistribution’, American Economic Review, LIX (1969), 542–57Google Scholar; Dacy, D. C. and Kunreuther, Howard, The Economics of Natural Disasters (New York: Free Press, 1969)Google Scholar; De Alessi, Louis; ‘A Utility Analysis of Post-Disaster Cooperation’, Papers in Non-Market Decision Making, III (1967), 85–90Google Scholar; De Alessi, , ‘The Utility of Disasters’, Kyklos, XXI (1968), 525–32CrossRefGoogle Scholar; De Alessi, , ‘Toward an Analysis of Postdisaster Cooperation’, American Economic Review, LXV (1975), 127–38Google Scholar. See more generally Baker, John, ‘Social Conscience and Social Policy’, Journal of Social Policy, VIII (1979), 177–206CrossRefGoogle Scholar; Thomas, and Wilson, Dorothy, The Political Economy of the Welfare State (London: Allen & Unwin, 1982)Google Scholar; and Margolis, Howard, Selfishness, Altruism and Rationality (Cambridge: Cambridge University Press, 1982).Google Scholar

7 Barry, Brian, ‘Justice Between Generations’ in Hacker, P. M. S. and Raz, J., eds, Law, Morality and Society (Oxford: Clarendon Press, 1977), pp. 268–84 at p. 279.Google Scholar

8 Becker, Gary S., A Treatise on the Family (Cambridge, Mass.: Harvard University Press, 1982).Google Scholar

9 Overvold, Mark Carl, ‘Self-interest and the Concept of Self-sacrifice’, Canadian Journal of Philosophy, X (1980), 105–18CrossRefGoogle Scholar; Batson, C. Daniel and Coke, Jay S., ‘Empathy: A Source of Altruistic Motivation for Helping?’ in Rushton, J. P. and Sorrentino, R. M., eds, Altruism and Helping Behavior (Hillsdale, N.J.: Lawrence Erlbaum Associates, 1981), pp. 167–87Google Scholar. All this bears out the argument, dating back at least to Butler, Bishop Joseph's Fifteen Sermons at the Rolls Chapel (London: Knapton, 1726)Google Scholar, esp. Sermon 9, that the altruist's object is the other's happiness itself and not just the satisfaction of the altruist's own desire for the other's happiness.

10 Thompson, William, Appeal of One Half of the Human Race, Women, Against the Pretensions of the Other Half, Men (London: Longman, 1825), p. 19.Google Scholar

11 Sugden, Robert, ‘On the Economics of Philanthropy’, Economic Journal, XCII (1982), 341–50.CrossRefGoogle Scholar

12 Harsanyi, John C., ‘Cardinal Utility in Welfare Economics and in the Theory of Risk-Taking’, Journal of Political Economy, LXI (1953), 434–5CrossRefGoogle Scholar. See similarly Harsanyi, , ‘Morality and the Theory of Rational Behavior’Google Scholar, in Sen, Amartya and Williams, Bernard, eds, Utilitarianism and Beyond (Cambridge: Cambridge University Press, 1982), pp. 39–62 at pp. 44–8.CrossRefGoogle Scholar

13 Harsanyi, , ‘Morality and the Theory of Rational Behavior’, pp. 44, 47Google Scholar. These passages echo earlier ones in ‘Cardinal Utility in Welfare Economics’, p. 435Google Scholar and ‘Cardinal Welfare, Individualistic Ethics and Interpersonal Comparisons of Utility’, Journal of Political Economy, LXIII (1955), 309–21 at p. 315 (italics in the original).Google Scholar

14 Goodin, R. E. and Roberts, K. W. S., ‘The Ethical Voter’, American Political Science Review, LXIX (1975), 926–28CrossRefGoogle Scholar; Benn, Stanley I., ‘The Problematic Rationality of Political Participation’, in Laslett, P. and Fishkin, J., eds, Philosophy, Politics and Society, 5th series (Oxford: Blackwell, 1979), pp. 291–312.Google Scholar

15 See, e.g., Robson, William, Welfare State and Welfare Society (London: Allen & Unwin, 1976).Google Scholar

16 Rawls, John, A Theory of Justice (Cambridge, Mass.: Harvard University Press, 1971), Sec. 86.Google Scholar

17 Paul Samuelson, for example, muses that ‘an essay could be written on the welfare state as a complicated device for self- or re-insurance’ in his paper on ‘An Exact Consumption-Loan Model of Interest with or without the Social Contrivance of Money’, Journal of Political Economy, LXVI (1958), 467–72 at p. 480Google Scholar. William Baumol similarly suggests that ‘a rule of fairness is a sort of insurance arrangement which selfish people accept to make sure they will not be mistreated, and pay for it by providing assurance to others that they, too, will not be mistreated’ in his paper on ‘Applied Fairness Theory and Rationing Policy’, American Economic Review, LXXII (1982), 639–51 at p. 640Google Scholar. See similarly Buchanan, James M. and Tullock, Gordon, The Calculus of Consent (Ann Arbor: University of Michigan Press, 1962), p. 193CrossRefGoogle Scholar and Goodin, Robert E., The Politics of Rational Man (London: Wiley, 1976), pp. 78–80, 115Google Scholar. Notice that our analysis is linked to the classic welfare-economic rationale for redistribution, insofar as people are willing to buy insurance if and only if income has diminishing marginal utility for them; see Olson, Mancur, ‘A Less Ideological Way of Deciding How Much Should be Given to the Poor’, Daedalus, CXII, (1983), 217–36.Google Scholar

18 ‘Environed as we are by risks and perils, which befall us as misfortunes, no man of us is in a position to say, “I … am sure … I shall never need aid and sympathy”. At the very best, one of us fails in one way and another in another, if we do not fail altogether. Therefore the man under the [falling] tree is the one of us who for the moment is smitten. It may be you tomorrow, and I the next day. It is the common frailty in the midst of a common peril which gives us a kind of solidarity of interest to rescue the one for whom the chances of life have turned out badly just now’. Sumner, William Graham, What Social Classes Owe Each Other (New York: Harper, 1883), p. 158.Google Scholar

19 Zeckhauser, Richard, ‘Risk Spreading and Distribution’, in Hochman, H. M. and Peterson, G. E., eds. Redistribution Through Public Choice (New York: Columbia University Press, 1974), pp. 206–28.Google Scholar

20 These differ from privately-organized insurance schemes in certain respects. Premiums, for example, are not calculated on an ‘individualized actuarial model’, as private underwriters would have attempted; and, of course, participation in these public schemes tends to be compulsory. All this is explicable in terms of government insuring against various particularly nasty contingencies, however; see Goodin, , Politics of Rational Man, p. 112Google Scholar. In other respects, however, these initial programmes were much more like insurance schemes than many that followed (family allowances, AFDC, etc.). The latter programmes tended to be available to anyone in need, and were paid for out of general-fund revenue generated by ordinary taxes. The insurance-like schemes were organized around separate trust funds, fed by premium-like taxes that were collected separately and earmarked for deposit to those trust funds only. The criterion of membership in the scheme, and hence of eligibility for benefits, was payment (by oneself or on one's behalf) of those special taxes-cum-premiums. See Titmuss, Richard M., ‘Models of Redistribution in Social Security and Private Insurance’, Commitment to Welfare (London: Allen & Unwin, 1968), pp. 173–87Google Scholar and Social Policy (London: Allen & Unwin, 1974), Chap. 7Google Scholar; and Burns, Eveline M., ‘Social Insurance in Evolution’, American Economic Review (Papers & Proceedings), XXXIV (1944), 199–211Google Scholar and ‘The Revolution of Social Insurance’, Social Service Review, IXL (1965), 129–40.Google Scholar

21 SirHelps, Arthur, The Claims of Labour: An Essay on the Duties of the Employers to the Employed, 2nd edn (London: William Pickering, 1845), p. 14.Google Scholar

22 As the eminent turn-of-the-century British campaigner for old-age pensions, Charles Booth, observes, ‘If the chances of poverty were equal, none could say that they lost while another gained by paying for a general scheme of pensions. They are not equal, “One of the few lessons” (says Mr Leslie Stephen) “which I have learnt from life, and not found already in copy books, is the enormous difficulty which a man of the upper classes finds in completely ruining himself even by vice, extravagance, and folly; whereas there are plenty of honest people who in spite of economy and prudence can scarcely keep out of the workhouse’; Booth, , Pauperism and the Endowment of Old Age (London: Macmillan, 1892), p. 201Google Scholar. Philosopher Gilbert Harman says more pointedly, ‘Since the rich and the strong could foresee that they would be required to do most of the helping and that they would receive little in return, they would be reluctant to agree to a strong principle of mutual aid’; Harman, , ‘Moral Relativism Defended’, Philosophical Review, LXXXIV (1975), 3–22 at pp. 12–13CrossRefGoogle Scholar. See similarly Benn, Stanley I., ‘The Rationality of Political Man’, American Journal of Sociology, LXXXIII (1978), 1271–6CrossRefGoogle Scholar. Zeckhauser, (‘Risk Spreading and Distribution’, pp. 215–17)Google Scholar likewise observes that in many cases ‘risk-spreading contracts are drawn up after some lotteries are [already] run’, which explains why risk-spreading arrangements fail to produce completely impartial, egalitarian outcomes.

23 Insurance companies have a good idea, too, and choose whether or not to insure individuals and what premiums to charge them on the basis of actuarial risks. When individuals have better information, they can outwit the actuaries. Then there will be a process of ‘adverse selection’, with good risks opting out, leaving only high-risk individuals in the insurance pool. Adverse selection acts as an impediment to the development of markets in insurance, which further inhibits effective redistribution through insurance.

24 Some can be explained as attempts to increase economic productivity. Turn-of-the-century Antipodean social reforms, the original British scheme for labour exchanges and unemployment insurance, and the socialization of production in Britain during the First World War are all examples of that. See Briggs, Asa, ‘The Welfare State in Historical Perspective’, Archives Européennes de Sociologie, II (1961), 221–58, esp. pp. 243–5CrossRefGoogle Scholar; Rimlinger, Gaston V., ‘Welfare Policy and Economic Development: A Comparative Historical Perspective’, Journal of Economic History, XXVI (1966), 556–71CrossRefGoogle Scholar; Beveridge, William H., ‘Labour Exchanges and the Unemployed’, Economic Journal, XVII (1907), 66–81CrossRefGoogle Scholar; and Tawney, R. H., ‘The Abolition of Economic Controls’, Economic History Review, XIII (1943), 1–30 at pp. 2–7.CrossRefGoogle Scholar Other welfare-state innovations can best be understood in the context of war preparations, as attempts to guarantee the quantity and quality of cannon fodder, either in this generation or the next; see, e.g., Titmuss, Richard M., ‘War and Social Policy’, Essays in ‘the Welfare State’ (London: Allen & Unwin, 1958), pp. 75–88 at p. 81.Google Scholar

25 Another factor, which fits equally well with our model (and which may in practice be hard to separate from the influence of the Second World War) is the pervasive uncertainty induced by the Great Depression. See Heclo, Hugh, ‘Toward a New Welfare State?’Google Scholar in Flora, and Heidenheimer, , eds, The Development of Welfare States in Europe and America, pp. 383–406, esp. 392.Google Scholar

26 Titmuss, Richard M., Problems of Social Policy (London: HMSO and Longman, Green, 1950), p. 506.Google Scholar

27 Indeed, according to Gowing, Margaret's ‘Obituary Notice’, Proceedings of the British Academy, LI (1975), 401–28 at p. 406Google Scholar, Titmuss's own London home was bombed twice during the war.

28 Titmuss, , Problems of Social Policy, pp. 506–7.Google Scholar

29 Titmuss, , ‘War and Social Policy’, p. 85.Google ScholarDouty, Christopher M., ‘Disasters and Charity: Some Aspects of Cooperative Economic Behaviour’, American Economic Review, LXII (1972), 580–90 at pp. 586–7Google Scholar offers a similar analysis of disaster relief more generally; but cf. De Alessi, , ‘Toward an Analysis of Postdisaster Cooperation’.Google Scholar

30 This approximates that behaviour which Abba Lerner believes welfare economics should always commend to governments; see his Economics of Control (New York: Macmillan, 1944).Google Scholar Of course, wartime governments also had various other less high-minded reasons for introducing insurance schemes during the war. One was to guarantee the stability of economic and financial institutions predicated on the assumption that real estate would (at least on average) retain its value. Another was that insurance premiums were a convenient way of soaking up money out of the private sector and dampening wartime inflation. Both points were made explicit in the Chancellor's speech introducing the British War Damage Insurance scheme and in the debate that followed; see Hansard's Parliamentary Debates (Commons), 5th series, vol. 367 (1940–1941)Google Scholar, cols. 1125–90 and 1243–99.The Australian Treasurer also explicitly evoked the latter consideration in proposing the introduction of unemployment insurance during the war, when of course most people would be employed and hence income into the fund could be counted upon vastly to exceed outflow; see Watts, Rob, ‘The Origins of the Australian Welfare State’, Historical Studies, XIX (1980), 175–98 at p. 190.CrossRefGoogle Scholar

31 Briggs, , ‘The Welfare State in Historical Perspective’, p. 221.Google Scholar

32 See, e.g., Furniss, Norman and Tilton, Timothy, The Case for the Welfare State (Bloomington: Indiana University Press, 1977)Google Scholar and Milton, and Friedman, Rose, Free to Choose (New York: Avon, 1981), pp. 82–118.Google Scholar

33 See, e.g., Williams, Gertrude, The Coming of the Welfare State (London: Allen & Unwin, 1967).Google Scholar

34 Addison, Paul, The Road to 1945 (London: Cape, 1975), p. 14Google Scholar, argues that ‘the Coalition proved to be the greatest reforming administration since the Liberal government of 1904–14’. Much the same is true of Australia; see Butlin, S. J. and Schedvin, C. B., War Economy 1942–45, Australia in the War of 1939–45, Series 4 (Civil), Volume 4 (Canberra: Australian War Memorial, 1977), Chap. 21.Google Scholar

35 Manchester Guardian Weekly, 12 12 1982.Google Scholar

36 SirBeveridge, William Henry, Social Insurance and Allied Services, Cmd. 6404 (London: HMSO, 1942).Google Scholar

37 See Titmuss, , Problems of Social PolicyGoogle Scholar, passim. Some of these innovations – such as the War Damages Act – could be interpreted as simply part of the war effort. Others – such as pensions and school meals – had no more than a tangential connection with the actual prosecution of the war.

38 Peacock, Alan T. and Wiseman, Jack, The Growth of Public Expenditure in the United Kingdom (Princeton, N.J.: Princeton University Press, 1961), p. 27.Google Scholar

39 This in turn arises from the ‘inspection effect’, which derives (in the first instance, anyway) from a desire to assure the quality of cannon fodder, as the preface to the revised 1969 edition of Peacock, and Wiseman, , Growth of Public ExpenditureGoogle Scholar, makes clear: ‘In wartime the tally taken of human resources for special purposes may incidentally reveal information about social, etc. conditions which was not previously available, and this new knowledge may produce a consensus of opinion in favour of new and larger public expenditures of particular kinds after the “return to normalcy”’. Hence, their explanation of the ‘inspection effect’ would seem to have more in common with the familiar forces discussed in footnote 24 than with the wartime uncertainty which is our focus here.

40 ‘Revenue determines expenditure’ is the conclusion of Alan Peacock's lecture on ‘The Welfare Society’, Unservile State Papers, no. 2 (1960). p. 14.Google Scholar See similarly Peacock, and Wiseman, , ‘Approaches to the Analysis of Government Expenditure Growth’, Public Finance Quarterly, VII (1979), 3–23 at pp. 14–17.CrossRefGoogle Scholar

41 Politicians and civil servants might, right from the beginning of the war, divert some of the revenues raised for fighting the war into their long-planned domestic projects. But this would be no more than a marginal phenomenon. For a vast majority of politicians and civil servants, winning the war would almost certainly have taken precedence over all other goals; and they would not care to endanger that outcome by more than marginally diverting revenues away from the war effort.

42 Peacock, and Wiseman, , Growth of Public Expenditure, pp. 62, 27.Google Scholar

43 Interpretation of the timing of changes in observed social services expenditure can be hazardous: delayed implementation can make effects lag far behind their causes; and the effects of anticipation can even make effects predate their causes, as Titmuss, , Problems of Social Policy, p. 54Google Scholar finds with the hospital services. Peacock and Wiseman's exhaustive study of historical patterns of The Growth of Public Expenditure in the United Kingdom sheds little light on the timing of wartime increases. Their statistical time series has large gaps for the war and its aftermath: it shows that social welfare expenditure is much higher in 1950 than in 1938, but the intervening years are a blank. These gaps might have been what led Peacock and Wiseman to construct an explanatory account of the war-related spending increase which treats the war and its aftermath as a single unit.

44 Titmuss, , Problems of Social Policy, p. 15.Google Scholar

45 Indeed, payments peak twice, first in 1942 and then not again until 1947, well after the war's conclusion. This is partly due to the ordinary time lags in reaching settlements and disbursing funds to claimants, and partly due to the fact that government intentionally held up payments to prevent labour and material needed in the war effort from being diverted to premature civilian rebuilding projects. Indeed, at the outset of the war the government proposed waiting until the war was over before making any payments, although it relented somewhat starting in 1940; see Titmuss, , Problems of Social Policy, pp. 282–3Google Scholar and ‘War Damage Balance Sheet’, The Economist. CIL (15 12 1945), 883.Google Scholar

46 These premiums reflected, at least in the first instance, government rather than public perceptions of the level of risks: premiums were fixed by the government, and participation in the schemes was for the most part compulsory (a voluntary scheme being run only for private chattels and movable property, such as household furniture). But there is no particular reason to suppose that there was any very great divergence between the government's perception of the risk and the public's. On the contrary, it seems entirely likely that the public based its perceptions on cues (such as War Risk and Damage premiums) given by the government itself.

47 Notice, also, that a wartime threat to capital gave rise to post-war protection against threats to labour. Specifically, those wartime schemes offering primarily protection against risks to property (the War Risks–War Damage insurance schemes) were displaced, after the war, into social welfare programmes primarily protecting against risks to persons.

48 A continuous series is available for local government expenditures on education, housing and poor relief; see Peacock, and Wiseman, , Growth of Public ExpenditureGoogle Scholar, Table A 22. This is only a small fraction of the total central plus local government social service expenditure, however – 8·1 per cent in 1938, 3·8 per cent in 1950. Hence this series is of little use for our purposes here.

49 Peacock, and Wiseman, , Growth of Public Expenditure.Google Scholar

50 Titmuss, , Problems of Social Policy, p. 55.Google Scholar

51 Peacock, and Wiseman, , Growth of Public Expenditure, pp. iv–xxv.Google Scholar

52 Foster, C. D., Jackman, R. A. and Perlman, M., Local Government Finance in a Unitary State (London: Allen & Unwin, 1980), pp. 80 ff.Google Scholar

53 Fraser, Derek, The Evolution of the British Welfare State (London: Macmillan, 1973), p. 193.CrossRefGoogle Scholar

54 Fraser, , Evolution of the British Welfare State, p. 195.Google Scholar

55 Titmuss, , Problems of Social Policy, p. 506.Google Scholar

56 Addison, , Road to 1945, pp. 271–8.Google Scholar

57 Harris, José, ‘Some Aspects of Social Policy in Britain During the Second World War’, in Mommsen, W. J., ed., The Emergence of the Welfare State in Britain and Germany, 1850–1950 (London: Croom Helm, 1981).Google Scholar

58 Suitably comparable cross-national data for other aspects of total social service spending, such as housing, health and education, were unavailable. We were also unable to obtain suitable data on the extent of programme coverage (i.e., the number of persons or percentage of the workforce covered). Even if the latter data did exist, however, they would constitute an ambiguous indicator, double-counting individuals who are covered by more than one programme; total expenditure level suffers no such ambiguity.

59 To fall under this definition, these schemes must also be either set up by legislation or have a public or semi-public body assuming obligations; the latter kind of body must also administer the scheme. See International Labour Organisation, ‘The Cost of Social Security’, International Labour Review, LIV (1952), 726–91Google Scholar for details.

60 Notice that, when Peacock, and Wiseman, (Growth of Public Expenditure, p. 33)Google Scholar say that ‘wars generate commitments that continue into peace’, it is just this sort of commitment (along with ‘debt commitments’) that they take as their paradigm case. Even Titmuss, who usually emphasizes other deeper connections, remarks similarly at one place; see Titmuss, , ‘War and Social Policy’, p. 80.Google Scholar

61 This is not, in fact, a bad choice of years. The former (1933) represents one of the last years of international quiescence before the gathering storm of war affected countries such as Japan, Spain and Italy. The latter (1949) effectively represents the settled post-war situation. By then, the uncertainty which had carried over the immediate post-war period had, with the onset of reconstruction, largely disappeared.

62 One indirect indicator which we considered and rejected is the amount of American nonmilitary aid per capita in the immediate post-war period. This financial aid (notably the Marshall Plan) was intended to promote reconstruction, and hence one would expect it to vary in proportion to the degree of destruction each country had experienced. Unfortunately for our purposes, political considerations entered too strongly into decisions about both the giving and the accepting of American aid for this to serve as a true indicator of wartime destruction.

63 Thus, uncertainty as indicated by this second variable is zero for non-belligerents, very low for countries (like Denmark) occupied within a few days of the beginning of the war and held by Germany for the duration, and very high for persistent belligerents like the United Kingdom and Japan. This indicator is not free from ambiguity. For example, how should resistance fighting on the national territory and civil conflict be treated? It was decided to exclude resistance fighting in all cases except Yugoslavia, where Partisans and Chetniks controlled large portions of the country during the entire period of notional Axis occupation. Civil conflict was treated as a state of belligerence.

64 ‘Time fighting or fought over’ may be controversial as an indicator of uncertainty, because it implies that there is no uncertainty associated with full occupation by another power. Were we to include ‘time occupied’, however, that would drastically reduce variation in this component of the Index: virtually all countries caught up in the war at all were either fighting, being fought over, or occupied for the duration.

65 Peacock, and Wiseman, , Growth of Public Expenditure, p. 54.Google Scholar

66 Cutright, Phillips, ‘Political Structure, Economic Development and National Social Security Programs’, American Journal of Sociology, LXX (1965), 537–50CrossRefGoogle Scholar; Jackman, , Politics and Social Equality.Google Scholar

67 Lenski, Gerhard, Power and Privilege (New York: McGraw-Hill, 1966)Google Scholar; Cutright, , ‘Political Structure, Economic Development and National Social Security Programs’.Google Scholar

68 Figures for the war years are unavailable, so we use the average of 1937 and 1949 figures for each country. The use of energy consumption as a surrogate for national income in indicating the resource base of a country is a well-established practice in cross-national social and political research; see Taylor, Charles L. and Hudson, Michael C., World Handbook of Political and Social Indicators, 2nd edn (New Haven, Conn.: Yale University Press, 1972), p. 291.Google Scholar Gross national product for 1950 was available for a limited number of countries in our sample; this showed a correlation of 0·84 with wartime energy consumption.

69 Olson, Mancur, The Rise and Decline of Nations (New Haven, Conn.: Yale University Press, 1982).Google Scholar

70 Piven, Frances Fox and Cloward, Richard A., Regulating the Poor (New York: Pantheon, 1971).Google Scholar

71 Stephens, James D., The Transition to Socialism (London: Macmillan, 1979)Google Scholar; cf. Jackman, , Politics and Social Equality, Chap. 6.Google Scholar

72 Beveridge, , Social Insurance and Allied ServicesGoogle Scholar, para. 460. This theme also emerged, for example, in the course of debates on the War Damage Insurance Act. The Chancellor, introducing the measure, called it ‘an instrument of justice and an act of solidarity’. Perhaps more revealing, however, was an interjection during the speech of an MP opposed to the Bill: George Benson, MP, asked (as what he thought to be a rhetorical question), ‘Why is it that the Lake District owner, who is practically safe, is compelled to share the burden of the Coventry owner [who is very much at risk from German bombers]?’; Hansard records as ‘an interjection from an Honourable Member on the opposite benches’ the reply, ‘We are all in the war’. See Hansard's Parliamentary Debates (Commons), 5th series, vol. 367 (1940–1941)Google Scholar, cols. 1137 and 1166.

73 See, e.g., Raiffa, Howard, Decision Analysis (Reading, Mass.: Addison-Wesley, 1968).Google Scholar

74 Uncertainty of a different sort might also explain the widespread tendency to adopt social security systems that shift costs and/or benefits far into the future: uncertainty about who will benefit and who will suffer naturally tends to increase with temporal distance.

75 An explanation in terms of shifting motivations really is required. It is not enough to say, as some economists might, that this is just a case of war precipitating a ‘market failure’ in insurance markets which the government stepped in to remedy. That leaves unexplained some of the most important particulars of the intervention, specifically, why governments failed to apply actuarial principles in setting war damage premiums; see SirWood, Kingsley, Hansard's Parliamentary Debates (Commons), 5th series, vol. 367 (1940–1941)Google Scholar, col. 126 and Hirschleifer, Jack, ‘War Damage Insurance’, Review of Economics and Statistics, XXXV (1953), 144–53.CrossRefGoogle Scholar

76 Tawney, , ‘The Abolition of Economic Controls’, p. 7Google Scholar, writes, ‘The most extensive and intricate scheme of state intervention in economic life which the country had seen was brought into existence, without the merits or demerits of state intervention being ever discussed. … Each addition to the structure was related to some immediate necessity of incontestable urgency. … Once the war was over, what had been a source of strength became a weakness. War collectivism had not been accompanied by any intellectual conversion on the subject of the proper relations between the state and economic life, while it did not last long enough to change social habits. With the passing, therefore, of the crisis that occasioned it, it was exposed to the attack of the same interests and ideas as, but for the war, would have prevented its establishment.’