The Cash Nexus: Money and Politics in Modern History, 1700-2000
Page 18
Table 3. The growth and structure of the London Stock Exchange, 1853–1990
* * *
Total value (£ millions)
UK government share (per cent)
* * *
1853
1,215
70.2
1863
1,683
53.6
1873
2,270
37.6
1883
5,677
24.0
1893
6,561
16.5
1903
8,834
13.4
1913
11,263
9.0
1920
16,626
32.6
1933
18,476
35.3
1939
18,507
35.7
1945
24,701
49.3
1950
25,063
54.9
1960
45,060
31.9
1970
107,414
15.0
1980
280,328
21.7
1990
2,098,492
5.9
* * *
Source: Michie, London Stock Exchange, pp. 88 f., 175, 184, 320, 322, 360 f., 419, 421, 440, 473, 521 f., 589 f.
Note: To 1933: Nominal values; from 1939: market values.
Yet all this may be a debate about a flawed concept. It is widely acknowledged already that the term ‘deficit’ is an ill-defined one. To take the British case, what contemporaries regarded as the bottom line of the central government’s budget (the balance of the ‘consolidated fund’) from the 1870s until the 1930s tended to understate the size of current surpluses by counting payments to the sinking fund as expenditure.122 The Treasury also made a somewhat arbitrary distinction between expenditure ‘above the line’ and ‘below the line’, which notionally but not exactly distinguished between current and capital expenditure. Moreover, the consolidated fund does not include the national insurance fund, nor does it include the borrowings of local authorities and public corporations (which were included in the Public Sector Borrowing Requirement, the measure of the deficit introduced in 1976).123 And this too is considered by some economists a measure inferior to the cyclically adjusted Public Sector Fiscal Deficit, which seeks to exclude the influence of public sector financial transactions (such as privatization) and the cyclical fluctuations of economic growth.124
There are even more profound definitional problems.125 In the modern dynamic theory of fiscal policy, the key concept is the government’s inter-temporal budget constraint. This means that the sum of the ‘generational accounts’ of those now alive plus those of future generations has to be equal to the sum of future government purchases plus the government’s net debt. Generational accounts represent the sum of the present values of the future net taxes (taxes paid minus transfer payments received) that members of a birth cohort can be expected to pay over their remaining lifetimes, assuming current policy is continued. The sum of the generational accounts of all members of all living generations is how much those now alive will pay towards the government’s bills. The government’s bills, on the other hand, are the present value sum of all of the government’s future purchases of goods and services plus its official net debt (its official financial liabilities minus its official financial assets, including the value of its public-sector enterprises). Bills not paid by current generations must be paid by future generations. This is the zero-sum nature of the government’s inter-temporal budget constraint. Essentially, existing debt must be fully funded in the long run by cumulative budget surpluses.
However, different choices of fiscal labels can alter the present generation’s accounts and the government debt by equal absolute amounts, leaving the next generation’s accounts and the government’s future purchases unchanged. Suppose, for example, that the British government had chosen in 1998–9 to label workers’ national insurance contributions a ‘loan’ and the additional Basic and State Earnings Related Pension benefits paid to workers in old age in recognition of those contributions ‘payment of interest and principal’ on those ‘loans’, less an ‘old age tax’ (levied at the time contributors receive their benefits). This alternative set of words would have increased the government deficit by roughly £45 billion, instead of the surplus officially claimed. The government’s debt would also have risen. However, so would the generational accounts of currently living generations, since their future ‘old age tax’ would now be included in their accounts. The burden on future generations would therefore remain the same. And the economic position of the present generation would also be unaffected by the change of labels. Each worker would have handed the government the same amount of money in 1998 and would receive the same amount of money from the government in the future.126
The fact that the government uses one set of words rather than another is therefore a matter of semantics not economics. Each set of words results in a different measure of the deficit. But there is nothing in economic theory to lead one to prefer one measure to another. This approach to public finance – known as generational accounting – is little more than a decade old, but it has already been adopted in more than twenty countries.127 We shall return to its distributional and political implications in Chapter 6.
DEBT SERVICE
The most economically important measure of public debt may therefore not be the current outstanding nominal amount of debt, but the relationship between present and future tax burdens. On the other hand, the most politically important measure of public debt is more likely to be the current cost of debt service as a proportion of government expenditure. This is certainly the most visible measure to a government struggling to make ends meet, for the simple reason that every penny spent on debt service – in effect, the ongoing cost of past policies – is a penny that cannot be spent on present policies.
When state budgets were relatively modest, debt charges could be immense. In fifteenth- and sixteenth-century German towns, debt service averaged around a third of total budgets. In princely states and kingdoms, there was wider variation. In the first half of the sixteenth century the state of Hesse paid between 2 and 9 per cent of total spending on debt service. The figure for Württemberg in the same period was 80 per cent. Somewhere in the middle was Spain, where by 1543 nearly two-thirds of ordinary revenue was going on interest on the juros.128 France too ended the sixteenth century with burdensome debts – some four-fifths of annual revenue was already assigned at the start of Henry IV’s reign129 – but thanks to the reforms of Sully, her debt burden declined in the course of the seventeenth century to around a fifth of total spending between 1663 and 1689. Naples, by contrast, paid as much as 56 per cent of the budget on debt charges in 1627.130 Papal debt service was also high, rising from 36 per cent in 1526 to a peak of 59 per cent in 1654.131 By comparison, eighteenth-century Austrian debt service was low, at between a quarter and a third of total spending.132
History provides plentiful examples of political crises due to the rising burden of debt service. The ability of German city-states to preserve their independence often hinged on this: thus Mainz, which by 1411 was paying almost half its total revenue to the holders of annuities, lost its independence in the fifteenth century, while Lübeck and Hamburg, where debt service was lower, did not.133 The Spanish monarchy’s difficulties in thelatesixteenthandseventeenthcenturieswerecloselyrelatedtorecurrent debt crises. As early as 1559 total interest payments on the juros exceeded ordinary revenue; and the situation was not better in 1584 when 84 per cent of ordinary revenue went to bondholders. By 1598 the proportion was back to 100 per cent.134 The Dutch Republic was able to sustain much higher absolute levels of debt than its continental rivals, yet paid relatively small amounts to service the debt. In the 1640s, for example, debt service accounted for just 4 per cent of the total budget
. But even here a limit was finally reached. By 1801, six years after the provincial and union debts had been consolidated into one, debt service amounted to 41 per cent of the budget. The French Republic which had overrun Holland in 1795 was, by contrast, unburdened by debt, for reasons to be discussed below.135
Pre-revolutionary France is perhaps the most notorious case of a state brought low by the costs of debt service. Between 1751 and 1788 interest and amortization payments rose from 28 to 49 per cent of total expenditure, or from just over a quarter of tax revenue to 62 per cent.136 In fact, the cost of debt service to France’s main military rival was not much less. Between 1740 and 1788 British debt charges rose as a proportion of tax revenues from 37 per cent to 56 per cent.137 But the key point is that France had a substantially lower debt than Britain both in absolute terms and as a proportion of national income. Between 1776 and 1782 French debt charges amounted to around 7.5 per cent of the total debt, compared with a figure of 3.8 per cent for Britain. In other words, the cost of servicing the same amount of debt was roughly twice as high for France. This crucial disadvantage was only partly due to higher payments for amortization; the main reason will be explored in the next chapter.
Figure 9 presents figures for debt service in relation to budgets since the early nineteenth century, showing that it was not until the 1870s that other major states approached Britain in this respect. The British data show that for almost the entire period between 1818 and 1854 more than half of gross central government expenditure was going on debt service, close to the debt burden carried by the French ancien régime on the eve of the Revolution. But Britain was able to reduce the burden of debt gradually from the late 1830s, as the figure shows, while French (and Italian) debt charges caught up as a result of the wars fought from Sebastopol to Sedan. From the end of the 1860s until the mid-1880s, Britain, France and Italy were all spending around a third of their budgets on debt charges. Rising expenditure on other civil and military functions caused the proportions to fall towards the Prussian level (below 10 per cent) in 1913, except in France where the figure remained just above 20 per cent.
The figure also makes clear how different were the burdens the four states carried after the First World War. Whereas in Britain and France debt service peaked at around 44 per cent of total government spending, in Italy the average figure for the 1920s was just under 18 per cent. In Germany – for reasons we shall soon see – debt service was just 2 per cent of total spending in 1925. Interest and debt repayments have mattered far less in Germany and France since 1945, though in both cases the share of total spending has been going up since the early 1980s. In Britain, debt service tended to fall from the 1950s to the 1990s, whereas in Italy the trend was in the opposite direction, culminating in the mid-1990s, when more than a fifth of total government expenditure was going on the national debt.
Figure 9. Debt service as a percentage of government expenditure, 1802–1999
Sources: Flora et al., State, Economy and Society, vol. i, pp. 381 ff., except: France 1802–22: Mann, Sources of Social Power, vol. ii, p. 373; UK 1802–1914: Mitchell and Deane, Abstract of British Historical Statistics, pp. 396–9; Prussia: Gerhoff, ‘Der Staatshaushalt’, p. 5; Jahrbuch für die Statistik des Preussischen Staats (1869), pp. 372–443, 466–545. All figures from 1982 to 1999 are from OECD.
Note: German figures for 1870 to 1914 are for general government; as are all figures from OECD. Other figures are for central government.
The obvious explanation for the declining importance of debt service as a proportion of government spending might simply be that government budgets in the nineteenth century were so small. As we have seen, the growth of the welfare state had barely begun in this period, so that payments to bondholders were the principal transfers made through national treasuries. This is certainly a part of the story, to which we shall return in Chapter 7. There are, however, other reasons why debt was so much less expensive for Britain than for her eighteenth- and nineteenth-century rivals; and why, at least for the developed economies of the West, debt today is relatively less expensive than in the past. These are the subject of the next chapter.
5
The Money Printers: Default and Debasement
‘To whom it may concern, this note of hand
Is worth a thousand ducats on demand …’
…
You signed: as if by sleight of hand, behold,
That night provided copies thousandfold,
And, so that all might have the boon to share,
We stamped the total series then and there.
Tens, Thirties, Fifties, Hundreds, all to date,
You cannot think how people jubilate.
…
None now has power to stay the flying chits,
They ran as quick as lightning on their way,
And money-booths kept open night and day,
Where every single note is honoured duly
With gold and silver – though with discount, truly.
GOETHE, Faust1
In 1912 the German Union of Women’s Suffrage held a well-attended meeting on the subject of ‘Inflation’. At that time, consumer price inflation in Germany – as measured in the price of food – was just under 5.3 per cent per annum. This was its highest level since 1880: the average annual inflation rate since the foundation of the German Reich in 1871 had been little more than 1 per cent. The Suffragists’ meeting was one of many expressions of public anxiety about high prices: as one newspaper commentator had remarked the year before, ‘Everyone talks about the rise in the cost of living.’2 But talking about inflation is not the same as understanding it.
In 1924 the UWS met again to consider not inflation but stabilization. As a result of the previous year’s disastrous hyperinflation – which had seen the annual rate of inflation peak at 182 billion per cent – the society’s assets were now worth precisely five marks and fourteen pfennigs. Though they had been denied the vote before and during the First World War, they had patriotically invested their funds in German government bonds.3 Those purchased during the war were now worth precisely nothing.
Twice in the space of twenty-five years, the German government debt all but vanished as a result of the total collapse in value of the currency. Twice those who put their faith in the credit of the German Reich were left with worthless paper. It was as if there had been a collective failure to read to the end of Goethe’s masterwork (though, as it happens, the Vice-President of the Reichsbank during the first hyperinflation was a distinguished Goethe scholar).4 In the scene in Part Two, Act I, from which the epigraph above is taken, the printing of money at first seems to bring prosperity. The paper money flows ‘to wine-shops, butchers, bakers, / With half the world as glutton merry-makers’. Clothiers, tailors and restaurateurs do a roaring trade. ‘Such paper-wealth’, declares Mephistopheles, ‘is practical.’5 But in Act IV Mephistopheles reveals that these were ‘bogus riches’; and the country on which he bestowed them has ‘collapsed in anarchy’:
With men both high and low emmeshed in feud,
Brother by brother murderously pursued.
Castle fought castle, town invaded town,
And guilds had plots to pull the nobles down,
Chapter and flock against the bishop rose,
And nowhere could men meet, except as foes,
In church they stabbed to kill, before the gate
The travelling merchant met a bloody fate …
So they … limped on, fell, rose again perhaps,
Then, losing balance, lurched to a collapse.6
Writing shortly before his death in 1832, Goethe probably had the French experience during the 1790s in mind. But this passage also foretells uncannily the early history of the Weimar Republic.
The German experience of inflation has been by any standards extreme. The memory of hyperinflation was still being cited as a factor in German politics in the late 1980s, as politicians sought to persuade voters that a new European currency would be as soun
d a currency as the deutschmark (which in fact depreciated by 75 per cent in the fifty years of its existence). Yet the experience of default through inflation is in many ways universal – as universal as the story of Faust. Since 1899 the price of a packet of cigarettes in Britain has risen by a factor of 15; the price of a loaf by a factor of 32; and the price of a pint of beer by a factor of 456. The average weekly wage has risen by a factor of 89.7 By contrast, a British government consol with a face value of £1,000 has actually fallen in price.
HOW NOT TO PAY
There are five ways to reduce transfer payments in the form of debt interest and repayment when they reach what is judged (politically) to be excessive. First, part or all of the debt can simply be paid off. One obvious way of doing this is by levying a one-off capital levy on the bondholders or, for that matter, all wealthy groups. Secondly, the interest paid on the debt can be reduced by legislative act, an operation known as a ‘conversion’. Thirdly, payments to bondholders can be suspended by fiat. Fourthly, an unanticipated rise in inflation can reduce the real value of both debt and interest payments, provided the debt is not index-linked or denominated in foreign currency (or gold). This has often been seen as the easier political option; and, as we shall see, twentieth-century governments found it hard to resist. The final option – the hardest but best way to reduce a debt burden – is to achieve an increase in the real rate of growth, though under certain circumstances the very existence of a large public debt may make this difficult.
The most politically ‘respectable’ way to reduce the real debt burden is by repayment, that is by running recurrent primary budget surpluses (meaning surpluses greater than current debt interest). Occasions when a debt has been wholly repaid are in fact relatively few. Between 1816 and 1834, to give one of the rare examples, the total US federal debt was paid back.8 However, both the United States and Britain regularly managed to reduce their total debts by running primary surpluses. Between 1822 and 1914 the British national debt was reduced by about a quarter in nominal terms as a result of a sustained programme of debt repayment. In the United States there were also debt reductions between 1805 and 1811, 1871 and 1893, 1920 and 1930, and again (though on a much smaller scale) between 1947 and 1953.9