The disfranchisement of the elderly, and other attempts to secure intergenerational justice

Philosophy and Public Affairs

Volumn 27, Issue 4, 1998, Pages 292-333

  Van Parijs, Philippe ^a  

^a NONE

Author keywords

[No Author keywords available]

Indexed keywords

EID: 4243057653     PISSN: 00483915     EISSN: 10884963     Source Type: Journal    
DOI: 10.1111/j.1088-4963.1998.tb00072.x     Document Type: Article

References (162)

1. Douglas J. Stewart, "Disfranchise the Old," New Republic 29, no. 8 (1970):20-22.
2. Here are just a couple of typical formulations: "In an aging population, the great danger is that the electorate will become more and more focused on the short term, for there will eventually be fewer and fewer voters who are parents of young children and more who are concerned with having the state provide either for their own aged parents or for themselves in retirement" (Phillip Longman, Born to Pay: The New Politics of Aging in America [Boston: Houghton Mifflin, 1987], p. 143).
3. "But the elderly are growing both richer and more numerous, and unless something is done to curb their expanding political power, programs to benefit them may yet become untouchable" (Alexi Bayer, "Let's Give Parents an Extra Right to Vote," New York Times, 4 May 1997).
4. Binstock describes and denounces some aspects of this in the U.S. (Robert Binstock, "Transcending Intergenerational Equity," in Economic Security and Intergenerational Justice: A Look at North America, ed. T. R. Marmor, T. M. Smeeding, and V. L. Greene [Washington, DC: The Urban Institute Press, 1994], pp. 155-85).
5. Several other essays in Marmor, et al., eds. (Economic Security and Intergenerational Justice: A Look at North America, ibid.) reflect on why this issue became more salient in the U.S. than in Canada. The age-inclusiveness of Canada's health-care system (in contrast to the restriction of Medicare to the elderly) and the far greater inequality among the elderly in the U.S. (and hence the conspicuous affluence of some of them) are likely to have played an important role.
6. In doing so, I put myself in good company: "For example, there are sensible proposals for what should be done regarding the alleged coming crisis in Social Security: slow down the growth of benefit levels, gradually raise the retirement age, impose limits on expensive medical care for only a few weeks or days, and finally, raise taxes now, rather than face large increases later. But as things are, those who follow the 'great game of politics' know that none of these sensible proposals will be accepted" (John Rawls, "The Idea of Public Reason Revisited," The University of Chicago Law Review 64 no. 3 [1997]:765-807).
7. The intergenerational dimension I found a particularly interesting aspect of this more general question, but the fact that this article focuses on it should not be taken to imply that I regard other aspects - for example, justice between wealth or skill categories, or between genders, ethnic groups, or regions - any less important. See Philippe Van Parijs, "Justice and Democracy: Are They Incompatible?" Journal of Political Philosophy 4, no. 2 (1996):101-17;
8. Van Parijs, "Should the European Union Become More Democratic?" in Democracy and the European Union, ed. A. Follesdal and P. Koslowski (Berlin and New York: Springer, 1997), pp. 287-301;
9. and Van Parijs, "Power-Sharing versus Border-Crossing in Severely Divided Societies: A Comment on Horowitz," in Designing Democratic Institutions. Nomos 41, ed. I. Shapiro and S. Macedo, forthcoming, for an exploration of some of these other dimensions.
10. Along these lines, see, e.g., John Rawls, A Theory of Justice (Oxford: Oxford University Press, 1972), pp. 284-93;
11. John Hartwick, "Intergenerational Equity and the Investing of Rents from Exhaustible Resources," American Economic Review 66 (1977): 972-74;
12. Brian Barry, "Justice Between Generations," in Barry, Liberty and Justice (Oxford: Oxford University Press, 1989), 242-58;
13. and Philippe Van Parijs, Real Freedom for All: What (if Anything) Can Justify Capitalism? (Oxford: Oxford University Press, 1995), pp. 38-41.
14. The conception of intergenerational justice that follows from this view is significantly different from the more generous "solidaristic" conception of Léon Bourgeois, Solidarité (Paris: Armand Colin, 1902),
15. which requires each generation to improve the situation of the next one just as previous ones contributed to improving its own situation. It is also crucially distinct from the meaner "equal exchange" conception which provides much of so-called "generational accounting" with a simple ethical ideal of equal "benefit ratios" (see, e.g., A. J. Auerbach, J. Gokhale, and L. J. Kotlikoff, "Generational Accounts: A Meaningful Alternative to Deficit Accounting," Tax Policy and the Economy 5 [1991]:55-110,
16. and Lawrence J. Kotlikoff, "Justice and Generational Accounting," in Justice Across Generations: What Does It Mean? ed. L. M. Cohen [Washington, DC: American Association of Retired Persons, 1993], pp. 77-93). But those who are committed to either of these alternative conceptions of intergenerational justice should find the considerations below no less relevant to their concerns.
17. As pointed out by Claus Offe ("Zusatzstimmen für Eltern. Ein Beitrag zur Reform von Demokratie und Wahlrecht?" in Zukunft wählen - Zusatzstimmen für Eltern?, ed. G. Grözinger and H. Geiger [Bad Boll: Evangelische Akademie, 1993], pp. 1-26, 9), in countries in which pension levels are highly sensitive to the completion of a full working career, this phenomenon is further amplified as a result of women giving up the idea of having (more) children because of a cost in pension rights far more than proportional to the immediate loss in earnings.
18. Figures and 1998 estimates for Belgium (assuming an unchanged minimum voting age of 18) have been kindly provided by Paul-Marie Boulanger and André Lambert (ADRASS, Ottignies). The estimates are based on the assumptions of 1.55 children per woman, a net immigration of 0.1 percent per year, and a life expectancy gradually rising to its "natural limit" of 90 years.
19. Bear in mind that, in some European countries, the average effective retirement age for men and women was well below 60 in the 1990s. Moreover, between the time of a particular election and the time at which it has an impact on age-sensitive policies, there may be a considerable time lag. With a median electoral age of 56, an average retirement age of 58, and a 3-year policy lag, an absolute majority of the electorate can expect to be in retirement at the time their votes produce their effects.
20. In Switzerland's 1991 national election, for example, the turnout was 52 percent among the over-65s, while it was only 30 percent among those aged 18 to 23 and 44 percent among those 24 to 39 (Silvano Möckli, "Demokratische Struktur und Volksabstimmungen," Hochschule St-Gallen: Institut für Politikwissenschaft, 1993, p. 3.
21. Also in Hilfe, die Schweiz schrumpft, ed. P. Füglistaler (Verlag Orell Füssli, 1994), pp. 13-25).
22. In the U.S. 1992 national election, the turnout was 70.1 percent among the over-65s, while it was only 38.5 percent among those 18-20 and 45.7 percent among those 21-24 (Matthew C. Price, Justice Between Generations: The Growing Power of the Elderly in America [New York: Praeger, 1997], p. 82).
23. In the U.S., the lobby of the elderly has been a mass-based movement since the mid-1960s (Henry J. Pratt, The Gray Lobby [Chicago & London: The University of Chicago Press, 1976], chap. 4).
24. It is far better organized than the family lobby, for example, despite the fact that there are nearly twice as many households with children as households of retirees (see Frank Levy and Richard J. Murnane, "Orphans of the Ballot Box," The New York Times, 6 June 1992, p. A23).
25. With over 33 million members aged 50 or more, the American Association of Retired Persons (AARP) is the biggest organization in the U.S. apart from the Catholic Church. One out of four registered voters is a member of it, and it can count on the involvement of 350,000 active volunteers (Price, Justice, pp. 88-89).
26. It may be true that, in the U.S. for example, the enactment of the major old-age policies is "attributable for the most part to the initiatives of public officials in the White House, Congress, and the bureaucracy," rather than to the lobby of the elderly (Binstock, "Transcending Intergenerational Equity," p. 165). But by no means does this prevent the aging of the electorate from significantly affecting the content of the platforms that candidates feel they have to put forward or the content of the policies that incumbents feel they can get away with adopting.
27. This was apparently the case through the 1970s and 1980s in the U.S. See Christine L. Day, What Older Americans Think: Interest Groups and Aging Policy (Princeton, NJ: Princeton University Press, 1990), pp. 41-52: "older people are nearly indistinguishable from younger adults (both middle-aged and younger categories) on most issues - including age policy issues."
28. Support for the U.S. social security system (versus private old-age insurance) may of course also reflect, apart from age-group-sensitive interests, ideological stances about redistribution that are unevenly distributed across birth cohorts, depending on the economic and cultural contexts in which each of these grew up. This might upset any simple correlation between attitudes on transfer systems for the elderly and age-related interests even if the latter were certain to have a significant causal impact. (On the relevance of generations as cohorts on political attitudes in Germany, see Matthias Metje, Wählerschaft und Sozialstntktitr im Generationswechsel. Eine Generationsanalyse des Wahlverhaltens bei Bundestagswahlen [Wiesbaden: Deutscher Universitätsverlag, 1994]).
29. On U.S. data, see Day, What Older Americans Think, p. 48.
30. According to a Belgian 1982 survey, for example, support for higher pensions went up monotonously from 61 percent among the under-25s to 80 percent among the over-65s, while support for higher child benefits peaked at 52 percent among those 25-34 and dropped to 31 percent among the over-65S (Paul-Marie Boulanger, "Vieillissement et gérontocratie dans une constitution démocratique," in Populations agées et revolution grise, ed. M. Loriaux, D. Remy, and E. Vilquin [Louvain-la-Neuve: CIACO, 1990], pp. 971-81, at p. 979).
31. In Switzerland's September 1990 national referendum on this issue, 47 percent of all those taking part voted in favor of the proposal (which was therefore rejected), compared with 57 percent among the voters aged 30-39 and 64 percent among those aged 18-29 (Möckli, "Demokratische Struktur," p. 5).
32. In the U.S., the proportion of households with children under 18 has declined from over half to slightly over a third between the 1950s and the 1990s (Levy and Murname, "Orphans of the Ballot Box").
33. See Rawls, Theory of Justice, pp. 60-61, 223-24,
34. on the (slightly qualified) immunity of universal suffrage and eligibility from consequentialist consideration, and Rawls, Theory of Justice, ibid., pp. 199, 359-61, 454 on the sharp contrast between the motivational assumptions required within market and democratic institutions.
35. In a less than ideal political world, however, Rawls (Theory of Justice, ibid., pp. 57, 198) allows this contrast to lose its sharpness.
36. See Niccoló Machiavelli, Discorsi sopra la prima deca di Tito Livio [1517], in N. Machiavelli, Opere, ed. E. Raimondi (Milan: Mursia, 1969), pp. 69-342, at 81-92;
37. and Quentin Skinner, Machiavelli (Oxford: Oxford University Press, 1981), pp. 64-71,
38. for clear formulations of this consequentialist approach to constitutional design, formulations of which can also be found in David Hume, Essays: Moral, Political and Literary [1741] (Indianapolis: Liberty Classics, 1985), p. 42;
39. John Stuart Mill, Considerations on Representative Government [1861], in On Liberty and Other Essays, ed. J. Gray (Oxford: Oxford University Press, 1991), pp. 203-467, at 298;
40. and indeed Rawls, Theory of Justice, pp. 57, 198.
41. John Roemer made me aware of the fact that this program could be viewed as a special case of the intellectual ambition behind so-called implementation theory - the attempt to design payoff structures in such a way that the individual agents' behavior will generate the socially preferred outcomes - at least providing one allows these outcomes to be characterized in nonwelfarist terms. Unlike implementation models with neat behavioral assumptions, however, our Rawls-Machiavelli program moves in the messy world of people "as they are or can realistically be made to be," on the lookout for context-dependent local institutional improvements.
42. I shall be using the term "constitutional" throughout in a broad sense that encompasses all the rules of the political game, including those (such as some electoral rules or the rules which govern the funding of political parties) which are not part of a country's constitution but of its ordinary laws.
43. I thereby pursue the inquiry I undertook, along other dimensions, in Van Parijs ("Justice and Democracy," and
44. "Should the European Union Become More Democratic?").
45. The uncompromising justice-consequentialist approach tested in this way is akin to the one sketched by Richard J. Arneson (Arneson, "Democratic Rights at National and Workplace Levels," in The Idea of Democracy, ed. D. Copp, J. Hampton, and J. E. Roemer [Cambridge: Cambridge University Press, 1993), pp. 118-48).
46. Mechtild Jansen, ed., Halbe-Halbe: Der Streit um die Quotierung (Berlin: Elefanten Press, 1986);
47. Françoise Gaspard, Claude Servan-Schreiber, and Anne Le Gall, Au Pouvoir Citoyennes: Liberté, égalité, parité (Paris: Seuil, 1992).
48. Wilma Rule and Joseph F. Zimmermann, eds., Electoral Systems in Comparative Perspective: Their Impact on Women and Minorities (Westport, CT & London: Greenwood Press, 1994);
49. Andrew Reynolds and Ben Reilly, The International IDEA Handbook of Electoral System Design (Stockholm: Institute for Democracy and Electoral Assistance, 1997), pp. 98-99.
50. Anne Phillips, The Politics of Presence (Oxford: Oxford University Press, 1995), p. 63.
51. Offe, "Zusatzstimmen für Eltern," p. 16.
52. Minimum voting age is still 19 in Austria and 20 in Switzerland, though (Joseph M. Vallès and Agustí Bosch, Sistemas electorales y gobierno representativo [Barcelona: Ariel, 1997], p. 53).
53. Joseph M. Vallès and Agustí Bosch, Sistemas electorales y gobierno representativo [Ibid., p. 44;
54. Jean-Pierre Perrin, "Un mollah outsider à la présidence de l'Iran," Libération 26 May 1997;
55. and Robert Ludbrook, Should Children Have the Right to Vote? (Sydney: University of New South Wales: National Children's and Youth Law Centre, 1996), p. 19.
56. For a presentation and discussion of these ideas, see Klaus Hurrelmann, "Für eine Herabsetzung des Wahlalters," in Jugend und Politik, ed. C. Palentien and K. Hurrelmann (Neuwied: Luchterhand, 1997), pp. 281-90;
57. Klaus Hurrelmann and Christian Palentien, "Jugendliche an die Wahlurnen! Argumente zur Verbesserung 'der politischen Partizipation der jungen Generation," Diskurs 2 (1997):38-45;
58. Christian Palentien, "Pro- und Contra-diskussion zu einer Veränderung des Wahlrechts," in Jugend und Politik, ed. Palentien and Hurrelmann, pp. 290-300;
59. and Hans Hattenhauer, "Über das Minderjährigenwahl-recht," Juristische Zeitung 1 (1996):9-16.
60. The lowering of suffrage to 16 was also proposed by a New Zealand Royal Commission in 1986, by the UK's liberal democrats in the late 1980s and by the Australian Youth Organisation AYPAC in 1994 (Ludbrook, Should Children Have the Right to Vote?, pp. 18-19).
61. More radical proposals for the extension of suffrage to all children had been made for example by John Holt (Escape from Childhood [New York: Penguin Books, 1974], p. 118);
62. and Hubertus von Schoenbeck, "Deutsches Kindermanifest," in Quellen und Dokumente der Antipädagogik, ed. Ulrich Klemm (Frankfurt am Main, 1980).
63. Many more modest proposals can also be found. For example, Campiglio ("Political Participation, Voting, and Economic Policy: Three Problems of Modern Democracies," in Understanding Democracy: Economic and Political Perspectives, ed. A. Breton, G. Galeotti, P. Salmon, and R. Wintrobe [Cambridge: Cambridge University Press, 1997], pp. 196-208, at p. 199) suggests that teenagers should have their own representatives (themselves not older than 30) with a political agenda restricted to "the problems of the young."
64. See Ursula Feist, "Niedrige Wahlbeteiligung - Normalisierung oder Krisensymptom?" in Protestwähler und Wahlweigerer, ed. Karl Starzacher, et al. (Cologne: Bundverlag, 1992);
65. for Germany; Möckli, "Demokratische Struktur," p. 3, for Switzerland;
66. Price, Justice, p. 82, for the United States. Voter turnout is particularly low among those 16 to 18 in Brazil
67. (Offe, "Zusatzstimmen für Eltern," p. 9).
68. Friedrich A. Hayek, Economic Freedom and Representative Government (London: Institute of Economic Affairs, 1973), Occasional Paper no. 39, pp. 19-21. Another, very different, way of working from both ends at the same time would consist (as suggested to me by Axel Gosseries) in introducing a competence test below and beyond a certain age, in such a way that a significant proportion of the under-18s would be enfranchised and a significant proportion of the over-70s disfranchised.
69. Stuart White and Andrew Williams suggested an analogous effect is achieved by attributing political power to an age-stratified random sample of citizens (at referendums or in assemblies), whose sampling bias favors younger age groups. (The coefficient of over-representation must equal the weight of one's vote in the parallel case of plural voting.)
70. Claimed by Jacques Lefèvre ("Le troisième âge: riche mais inexploité," Le Soir, 5 Aug. 1997, p. 15) to have been unsuccessfully put to a referendum in a Swiss canton.
71. Möckli, "Demokratische Struktur," p. 13.
72. Gerd Grözinger,"Achtung, Kind wählt mit! Ein Beitrag zur allmählichen Aufhebung der Diktatur der Gegenwart über die Zukunft," Blätter für deutsche und internationale Politik 10 (1993):1261-67, at 1265.
73. These are figures for 1994, derived, as are the figures in the next paragraph, from John W. Wright, The New York Times Almanac 1998 (New York: Penguin Books, 1997), p. 396.
74. Similarly, in Nasser's Egypt, both men and women had the right to vote, but only men the obligation (Asad Abukhalil, "Women and Electoral Politics in Arab States," in Electoral Systems in Comparative Perspective: Their Impact on Women and Minorities, ed. W. Rule and J. F. Zimmermann [London and Westport, CT: Greenwood Press, 1994], pp. 127-37, at p. 131), and in Belgium, where voting is compulsory for Belgian citizens at all elections, E.U. citizens now entitled to vote for local councils and the European Parliament are only subjected to this obligation if they choose to register as electors.
75. A further option, suggested to me by Erik Olin Wright, would remove altogether the problem that arises from the young turning up in lesser numbers. It consists in giving each age group a weight proportional to its total size in the electorate, rather than to the fraction of it that actually votes: "one could have a vote-inflater attached to each vote that is a function of the proportion of the age group in the electorate which simply multiplies the vote of a given 18-year-old, so that the total of 18-year-old votes is proportionate to their number in the population. (If 50 percent of 18-year-olds vote and 66 percent of 25-year-olds vote, then each 18-year-old vote gets a weight of 2 and each 25-year-old vote a weight of 1.5)." This vote-inflater technique is likely to create a self-regulating dynamics favoring the equalization of turnout rates - since my vote matters more if I belong to an age group with a low turnout, candidates and parties will be particularly keen to get my vote out - but even if it did not, the younger electors' lower turnout no longer translates into a lesser political weight. Even with a small number of age categories, the ex post calculation of inflation rates obviously makes for some sizeable complications. But the fact that this proposal, unlike those mentioned in this paragraph but like those mentioned in the next one, requires a separate counting of the votes cast by each age group should not be a prohibitive obstacle. What about extending the technique to the gender or ethnic or class divide?
76. See Möckli, "Demokratische Struktur," p. 12.
77. Bruce Ackerman, "Crediting the Voters: A New Beginning for Campaign Finance," The American Prospect 13 (1993):71-80.
78. Here again, and for reasons analogous to those spelled out in connection with the first family of proposals, the focus is on the electorate rather than the representatives. Introducing a parents' quota among the latter (a possibility mentioned, though not endorsed, by Claus Offe, "Zusatzstimmen für Eltern," p. 10) would be just as inadequate as the quota for the young mentioned above (indeed, even more so, as no analogous impact on turnout can be expected). It would leave electoral pressures unchanged, while not guaranteeing a better defense of the interests of parents and their children. Indeed, these interests may be far more effectively defended by an energetic childless person trying hard to tap the parents' votes than by some of the latter's overstretched, exhausted peers.
79. As pointed out by Alfred Sauvy (Bien-être et population [Paris: Edition sociale française, 1945], p. 214), there are countries, including France, in which the size of a constituency is measured (for the sake of seat allocation) by the size of its population, not of its current electorate, and extending the suffrage to all children would be consistent with this practice.
80. Even without giving parents extra votes, our electoral systems are arguably already assuming now that children's interests are represented through their parents: "Their parents have the vote, and we trust them to use it (at least in part) to protect their children's interests as well as their own. . . . There is nothing illegitimate, in those circumstances, in letting such others speak on the child's behalf (Robert E. Goodin, "Enfranchising the Earth, and Its Alternatives," Political Studies 44, no. 5 [1996]:835-49, at p. 843).
81. Moreover, in some countries at any rate, parents are entitled to vote on behalf of their children qua shareholders (Konrad Löw, "Es gibt kein allgemeines Wahlrecht," Frankfurter Allgemeine Zeitung, 11 July 1997).
82. In 1960, 35.7% of the U.S. population was under 18; in 2000, only 25.7% is expected to be (Wright, New York Times Almanac, p. 276).
83. See André Toulemon, Le Suffrage familial ou suffrage universel intégral (Paris: Librairie du Recueil Sirey, 1933), pp. 121-22.
84. A de facto approximation to a more general form of parental vote is the plural voting system that was in place in Belgium between 1893 and 1910. Universal male suffrage was combined with an extra vote for married (or widowed) men, providing they were also taxpayers, and most of these married men (given that the average number of children per family was then larger than now and life expectancy shorter) had at least one nonvoting child at home (see, e.g., Andrew McLaren Carstairs, Short History of Electoral Systems in Western Europe [London: George, Allen & Unwin, 1980], p. 102).
85. Lasserre, De la Réforme et de l'organisation du suffrage universel (Paris: Victor Palmé, 1873);
86. Tarde, "Le suffrage 'dit' universel," in Etudes pénales et sociales (Paris, 1892): "For while children mean future and hope, women are above all the children's mothers, and the interest of the nation is that its statesmen should worry, not about the present generation, on which the thought of adult men usually gets stuck, but about posterity"
87. (see Toulemon, Le Suffrage, 108-9, 111-13, 200-201; on the "near miss" see pp. 115-39 and 217.)
88. Sauvy, Bien-être et population, pp. 213-14;
89. Adolphe Landry, Traité de démographie (Paris: Payot, 1949), p. 634;
90. Charles F. Westoff, "Marriage and Fertility in Developed Countries," Scientific American 239, no. 6 (1978):51-57, at p. 56.
91. The Front national's web site phrases the proposal as the attribution of extra votes to families in proportion to the number of minor children. Support in France is broader than the extreme right, however: just before the 1997 general election, the conservative representative Christine Boutin (from President Chirac's Gaullist party) gathered the support of 123 members of the French Assembly from various parties around a proposal that included, in vaguer terms, some sort of family vote. In Belgium, the right-wing Flemish-nationalist party has also picked up the idea: "Children, on whom policy exerts a great influence, are not democratically represented. Therefore, the Vlaams Blok advocates that parents should vote on behalf of their children until they reach majority" (Gerolf Annemans, et al., Een Kenze voor het leven [Brussels: Vlaams Blok, 1998], p. 9).
92. Konrad Löw, "Das Selbstverständnis des Grundgesetzes und wirklich allgemeine Wahlen," Politische Studien 213 (1974):19-29.
93. Heiko Schultz, "Mit Familienwahlrecht Einfluß nehmen," Sozialdemokratischer Pressedienst, 25 Aug. 1992;
94. Lore Maria Peschel-Gutzeit, "Unvollständige Legitimation der Staatsgewalt oder geht alle Staatsgewalt nur vom volljährigen Volk aus?" Neue juristische Wochenschrift 43 (1997):2861-62.
95. For further details, see Dieter Suhr, "Transferrechtliche Ausbeutung und verfassungsrechtlicher Schutz von Familien, Müttern und Kindern," Der Staat. Zeitschrift für Staatslehre, öffentliches Recht und Verfassungsgeschichte 29 (1990):69-86;
96. Grözinger,"Achtung, Kind wählt mit!";
97. Konrad Löw, "Verfassungsverbot Kinderwahlrecht? Ein Beitrag zur Verfassungsdiskussion," in Familie und Recht 1 (1993):25-28 ,
98. and Löw, "Es gibt kein allgemeines Wahlrecht";
99. and Hattenhauer, "Über das Minderjährigenwahlrecht," as well as the proceedings of a conference on the subject gathering left-of-center academics and activists at the Akademische Akademie Arnoldshain (ed. Gerd Grözinger and Helmut Geiger, Zukunft wählen - Zusatzstimmen für Eltern? [Bad Boll: Evangelische Akademie, 1993]).
100. Stein Ringen, "In a Democracy, Children Should Get the Vote," International Herald Tribune, 14-15 Dec. 1996;
101. Paul E. Peterson, "An Immodest Proposal," Daedalus (1996):151-74;
102. Bayer, "Let's Give Parents an Extra Right to Vote";
103. Campiglio, "Political Participation, Voting, and Economic Policy," 198-99;
104. Manuel Carballo "Extra Votes for Parents?" Boston Globe, December 17, 1981, p. 35. In his short piece in the Boston Globe, Carballo writes: "I am left with an uncomfortable sense of imbalance in our political system. In a society all too ready to live for the present, how do we create a political force for our children's pensions?" Here is his answer: "My proposal is quite simple. Give parents a vote weighted by their number of minor children. Two parents with two children get four votes. One parent with one child gets two votes." According to Harvard University's web site, "throughout a life dedicated to public service, Carballo was committed to serving the poor, the vulnerable and the public at large", so much so that Harvard's Kennedy School of Government set up a Manuel Carballo Memorial Prize to encourage "innovations for improving the quality and effectiveness of programs to serve the poor and disadvantaged." This is worth remembering, just in case anyone tried to disparage the idea by exhibiting such embarrassing supporters as France's Le Pen or Germany's Dyba.
105. Article 8 of the law proposal by Sallies, et al., of the Fédération républicaine, gave each father as many extra votes as he currently had minor children, and one extra vote for as long as he lived (Toulemon, Le Suffrage, p. 217).
106. Ringen, Citizens, Families and Reform (Oxford: Oxford University Press, 1997).
107. Alternatively, one could restrict the same assignment formula using the children's rank order, not among all children born in the family, but only among those of them who are still under voting age. Not unlike Sauvy's scheme mentioned above, this variant could be interpreted as making fathers and mothers take turns in proxying for their children (except the first one) as these grow older (even though, for the sake of administrative simplicity, it may be better to severe the individual proxy link between parent and child and simply give the mother one more vote than the father if they have an odd number of minor children). Note too that this variant involves a much greater bias in favor of mothers: at any one time, mothers would have at least as many votes as fathers in all, not only in most households.
108. In the U.S., nearly a third of all children were not living with both their parents in 1995 (see Wright, New York Times Almanac, p. 284).
109. Several of these difficulties are articulated by Claus Offe ("Zusatzstimmen für Eltern," fn. 3).
110. Note that all of these justifications, in the interpretation I give them, are consequentialist. It would be absurd to claim that this unequal distribution of voting rights among adults according to the number of their children is just in itself: what about those who could not find a partner, for example, or those who cannot have children, or those whose children have died? It can only be shown to be just (if at all) by virtue of its indirect effects on the distribution of resources.
111. André Toulemon, "Influence du vieillissement de la population sur la composition du corps électoral," in Actes des journées pour l'étude scientifique du vieillissement de la population (Paris: Alliance nationale contre la dépopulation, 1948), pp. 107-15, quote at p. 114.
112. In his earlier book, Toulemon (Le Suffrage, pp. 200-201) quotes at length the philosopher and psychologist
113. Tarde ("Le suffrage 'dit' universel"): "The electoral peak would be the 45-50 age slice, i.e. the culmination point, not of imagination and love, but of experience and political capacity among most men. . . . The reform I am dreaming of would have the indisputable advantage of making the numerical weight of a person's vote proportional . . . to his physical or mental strength and to his civic importance throughout his life." 56. One may wish to replace the most plausible conjecture (3) by the following variant: (3) Voters are guided by their own self-interest and that of the other people with whom they actively interact outside market relations and with whom they thereby develop close emotional ties. As persuasively pointed out to me by Erik Wright, the extent to which the elderly are firmly integrated into multigenerational communities - not just multigenerational kinship structures - then becomes of decisive importance: "If the elderly live in settings where children and young adults help them and interact with them, then [they] will develop a stronger sense of obligation towards future generations." Even more than the child-spreading policies to be considered in this section, packages of policies (housing, town planning, public transport, social services, health care) that encourage socially active generationally mixed neighbourhoods - rather than segregated old people's homes and condominiums - could then be expected to boost the political potential of intergenerational justice.
114. In the U.S., the proportion of childless women in the 40-44 range has increased from 10.2% to 17.5% between 1976 and 1994, while the total fertility rate went up from 1.8 to 2.0 children per woman in the same period (Wright, New York Times Almanac, p. 282).
115. Thomas Büttner and Wolfgang Lutz, "Estimating Fertility Responses to Policy Measures in the German Democratic Republic," Population and Development Review 16, no. 3 (1990):539-55, at pp. 540-43. Less clear-cut, but still quite spectacular, is the case provided by Italy and Sweden swapping positions in the European fertility league between 1970 and 1990. Italy went down from about 2.4 children per woman in 1970 to 1.7 in 1980 and 1.3 in 1990, while Sweden went first down from 1.9 in 1970 to 1.7 in 1980 and then up to 2.1 in 1990 (far above the European Union average of 1.5). It is hard to resist the temptation to relate this swapping of positions to a number of striking policy differences. Child benefits are generous and kept pace with inflation in Sweden, but now hardly exist in Italy, where annual government expenditure on children outside education is about $400 per capita, compared to $1800 in Sweden (1992 figures). Italy's maternity leave of 20 weeks at 80% of pay is not bad at all on European standards, but no match to Sweden's full year (or more) of parental allowances (for either mother or father) with a 70 to 90 % rate of salary replacement - probably the world's most generous scheme. Moreover, the world of work is so organized that the rate of women's participation in the labour force is 85% in Sweden (with two fifths in part-time jobs), double the corresponding Italian rate.
116. (See Jean-Claude Chesnais, "Fertility, Family, and Social Policy in Contemporary Western Europe," Population and Development Review 22, no. 4 (1996):729-739, at pp. 730-33;
117. and Anne Hélène Gauthier, The State and the Family: A Comparative Analysis of Family Policies in Industrialized Societies (Oxford: Oxford University Press, 1996), pp. 174-75).
118. Note, however, that since 1990, the difference has been shrinking , with total fertility rate down in 1995 from 1.3 to 1.2 in Italy, but from 2.1 to 1.7 in Sweden (Chesnais, "Fertility," 730). Note too that Italy has a particularly developed preschool system (with over 85% of the children between three and school age in subsidized institutions compared to 80% in Sweden (Gauthier, The State, p. 181) and that Italy, unlike Sweden, has a tax credit system for dependent children. For this reason, the difference between Sweden and Italy is less sharp than non-educational government expenditure figures suggest.
119. Didier Blanchet and Olivier Ekert-Jaffé, "The Demographic Impact of Family Benefits: Evidence from a Micro-Model and from Micro-Data,"in The Family, the Market and the State in Ageing Societies, ed. J. Ermisch and N. Ogawa (Oxford: Clarendon Press, 1994), pp. 79-103, at pp. 92-93.
120. Anne Hélène Gauthier and Jan Hatzius ("Family Benefits and Fertility: An Econometric Analysis," Population Studies 51, no. 3 [1997]:295-307, at p. 302) show that a 25% increase of the benefits given for the first two children would raise total fertility by 0.01 child per woman in the short run, by 0.07 in the longer run. Given that, in the countries considered, the average level of benefits for a two-child family was about 5% of average earnings and that the average total fertility rate was 1.71, this implies that a benefit increase by 1% of average earnings increase births by nearly 4%. Earlier, using data for 11 European countries in the 1969-83 period,
121. Blanchet and Ekert-Jaffé ("Demographic Impact," p. 93) came up similarly with a significant and positive regression coefficient between the total fertility rate and a family policy index consisting of a ratio of the weighted average of child benefits to the average wage: a 25% increase in the latter index increases the total fertility rate by 0.04 child per woman. (The total fertility rate in a given calendar year is the expected number of children per woman, with each woman experiencing each year between the ages of 10 and 50 the probability of giving birth exhibited in the current calendar year by women in the corresponding age categories [between ages 10 and 50].)
122. In the case of families with average earnings and two children, from 0% in Italy or the U.S. and 0.3% in Spain to 10.4% in Belgium or 11.3% in Austria (Gauthier, The State, p. 166).
123. This result conflicts with L. A. Whittington's econometric analysis of the impact of tax relief on birth rates, on the basis of 1979-83 U.S. individual sample data (Whittington, "Taxes and the Family: The Impact of the Tax Exemption for Dependents on Marital Fertility," Demography 29, no. 2 [1992]:215-26).
124. The average annual nominal tax value of the exemption for a family in the 15% bracket was about $300 per child, i.e., a modest 8% of the estimated cost of a child. Depending on the tax bracket in which it fell, the real tax value (base year 1983) to a family ranged from $0 to $812. For a given income, it varied over the period, owing to a fall in the top bracket tax rate from 70% to 50%, to a rise in the exemption and to inflation (Whittington, "Taxes," pp. 216-17). The results support the conjecture of a significant positive impact of the average value of the tax exemption. However, because of the short time span, they cannot rule out that this may simply be a timing effect
125. (Whittington, "Taxes," p. 223), without any lasting impact on either total fertility or rate of childlessness.
126. Even this short term impact, however, given the modesty of the amounts involved (see Gauthier, The State, p. 170 for an international comparison) and the fact that they are inversely related to income, remains a challenge to the cross-national negative results.
127. Gauthier and Hatzius, "Family Benefits," p. 300. This is not necessarily a blow for those countries - such as France or Germany
128. (Blanchet and Ekert-Jaffé, "Demographic Impact," p. 90) - which chose to pursue natalist aims by generously focusing benefits on the third child: focusing on the third child is so much cheaper than focusing on the first that it may well yield better natalist value for its money, despite the greater demographic impact of a given per-capita increase of the benefit for the first child than for the third.
129. The presumption might seem to be, more specifically, in favor of benefits targeted at poorer households, since it is hard to imagine how the demographic impact could not be greater, for a given cost, if benefits were concentrated on lower-income families than spread more thinly over all families. But one must realize that it is a serious mistake to reason about redistribution schemes in the same way as one reasons about other expenditure programs (a point well put, for example, by Daniel Shaviro, "The Minimum Wage, the Earned Income Tax Credit, and Optimal Subsidy Policy," The University of Chicago Law Review 64, no. 2 (1997):405-81). Whereas the cost of these programs matches an opportunity cost in terms of other things one could have done with the same resources, increasing or decreasing the "cost" of a redistribution scheme is more appropriately described as a shifting of the profile of marginal and average net tax rates applying to the various components of the population. In particular, rather than as a "cheapening" of an expenditure program, the phasing out of benefits as family income increases is more appropriately described as the subjection of households with children, and hence de facto in most cases their female secondary earners, to a higher effective marginal rate of taxation than childless households: any additional euro earned is not only subjected to the explicit rate of tax but also to the rate of benefit withdrawal. There might conceivably be reasons for taxing mothers at a higher marginal rate than other workers, but they are most unlikely to have anything to do with the encouragement of a first birth.
130. An econometric study by John Ermisch ("Economic Analysis of Birth Rate Dynamics in Britain," Journal of Human Resources 23, no. 4 (1998):563-76, at pp. 571-75) on the basis of individual British Census data for 1971-85, that higher hourly wages for women (relative to men's weekly wages) and higher housing costs (relative to the cost of living) were the main factors behind the increase in the proportion of women remaining childless. Leaving aside, for the moment, the influence of wages, his results suggest that doubling housing costs would add another 7% of childless women, while reducing family size by 0.16 child only, and doubling the parity-independent child benefits would reduce the proportion of childless women by about 3.5%, while increasing average family size by 0.17 child. If the aim is to reduce childlessness (or stem its progress) rather than to increase (or maintain) family size, child benefits look less appropriate a tool than keeping housing costs down, as the latter has a much more powerful effect on childlessness than on average family size. However, acting on housing costs (rents, mortgage interest rates, housing subsidies) may well just be a clumsy, distortionary, truly costly way of providing implicit child benefits at a decreasing rate as family size increases. The same asymmetric effects should therefore be expected from increased cash benefits if they were focused on the first child, rather than parity-independent.
131. See, for example, Dieter Birnbacher, Verantwortlichkeit für zukünftige Generationen (Stuttgart: Reclam, 1988).
132. In the more ambitious versions, the Guardian would be empowered to take governments to court on the basis of clauses in a constitution or in an international treaty which protect future generations against both the governments' actions and their inaction. This may be thought to be question-begging, as it presupposes that the constituent or treaty-endorsing bodies are sufficiently driven by a sense of intergenerational justice to enshrine some features of what it commands in legal clauses that can be enforced even against governments and assemblies. However, while it is obvious that this device could not work on its own, it could nevertheless prove quite effective by virtue of the fact that representative assemblies may be willing to adopt principles in a certain form and at a certain time, which they may be under pressure to abandon when confronted to specific issues.
133. See Adam Swift, Gordon Marshall, and Carole Burgoyne, "Which Road to Social Justice?" Sociology Review 2, no. 2 (1992):28-31.
134. The sheer fact of systematically bringing the intergenerational issue into the open may also exert a civilizing influence on the monopoly-power-wielding generation through a mechanism that does not rely on belief formation. Hover strong the self-interested pressure on the representatives of this generation, they may shy away from publicly dismissing a perfectly audible strong case on behalf of the unrepresented (whether other countries or future generations) out of anticipated shame for the moment these will find out that their interests were deliberately ignored.
135. See, e.g., Gerald Curtis, "The Unintended Effects of Japan's Electoral Reform," paper presented at Yale University, Department of Political Science, 25 Feb. 1998,
136. and Donald L. Horowitz, "Constitutional Design: An Oxymoron?" in Designing Democratic Institutions, Nomos 41, ed. I. Shapiro and S. Macedo (New York: New York University Press, 1999).
137. This concern is expressed by Offe ("Zusatzstimmen für Eltern," pp. 21-22) in connection with the proxy vote for children.
138. That there is ground for concern is strongly substantiated by the virulent attack on the "civil servant" (versus self-interest-seeker) conception of the elector by the most articulate advocate of the parents' vote André Toulemon (Le Suffrage, pp. 179-89, esp. pp. 184-85): "When voting, the elector does not attempt to hide that he defends his interests and nobody blames him for it; quite the contrary, in order to catch his votes, the most honest and even the wisest candidate endeavors to show the elector that his interest, well understood, commits him to accept his program and his person. Whoever would tell the electors 'Vote for this program, even though this will be in your interest neither now nor later; free yourself of the selfishness that is natural to any well born creature,' would rightly be considered a madman or an imbecile; for it is obvious that the electors have indisputably the right to vote in defence of their interests."
139. The possibility of this counterproductive effect was pointed out to me by Andrew Williams. It is arguably documented by the fact that in the 1990 Swiss referendum on the abandonment of nuclear energy, support went down monotonously from 64% for the 18-29-year-olds to 32% for the 50-59-year-olds, but went up again to 46% for the over 60s (Möckli, "Demokratische Struktur," p. 5).
140. This possibility displays a standard case of fallacy of competition. If, for any particular type of person (in terms of propensity to care about the future), votes are distributed according to the number of children, the overall time horizon is lengthened. Also, if there were no correlation between type and number of children, such a proportional distribution of votes would lengthen the time horizon. But if there is a strong correlation, voting power is being shifted across types as well as within types to such an extent that the net effect may be a shorter overall time horizon.
141. More contingently, having more children may also be strongly correlated, in some countries, with membership in religious communities - say, Mormon, Hassidic, Amish, catholic traditionalist, or Islamic fundamentalist - which tend to adopt political attitudes - for example, in favor of state-imposed morality or against State-organized social policy - inimical to other dimensions of the full ideal of social justice (understood as some liberty-constrained maximin). The parents' vote would boost the power of these communities and may therefore, under certain demographic and institutional conditions, badly damage a country's overall political potential for social justice. This possibility too must be paid the attention which Erik Wright convinced me it deserves.
142. Note, however, that in the most extensive cross-country study "maternity leave (duration and benefits) did not appear to be significantly related to fertility" (Gauthier and Hatzius, "Family Benefits," p. 304) - contrary to what was suggested by the East/WestGermany and Italy/Sweden comparisons
143. (Büttner and Lutz, "Estimating Fertility Responses";
144. Chesnais, "Fertility").
145. Ermisch's ("Economic Analysis," pp. 571-75) estimates for the U.K. suggest that a 35% increase in women's hourly wages relative to men's (similar in magnitude to the increase in the 1971-85 period) would add about 7% to the (then) current 16% of childless women and depress average family size by 0.3 child from its current level of 2.0. Note, however, that, in Gauthier and Hatzius's cross-national study
146. ("Family Benefits," p. 300), women's wages display a slightly significant positive relationship with fertility levels, which suggests that the opportunity cost of large families, for given women's wage rates, may be very different depending on the extent to which being the mother of more than two children means giving up one's career altogether (availability of child care and parttime jobs, extended parental leave with a right to return, etc.).
147. Though possibly at a flat rate (irrespective of a woman's current wages and past career) and rather in the form of benefits not contingent upon the interruption of work (so that they can, for example, be used to top up wages in order to improve one's housing rather than to enable one of the parents to stay at home).
148. An alternative way of removing the side effect (suggested to me by Andrew Williams) would consist in enforcing substantial paternity and maternity leaves of equal length. This would get rid of the gender bias, but might further contribute to the shrinking of the number of families with at least one child (by strengthening the base for the male partner's veto power), thereby killing the solution at the same time as the defect in the solution.
149. This is of course, with or without public-interest embroidering, the distinct form of instrumentalism that drives attempts by political parties to modify the rules of the game for their own benefit. This partisan instrumentalism is vulnerable in a way in which "Machiavellian" instrumentalism is not. Think, for example, about transitions from "first past the post" to proportional representation (PR). When Belgium became the first country to make the move in 1899, Vandenpeereboom's Catholic government
150. Similarly, "the French Socialist Government's decision in 1986 to switch from their existing Two-Round System to PR was widely perceived as being motivated by partisan reasons, and was quickly reversed as soon as the government lost power in 1988" (Reynolds and Reilly, eds., International IDEA Handbook, p. 123),
151. in sharp contrast, for example, to New Zealand's 1993 well-prepared switch from First Past the Post to list PR, also motivated by the expected consequences but in nonpartisan fashion (Nigel Roberts, "New Zealand: A Long-Established Westminster Democracy Switches to Proportional Representation," in Reynolds and Reilly, eds., International IDEA Handbook, pp. 129-31).
152. A survey conducted at the 1990 West German election showed that the electoral scores of Social Democrats and Greens would go up slightly, while those of the Christian Democrats and the Liberals would shrink slightly, if parents were given proxy votes for their children (Grözinger, "Achtung, Kind wählt mit!" p. 1266).
153. In contrast, a lowering of the minimum voting age to 16 would seem to have a more clearly favorable effect for Greens and Social Democrats (Klaus Hurrelmann, "Mit 16 Jahren an die Wahlurne?" Reutlinger General Anzeiger, 23 Feb. 1996).
154. Even in the 1980s there would have been left majorities throughout Europe, had only males been allowed to vote! (Offe, "Zusatzstimmen für Eltern," p. 20).
155. Grözinger ("Achtung, Kind wählt mit!" p. 1267) points out, in response to a similar objection to his proxy-vote proposal that it may be rational for a majority to bind itself through constitutional change in order to protect itself against the weakness of its will. This is another way of formulating the second case just mentioned.
156. See again Grözinger ("Achtung, Kind wählt mit!" p. 1267), who similarly hints at the diversity of the reasons different categories of people may have for supporting his proxy vote proposal.
157. Elster, "Comment on van der Veen and Van Parijs," Theory and Society 15 (1986):709-22;
158. Elster, "Arguments for Constitutional Change: Reflections on the Transition to Socialism," in Constitutionalism and Democracy, ed. J. Elster and R. Slagstad (Cambridge: Cambridge University Press, 1988), pp. 303-23 .
159. Giovanni Sartori, Comparative Constitutional Engineering: An Inquiry into Structures, Incentives and Outcomes (London: Macmillan, 1984), p. 202.
160. See Rawls, Theory of Justice, 359-61: "A peculiarity of the ideal market process, as distinct from the ideal political process conducted by rational and impartial legislators, is that the market achieves an efficient outcome even if everyone pursues his own advantage." In contrast, "there seems to be no way of allowing [citizens and legislators] to take a narrow or group-interested standpoint and then regulating the process so that it leads to a just outcome." Does this mean that the same holds for Rawls as for Pascal's God: "Un peu de pensée éloigne de [lui], beaucoup y ramène"?
161. The absence of any such restriction distinguishes the radical instrumentalism defended here from a milder "epistemic" variant that might be ascribed to David Estlund ("Democracy without Preference," Philosophical Review 3 [1990]:397-423)
162. or Thomas Christiano (The Rule of the Many: Fundamental Issues in Democratic Theory [Boulder, CO, and Oxford: Westview Press, 1996]).