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

by Niall Ferguson

  Still, it is probably true that authoritarian regimes between the wars were more ambitious in this area, not least because they were less respectful of the traditional fiscal orthodoxy that had helped kill off prototype welfare states like the Weimar Republic.99 Though rearmament had come to dominate the German economy by 1938, the Nazis initially pursued job-creating policies that were not mere spin-offs of rearmament, spending up to 5 billion reichsmarks on job creation to the end of 1934 and devoting still more to the construction of 4,000 kilometres of Autobahn, a programme which employed 120,000 workers at its peak. ‘Each measure’, declared Hitler in July 1933, ‘is to be judged [according to the criteria]: what are its consequences? Does it create more employment or does it create more unemployment?’100 As is well known, the regime’s policy of state investment in infrastructure and armaments had achieved full employment by the mid-1930s, though historians continue to argue about the relative importance of civilian job-creation schemes and rearmament.

  Welfare too had been a Nazi preoccupation even before Hitler came to power. In 1931 Goebbels took the Berlin-based Nazi People’s Welfare Association under his wing; after May 1933 it spread to cover the whole Reich, swallowing up private charitable institutions in the process. By 1939 it covered over half of all households and was second in its membership only to the German Labour Front. Of course, National Socialist conceptions of welfare were distinctive, not least because ‘ethnic aliens’ were systematically excluded and an ethos of public activism was encouraged to promote the sense of ‘ethnic community’ Hitler craved. But in other respects there were disquietingly ‘modern’ aspects to the Nazi welfare state: the compulsory deductions of Winter Aid from pay slips, the procreation-friendly child allowances, the subsidized ‘Strength through Joy’ pleasure cruises and holiday camps.101

  The welfare state was thus no invention of William Beveridge, nor of the 1945 Labour government which implemented the recommendations of his famous report. Most of the key elements of that government’s economic policy – progressive taxation, national insurance, publicly funded education and state ownership of key industries – predated the 1940s. Even the notion that fiscal policy could be geared to maintain full employment had occasionally been put into practice before Keynes gave it intellectual respectability in his General Theory. What was new in Britain after 1945 was the aim of universal coverage, to allow the abandonment of means-testing. The implication of this was that, unless national insurance contributions were regularly adjusted to take account not only of inflation but also of the demand for health care or unemployment benefits, entitlements would almost inevitably outstrip what claimants could expect to receive under a real system of insurance. The breaking of the link between contributions made and entitlements received was to prove the crucial flaw of the British welfare system.

  The original effect of national insurance was in fact regressive so long as contributions (and benefits) were flat-rate.102 The first deviation came in 1959 when earnings-related supplements were introduced for both pensions and contributions, significantly increasing the progressivity of the tax system. The second came in the 1980s, when the Conservatives did away with the link between earnings and state pensions. This has significantly reduced the British government’s liabilities for state pensions compared with many other European countries: by the mid-1990s, the effective cut in pensions implied a saving of over 3 per cent of GDP.103 But the fact that the link between earnings and national insurance contributions was not broken amounted to a further step towards treating national insurance contributions as a shadow income tax.104 It is too seldom pointed out that, although she cut income tax rates, Mrs Thatcher raised the standard rate of employees’ national insurance contributions from 6.5 per cent to 9 per cent. For those on half average earnings, national insurance contributions became almost as burdensome as income tax.105 In 1949 income tax accounted for a third of all taxes, national insurance contributions for less than a tenth. By 1990 the proportions were, respectively, 28 per cent and 18 per cent. Even then, the rule whereby the fund should not fall below one-sixth of total national insurance expenditures would have been broken in the 1990s without ‘top-up’ payments from the Treasury.

  In the same way, the creation of a National Health Service funded out of taxation but supposedly ‘free at the point of use’ has imposed all the rising costs of an ageing population (to say nothing of increasingly sophisticated medical treatments) directly on central government finances. According to government estimates in the 1980s, an extra 1 per cent expenditure per year in real terms was necessary to maintain real spending per head at a constant level; the figure for the 1990s was closer to 2 per cent.106 But the pressure on governments of both parties to restrain total public spending means that from time to time such a real increase is not achieved. In effect, the NHS is a system of central rationing – to call it planning would be to flatter successive ministers – the effect of which is to hold total expenditure on health as a proportion of GDP substantially below the European and North American averages.107

  FROM WARFARE TO WELFARE

  Taken together, the processes described in Chapters 1 to 3 help to explain the transformation of the warfare state into the welfare state. The processes of parliamentarization and bureaucratization were first made necessary by the cost of war. But in the twentieth century they developed a momentum of their own, increasingly diverting resources away from military towards civilian employment and redistributive transfers.

  Perhaps the simplest way to illustrate the extent of the transformation is to compare government finance in Britain in 1898 and 1998. In 1898 gross public expenditure was equivalent to just 6.5 per cent of GDP. In 1998 the comparable figure (total managed expenditure as a percentage of GDP) was 39 per cent. In 1898 the biggest item of the budget was defence (36 per cent), followed by debt service (21 per cent) and civil government (20 per cent). Just over 10 per cent was spent on ‘Education, Art and Science’. In 1998 the biggest outlays went on social security (30 per cent), health (17 per cent) and education (12 per cent). The two biggest items of a century before, defence and debt service, now account for just 7 per cent and 9 per cent of spending. No less striking are the changes on the revenue side. The biggest sources of gross public revenue in 1898 were the excise (29 per cent), followed by customs (19 per cent), the income tax (15 per cent) and death duties (13 per cent). In 1998 the main sources of government revenue were income tax (26 per cent), national insurance contributions (16 per cent) and VAT (also 16 per cent). Inheritance tax now brings in less than 1 per cent of total revenue; customs duties a mere half of 1 per cent.108

  As these figures show, there is nothing novel in the idea of the budget as an instrument of redistribution: the high proportion of debt service in 1898 represented a transfer not much smaller in relative terms than the social security system in 1998. It is the nature of the transfer that has changed, as we shall see: from a system that was socially regressive, as a consequence of the way it financed its wars, to one that regards the reduction of material inequality as its primary function.

  In the light of the seemingly inexorable growth of welfare spending, we may well ask: is there ‘a limit to taxable capacity’?109 As Calvin Coolidge is said to have remarked, ‘Nothing is easier than spending the public money. It does not appear to belong to anybody. The temptation is overwhelming to bestow it on somebody.’ Yet even the most dirigiste social democrats had to draw the line somewhere if there was to be a meaningful distinction between their creed and outright socialism. Thus the Labour Home Secretary Roy Jenkins declared in 1976: ‘I do not think you can push public expenditure significantly above 60 per cent and maintain the values of a plural society with adequate freedom of choice. We are here close to one of the frontiers of social democracy.’110

  In fact, there is no need for politicians to devise such ‘lines in the sand’ for themselves. For there are real economic constraints that explain why the state’s expenditures and employment tend not to rise fa
r above, respectively, a half of output and a third of employment. One of these we have already seen: the limit on how much can be raised in taxation to finance expenditure before diminishing returns set in, not only in terms of revenue, but in terms of aggregate economic growth.111 We now turn to the other variable in what economists call the ‘inter-temporal budget constraint’: the limit on how much a state can borrow.

  SECTION TWO

  PROMISES TO PAY

  4

  Mountains of the Moon: Public Debts

  Great is Bankruptcy. CARLYLE

  If you walked up 6th Avenue to 43rd Street in Manhattan, you used to see a thirteen-digit number on a billboard above you. The last time I saw it (on 17 October 1999), the number was:

  5,601,723,423,979

  Above it are three words: ‘Our National Debt’. Before it is the dollar sign. And below it there are two small calculations: ‘Your Family Share: $73,192’ and ‘Increase per second: $10,000.’

  It is a little piece of history, that sign: a relic of the once acrimonious debate about American public finance which played such an important role in the politics of the late 1980s and early 1990s. In 1986 critics of Ronald Reagan’s fiscal policy took out an advertisement to warn readers of the New York Times that the burgeoning debt would lead to ‘the death of [the] Republic’. Paul Kennedy cited the total debt for 1985 (then a mere $1.8 trillion) as one indicator of impending American overstretch, adding darkly: ‘Historically, the only other example which comes to mind of a Great Power so increasing its indebtedness in peacetime is France in the 1780s, where the fiscal crisis contributed to the domestic political crisis.’1

  If that $10,000-per-second figure was to be believed, the national debt would have become a fourteen-digit number by the beginning of the year 2001: ten trillion dollars. Yet President Clinton was able to claim in February 2000 that, under his last budget plan, all US public debt would be repaid by 2013,2 which might seem to imply an ‘increase per second’ of the order of minus $1,000. Whatever happened to the American debt crisis? To answer this question it is necessary to set those thirteen digits above 6th Avenue in a rather broader historical and economic perspective. A long-run view of public debt reveals that an apparently large ‘mountain’ of debt may be far from disadvantageous, provided the institutions of a country’s financial system are equal to the task of its management. In the uneven geographical development of these institutions – a bove all, the institution of a funded national debt – lies one of the keys to modern history.

  THE ORIGINS OF PUBLIC DEBTS

  Though the history of private debt may be traced back as far as the second millennium BC, the history of public debt is much shorter.3 Neither ancient Greece nor ancient Rome had formalized public debts. Nor did the early Abbasid caliphate, though the central treasury in Baghdad still had to borrow for short periods in anticipation of tax receipts, illegally paying interest or rewarding lenders with non-cash privileges.4 The late development of public debts is somewhat surprising, since in the modern world states are generally (though not always correctly) seen by investors as less likely than private debtors to default on loans. Many of the essential institutions of credit predated large-scale public borrowing. Ways had been found to circumvent the laws against usury – condemned alike by Catholicism, Protestantism and Islam – by the early Middle Ages; bills of exchange were in use in Genoa from the twelfth century, and the first negotiable bills, which could be transferred to a third party through endorsement, date from the fourteenth century.5 As we shall see, however, the early modern risk premium was more often paid by rulers than by merchants.

  It was the simple fact of taxation – of more or less predictable revenue streams – that provided the basis for the earliest systems of public debt in medieval Italy. The Venetian public debt, which originated in the twelfth century, was secured on the state salt monopoly, the revenues of which were earmarked for debt service and redemption. In the thirteenth century the increasing use of forced loans (prestiti) as a form of taxation further increased the importance of the debt. Something similar happened in the sixteenth century, when the Monte Nuovo6 was established to administer the repayable tax known as the decima. In Genoa, the salt tax revenues themselves were sold at auction to comperisti, a system which, in the fifteenth century, was put under the control of a quasi-public bank, the Casa di San Giorgio.7 A similar system evolved in Florence, where the communal debt, administered by the Monte Comune, was systematically increased by the fisc’s heavy reliance on forced loans (prestanze). An important development here was the transferability of claims on the Monte, which could be sold to other citizens freely or, with authorization, to outsiders.8 In 1526 a Monte della Fede was established to manage the papal debt.9

  North European city-states evolved somewhat different arrangements based on the sale of perpetual, terminable or life annuities. In each case, an investor lent his capital to the state in return for a stream of income. In the case of a perpetual bond, that income stream was notionally infinite: the state would go on paying a percentage of the face value of the bond for ever, but of course the investor never got his capital back. A terminable bond, by contrast, paid interest and repaid capital for a fixed period, after which the bond ‘matured’. Life annuities, then as now, paid interest only for the duration of the investor’s or another specified life. From the late fourteenth century, Cologne offered perpetual but redeemable annuities paying from 5 to 5½ per cent.10 Such redeemable bonds were usually called ‘purchases of money’ or ‘sales of dues’, and tended to be secured on a piece of immovable property like a town; interest was called a ‘gift’ to circumvent the usury laws. Dutch cities, on the other hand, issued liffrenten (lifetime annuities) and losrenten (perpetual loans). In 1586 the Receiver General of the Union between the Dutch states began issuing obligaties, which were more easily transferable than urban bonds – and hence more attractive to investors, who might wish to liquidate their investment before a bond matured. However, the greater part of Dutch borrowing in the subsequent centuries was done at the regional level, mainly by the province of Holland, since it was the provinces that controlled the bulk of tax revenues.

  Medieval monarchs, by contrast, tended to rely on loans from wealthy banking families to finance their deficits. Siennese and Florentine bankers lent to the kings of England; Tuscan bankers to the Roman curia; South German bankers to the Habsburgs; Swiss and Italian bankers to the French.11 The Spanish crown turned first to Genoese merchant bankers (hombres de negocios), then to Portuguese marranos.12 It made sense to rely on international financiers when, very often, the money was needed to pay for armies fighting abroad.13 But it is important to remember that these were often little more than personal loans to individual rulers, like the £300,000 borrowed by Edward III.14 Only in Catalonia in the late fifteenth century was there anything like the system that had evolved in the Italian and German city-states. The Catalan system guaranteed investors regular interest out of revenues that were earmarked for the purpose (hypothecated) and managed by a special commission.15

  Haltingly, in the course of the sixteenth century, the other European monarchies learned to mimic the techniques of urban public debt. In France, for example, the Paris Hôtel de Ville issued perpetual 8 per cent annuities known as rentes. The money was handed to the crown in return for certain royal revenues being assigned to the Paris Receiver General; the advantage to investors was that the General Farm paid the interest payments directly from its coffers, rather than via the less than reliable royal fiscal administration.16 The volume of rentes grew substantially in the course of the seventeenth and eighteenth centuries: by the 1780s Necker put the capital sum at around 3.4 billion livres, and this may well have been an underestimate.17 The Spanish crown developed a two-tier system of short-term, high-interest loan contracts (asientos) and long-term, lower-interest bonds assigned on ordinary revenues (juros), which by the 1560s had become transferable, and which could be purchased in perpetual, lifetime or redeemable forms.18
Likewise, life and perpetual annuities in the Habsburg Netherlands in the 1540s were serviced by the revenues from excise and property taxes.19

  An important innovation which spread from Italy throughout Europe in the course of the seventeenth century was the public bank. Here it is important to distinguish between two functions that were originally performed by distinct institutions: the management of the state’s debt and the management of forms of money other than coinage (which tended to be entrusted to a separate mint), in particular the system of clearing that was so vital to the development of large-scale commerce. Although there were forerunners of these public banks in Genoa and the Florence of the Medicis, the first true public banks were the Banco della Piazza di Rialto (founded in 1587), which reformed the Venetian currency and payments system by accepting deposits, effecting transfers between accounts and accepting bills of exchange payable to its clients; and the Banco del Giro (1619), which converted a part of the Venetian state’s short-term debt into interest-bearing and transferable bonds (partite).20 The Amsterdamse Wisselbank (1609) performed similar functions to the Rialto Bank, but also dealt in bullion and minted coins. It was soon imitated in Middleburg (1616), Hamburg (1619), Delft (1621) and Rotterdam (1635); and later in Austria (Wiener Stadtbank, 1703), Denmark (Kurantbanken, 1736), Sweden (Riksen Ständers, 1762), Prussia (Königliche Giro- und Lehnbank, 1765) and Russia (Assignationsbank, 1768). The Sverige Riksbank in Sweden (1668), on the other hand, was more like the Venetian Giro Bank, as was the Bank of England (1694). Unlike the Amsterdamse Wisselbank, the Bank of England’s primary function was to manage the government’s debt. However, its regional monopoly on note issue and its extensive commercial business gave it a natural interest in (and hence, over time, responsibility for) currency stability, which meant maintaining the convertibility of paper notes into specie.21