THE ORIGINS OF THE PUBLIC FINANCE FRAMEWORK

The secret of Holland's seventeenth-century fiscal system lay in its mediated structure. Urban public finance facilities dating back to the Middle Ages were made to serve provincial needs through careful grafting which, because the cities held power in the Estates, helped to inspire confidence that financial demands remained reasonable, the money well spent, and payments met promptly.Footnote ⁶ The literature interprets this arrangement as a static equilibrium resulting from taxpayers doubling as debt holders and political decision makers, and motivated by considerations of citizenship.Footnote ⁷ However, we emphasize that the arrangement was really dynamic, a continuous bargaining process between local and central levels, between taxpayers, investors, and political representatives, and between urban and provincial officials, over fiscal policy.

The grafting started in 1482, when the six biggest cities jointly floated a loan on their account, but on behalf of the province.Footnote ⁸ The provincial loans issued during the 1540s were also controlled by the local tax receivers, members of the central administration but part of the local elite.Footnote ⁹ They doubled as debt managers, selling annuities and paying interest and redemptions from their tax receipts. At this stage, voluntary buyers were found only in Amsterdam; elsewhere office holders or wealthy citizens were forced to buy in proportion to their census contribution.Footnote ¹⁰

For the loans issued during the 1550s, Holland could do without forced purchases.Footnote ¹¹ The issues consisted of a mix of life annuities, more expensive but self-extinguishing, and losrenten, annuities redeemable at the debtor's wish. Redeemable annuities appear to have served a specific segment, investors seeking a secure store of wealth for long-term holding; transfers were rare.Footnote ¹² To supplement the annuities loans the Estates occasionally raised money by selling obligaties or promissory notes issued on the credit of a corporation or one of its officials pledging particular revenues to redeem a specified debt.Footnote ¹³ Being basically unfunded, they carried a higher interest rate than annuities.

During the 1560s Philip II tried to centralize and increase taxation. The effort, combined with rising religious tensions and an economic slump, sparked the Dutch Revolt. The military emergency pushed Holland into a serious fiscal crisis.Footnote ¹⁴ Interest payments were suspended from 1572 and forced the Estates to borrow from army commanders and suppliers at very high interest rates. Confiscated émigré and church property compensated the most demanding creditors.Footnote ¹⁵ The emergency also shifted the balance between cities and province. To ease the pressure, the Estates seized control over local excises from the cities in 1574. Initially conceived as an emergency measure, this administrative change transformed the patchwork of local excises into a uniform tax system of common means (gemeene middelen) across the whole province, a tax revolution which necessarily complemented the earlier shift upwards of borrowing power.Footnote ¹⁶ The rebel provinces which signed the Union of Utrecht (1579) accepted to introduce a similar system.

However, the new fiscal regime's full potential remained as yet unrealized because mobilizing private savings proved difficult. A forced loan was contemplated in 1581 but not carried through. In 1585 Holland's towns and villages had to borrow on their own credit to anticipate the revenues of a wealth tax.Footnote ¹⁷ A loan from the English Crown helped to make ends meet.Footnote ¹⁸ The province also continued to rely on expensive stopgaps such as short-term loans at punitive rates from merchants and army suppliers.Footnote ¹⁹ Loans raised by the cities supplied the rest. During the years 1584–1604 Amsterdam borrowed about 1 million guilders for the province, and Delft probably did likewise.Footnote ²⁰

LEARNING TO MANAGE THE MARKET, 1588–1621

During 1588 Holland's prospects changed dramatically. The Great Armada was defeated and the Spanish army's advance stalled, enabling the republic to secure its frontiers during the subsequent decade. After the immediate danger had subsided, Holland reorganized its public finance. Rising tax receipts gradually restored Holland's credit, enabling the Estates to cut issuing rates from 12 percent in 1590 to 8.3 percent in 1597. Though still above its own previous low, and high compared to some states, this was the same rate Holland had paid during the late 1550s and on a par with Geneva, so it probably reflects the market price for small, fledgling Protestant regimes.Footnote ²¹ The Estates also converted bills issued during the 1580s into annuities issued in the name of the gemeene land, i.e., the province.Footnote ²²

By 1594 Holland resumed sales of annuities through its 18 local tax receivers.Footnote ²³ These officials had considerable leeway.Footnote ²⁴ Provincial officers preferred to issue redeemable annuities or life annuities, but they allowed receivers to sell obligaties or bills if needed. This article was formally short-term, up to twelve months, akin to treasury bills in Anglo-Saxon parlance but, being habitually rolled over, tended to become long-term, as will become clear below. During 1594–1598 the receivers raised 850,000 guilders in annuities plus nearly 640,000 guilders in bills, but booming private business then raised the stakes, so Holland switched to forced loans supplemented by bills.Footnote ²⁵ Between 1599 and 1603 Holland borrowed 900,000 guilders through voluntary sales of annuities, 3.2 million in forced loans, plus 2.4 million in bills.Footnote ²⁶ Between 1604 and 1609 Holland primarily sold life and redeemable annuities, with bills occasionally mixed in. When the 1609 Truce with Spain took effect, Holland's debt had reached 14.4 million guilders, of which 5.2 million guilders in bills.Footnote ²⁷

The public credit recovery was highlighted by declining interest rates. The issuing rate for redeemable annuities dropped from 8.3 percent in 1597 to 7 in 1605 and to 6.25 in 1608, only slightly trailing the rates paid by Amsterdam merchants and the Dutch East India Company (VOC) for short-term loans (Figure 1).Footnote ²⁸ The Estates succeeded cutting interest costs further several times with a withholding tax on redeemable annuities of 0.5 percent in 1596, and 1 percent in 1601, 1604, and 1607.Footnote ²⁹ With maximum rates of 9 percent, bills paid slightly more than annuities (Figure 1), a consequence of the bills being sold on the receivers' personal credit.Footnote ³⁰ The rating difference gradually disappeared, however. From 1597 buyers could ask the Estates to guarantee the bills; in 1605 the Estates reduced the receivers' commission for selling bills from 1 to 0.5 percent, and from 1607 redeemable annuities and bills had identical issuing rates.Footnote ³¹ From 1608 these stood at 6.25 percent, a rate last obtained in 1556, putting Holland in the European middle band, above Genoa, Venice, Castile, and France, but below Naples, Rome, Bologna, and fellow Protestant Britain and Geneva.Footnote ³²

FIGURE 1

INTEREST PAID ON VOC BONDS, PRIVATE BONDS IN AMSTERDAM, AND HOLLAND'S REDEEMABLE ANNUITIES, 1595–1688

Sources: Private bonds: Gelderblom and Jonker, “Completing,” appendix, ACA 5060, Inv. Nrs. 50, 51, 162; ACA 235, Inv. Nr. 288, 291–295; ACA Oud-Rechterlijk Archief, Inv. Nrs. 482–493; VOC Bonds 1602–1608: NA 1.04.02 Inv. Nr. 7142, 1608–1688: Van Dam, Beschryvinge; and Holland's redeemable annuities: Liesker and Fritschy, Gewestelijke.

Economic growth boosted the recovery of Holland's credit. Between 1590 and 1609 fiscal revenues doubled.Footnote ³³ Moreover, the expanding economy boosted the supply of savings; the average annual increase of the wealth of Amsterdam merchants, for instance, may have been as high as 12.5 percent in this period.Footnote ³⁴ Potential public debt buyers had plenty of alternative investment opportunities, however. Both the booming European trade and the first intercontinental expeditions yielded much higher returns.Footnote ³⁵ Hence to accept the large loans of 1599–1603, investors needed prodding by forced issues for about half of the total. As before, such loans were based on a census of the entire population. Commissioners assessed every household in detail, after which those possessing 2,000 guilders or more had to accept a loan for 0.5 percent of their wealth in redeemable or life annuities.Footnote ³⁶ To sweeten the pill, the loans were exempt from the withholding tax, but the receivers still needed the threat of suspending interest payments on existing loans for investors to swallow their allotted portion.Footnote ³⁷

Both provincial and local authorities had resorted to such census-based levies before, first during the 1540s and again during the 1570s and 1580s emergency.Footnote ³⁸ This time, however, the Estates clearly intended to create a regular fiscal resource. Revised and updated in 1599 and 1602, the census ledgers enabled the receivers to subdue investors holding out for better conditions, offering them the option between a rinse (voluntary loan) and a shave (forced loan), with the potential third option, a haircut (levy), remaining unused until the 1620s. The ledgers harnessed the market so well that officials hailed them as the foundation of Holland's finances.Footnote ³⁹ However, as static appraisals of real assets, the ledgers were quickly obsolete, expensive to update, and unable to capture the fast growing liquid wealth from commerce.

To tap this commercial wealth, Holland raised substantial amounts in bills between 1588 and 1609. With a maturity of up to 12 months, bills offered investors the advantages of a slightly higher interest rate plus greater control over their money than annuities. Redemption of annuities was at the authorities' discretion. Transfer of ownership passed through an official register, for which a fee plus a 3.3 percent duty for transferring annuities was charged.Footnote ⁴⁰ With bills, the boot was on the other foot. Billholders could demand repayment at the end of term and effect transfers without formalities or extra cost, because bills were, in effect bearer paper, transferable by endorsement, really liquid equivalents of time deposits.Footnote ⁴¹ As such they were ideally suited to merchants with their constantly shifting liquidity needs, so the receivers in Amsterdam, Rotterdam, or Delft found it easier to sell bills than annuities. For their part, the Estates constantly urged receivers to sell bills only if no takers could be found for annuities, probably because they realized that its creditors could cause a public finance crisis at any moment by demanding their money back.Footnote ⁴² Presumably, the receivers countered that danger by staggering interest payments to match the flow of revenues coming in.

The loans raised between 1588 and 1609 were puny by later standards but the forces shaping issuing policy were the same. The Estates tailored amounts to be raised, the instruments, and their price to perceived market conditions. By letting the receivers mediate the preference mismatch between the province and its creditors, the Estates could raise loans as quickly, as cheaply, and as close to its funding needs as local markets allowed. Moreover, by setting maximum loan issuing rates, the Estates gave officials an incentive for bargaining with investors.Footnote ⁴³ The 1609 Truce with Spain altered the stakes of both parties. Funding pressure eased and debt issues remained comparatively small, creating room for the province to convert all bills and redeemable annuities into 6.25 percent redeemable annuities. The conversion proceeded slowly because investors bridled at the interest rate cut.Footnote ⁴⁴ It nevertheless succeeded because investors disliked the alternative, repayment, and because the Estates exempted the new annuities from transfer tax, a typical mix of threats and incentives made possible by the temporary financial breather.Footnote ⁴⁵ In similar circumstances, the attempted conversion of the Estates General's debt failed for a lack of money.

THE FREE MARKET'S STRETCH, 1621–1648

What policy choice the Estates of Holland may have had in the 1610s disappeared during the much shorter but massively more expensive second phase of the Eighty Years' War. Between 1621 and the Peace of Westphalia in 1648, Holland's debt rose more than fivefold to 125.5 million guilders, not counting a floating debt of 7.2 million.Footnote ⁴⁶ Letting the debt rise so quickly required a firm confidence in the country's fiscal resilience, which the economic boom clearly inspired. As before, the Estates preferred issuing annuities, but tweaked financial policy in two respects: they used wealth levies rather than forced loans to supplement voluntary purchases of annuities, and from the late 1620s they increasingly relied on bills rather than annuities. By 1648 bills amounted to 73 million guilders, or 60 percent of Holland's debt (Figure 2).

FIGURE 2

THE ESTIMATED SIZE AND COMPOSITION OF HOLLAND'S DEBT, 1590–1720 (million guilders)

Source: See the Appendix.

Presumably, there were three reasons for using wealth levies rather than forced loans. First, despite new taxes military expenditure outpaced revenues, so any measure slowing the growth of interest charges was welcome.Footnote ⁴⁷ Second, forced loans enabled taxpayers who needed liquidity to sell their paper, but rising wealth had lessened their need to do so. Third, the Estates feared pushing issuing rates upward by floating too much debt. The alternatives were first debated in 1622, when a proposed forced loan was dropped in favor of a wealth levy.Footnote ⁴⁸ The following year Delft objected to a free loan because local investors would demand higher interest rates, suggesting wealth taxes or forced loans instead.Footnote ⁴⁹ The yield from wealth levies began to drop, however, because the 1602 ledgers were too old, shifting the burden away from merchants with their liquid wealth to property owners in less dynamic sectors.Footnote ⁵⁰ The ensuing debate shows the Estates unable to find the lesser evil among the growing interest burden, wealth levies, and forced loans.Footnote ⁵¹ When in 1625 the Estates finally did issue a forced loan of 0.5 percent on wealth assessed, they drafted a careful compromise between conflicting interests. The annuities issued in return carried 3 percent interest during the first six years, rising to the going rate of 6.25 percent after that; they were also free from the 3.3 percent transfer duty during the first three years. To ensure a fair balance between wealth owners, the Estates also had the 1602 ledgers updated, so the loan was clearly intended as a one-off.Footnote ⁵² So it remained. Though the Estates debated the option of a forced loan again in 1626, 1627, 1628, 1630, and in 1633–1635, each time they decided to levy a wealth tax instead, sometimes made more attractive by letting the taxpayers have annuities for their contribution if they paid before a set date.Footnote ⁵³ By 1635 the tax yield had started to decline again, but now the Estates ensured a light touch for mercantile wealth by not updating the 1625 ledgers.Footnote ⁵⁴ Taxing wealth remained a last resort, occurring on average once every three years until 1650.Footnote ⁵⁵

By the mid-1630s the Estates could afford the policy change from forced loans to wealth levies and bills because the market absorbed bill issues with remarkable ease. The administrative groundwork for the transition was done in 1628, when the province substituted its own credit for that of the receivers. Henceforth, new bill issues were backed by a certificate of indemnity for the receiver pledging all provincial revenues for servicing the debt.Footnote ⁵⁶ The administrative tightening up made little difference for bill buyers, since receivers had provided such a guarantee on demand since 1597, but it did signal the Estates' commitment to the new issuing policy. Using bills on such a scale meant building up a perpetual debt with short-term instruments. Before the Truce of 1609 the debt-tax ratio had barely touched three, that is to say, the debt could have been redeemed with three years' revenues. Now the rate shot up to 12 by 1650 (Figure 3), rendering debt retirement unlikely given the frequency of European wars. Consequently, the bills turned from short-term paper into de facto perpetuals. This quasi-consolidation took place gradually and without any change to status or price of the bills, creating a remarkable anomaly in the history of public finance: a rapidly growing debt financed largely with short-term instruments at issuing rates that were amongst the lowest in Europe.

FIGURE 3

THE RATIO OF DEBT TO TAX REVENUES IN HOLLAND, 1600–1720

Source: See the Appendix; and Liesker and Fritschy, Gewestelijke.

How to explain this anomaly? By 1630 Holland combined a strong credit record with a booming economy and savings outpaced the rise in public debt. Historians agree that the period up to the 1650s was the most dynamic phase of the republic's economic growth.Footnote ⁵⁷ Moreover, the net debt increase between 1621 and 1648 more or less equaled the interest paid out during that period, with net annual transfers from investors to the province seldom exceeding 1 million guilders (Figure 4). The ballooning debt could thus be financed entirely by reinvesting interests paid out.Footnote ⁵⁸ Investors obliged because the bills were always redeemable at par and at short notice but as their savings accrued, few billholders exercised the option. In 1621, 1622, and 1623 redemptions amounted to only 56,000, 84,000, and 148,000 guilders.Footnote ⁵⁹ By 1635 the roll-over of bills was so common that the provincial receivers lost their small fee for it.Footnote ⁶⁰ Declining market rates in the 1630s led to plans for a conversion of redeemable annuities and bills from 6.25 percent to 5. When preparing this conversion in 1640, the Estates deemed 800,000 guilders enough to cover expected redemptions on an estimated total debt of 94 million.Footnote ⁶¹

FIGURE 4

ESTIMATED NET ANNUAL TRANSFERS FROM THE PROVINCE OF HOLLAND TO ITS CREDITORS, 1600–1720 (interest paid out minus new debt issues)

Source: See the Appendix; and Liesker and Fritschy, Gewestelijke.

Investors had little to gain from a formal consolidation and a lot to lose (in particular the liquidity from the option to redeem at the tax receivers' offices), as the fate of redeemable annuities showed. With a transfer tax and related formalities, Holland effectively locked its annuities into a relatively small segment. Data from Gouda demonstrate this phenomenon to good effect (Figure 5). The annuities transfers recorded by the city secretary show little activity until the 1650s. In 16 years between 1613 and 1650 there were no transfers; in 12 more years, only one or two annuities changed hands. The nine remaining years saw three or four transfers, exceptionally five or seven, for amounts in the hundreds of guilders rather than in the thousands. Moreover, the few recorded prices were invariably at par. The annuities clearly catered for a specific market segment, i.e., investors holding the paper long enough to cancel the penalty on transfers. This segment absorbed no more than the 50 million guilders reached by the early 1640s, anchoring prices at par until the 1672 emergency forced a fire sale.

FIGURE 5

TRANSFERS AND PRICES OF HOLLAND'S REDEEMABLE ANNUITIES RECORDED BY GOUDA'S TOWN SECRETARIES, 1600–1720

Source: Gouda Archives, Oud-Rechterlijk Archief, Inv. Nrs. 482–493.

If investors could only lose from consolidation, the Estates did not stand to gain. Their debt management depended entirely on continuously rolling over short-term paper at interest rates that could hardly be lower and by now they probably knew that the likelihood of a roll-over crisis was minimal.Footnote ⁶² Consequently, we must see Holland's anomaly as an equilibrium of interests achieved on one side by rapid economic growth and wealth accumulation and on the other by the soaring expenditures needed to defend that prosperity. The equilibrium depended on the receivers' dual-action safety valve, which gave investors an exit option while buffering roll-over crises.

TRAPPED IN EQUILIBRIUM, 1648–1672

The 1648 peace with Spain did not bring an end to fiscal concerns. Indeed, at 7 million guilders annually, interest payments absorbed 60 percent of revenue.Footnote ⁶³ Moreover, the debt's short-term character sat uneasy after an incident in 1653. When the first Anglo-Dutch War (1652–1654) appeared to go against the republic, billholders giving notice squeezed several receivers, notably the one in The Hague. The bottlenecks were solved by transferring funds between offices and by having receivers persuade creditors to leave their money untouched.Footnote ⁶⁴

Though the roll-over crisis did not pose a serious threat and passed quickly, the experience added impetus to formulating three financial policy priorities during the 1650s and 1660s: conversion, redemption, and consolidation. The Estates tackled the first two issues in tandem and started exploring the options for reducing the interest burden through cutting the debt. In August 1652 Johan de Witt, on his way to Holland's most powerful office of grand pensionary, presented radical plans. The interest on redeemable annuities and bills would be reduced from 5 to 4 percent for at least one year, with the savings going into a redemption fund. This proposal caused such uproar that the subject had to be shelved. The ongoing war with Britain prevented debt reduction anyway, but to avoid a further increase the Estates raised the land tax by 50 percent and reintroduced a 0.5 percent wealth tax, revising the ledgers in 1654 to boost the yield.Footnote ⁶⁵

The war with England taught some harsh realities: Holland's financial position required further billholders' sacrifices. Wealth levies became a regular tax, raised in all but nine years until 1722. There was room for an interest rate cut, too, because the savings which had earlier absorbed the fast growing debt pushed interest rates down now that debt creation had stopped. Between 1650 and 1653 net annual transfers from Holland to investors amounted to 7 million guilders, subsequently dropping to around 6 million (Figure 4), money which, with the Dutch economic spurt slowing down, could not easily find alternative employment. Consequently private interest rates started to fall: to 4.5 percent by 1655, and less than 4 percent five years later (Figure 1). By 1660 Amsterdam's orphans' trustees (weesmeesters) lent at 3.5 to 4 percent, while the city's leading merchants borrowed at 3 to 3.5 percent.Footnote ⁶⁶ The VOC lowered its bond rates from 4 to 3.5 percent between 1662 and 1665.Footnote ⁶⁷

Falling interest rates combined with the war lessons to muster the political will for a billholders' sacrifice. In March 1655 bills were said to trade at 105–107 percent of par, the first instance we have of bill prices.Footnote ⁶⁸ The Estates used these high prices to force through an interest rate reduction to 4 percent in August, earmarking the saved expense for a redemption fund that would extinguish the debt in 40 years.Footnote ⁶⁹ Foreigners would be repaid first, then holders who had given notice in 1653, followed by investors living outside Holland, and in the last resort inhabitants proper.Footnote ⁷⁰ The profoundly unpopular redemption would thus hit the delegates and their constituents somewhat later but investors need not have worried. To achieve its goal of repaying the debt in 40 years, the province should have redeemed over 3 million guilders annually. The debt, however, declined by just that amount between 1655 and 1670, partly due to the fact that in 1664, with the clouds gathering for the Second Anglo-Dutch War (1665–1667) redemption funds were diverted into war spending.Footnote ⁷¹

After conversion and redemption, the Estates turned their attention to consolidation. During 1657 and 1658 they repeatedly discussed the wholesale repackaging of bills into annuities.Footnote ⁷² No decision was taken, presumably because delegates realized that, with nothing to attract investors and little to gain for the government, consolidation made no sense. Nor was it viable. In Gouda, the number of annuities transferred rose markedly during the 1650s and 1660s, reaching double figures for amounts totaling 6,000 to 20,000 guilders annually (Figure 5). After 1655 prices dipped under par on a few occasions: 98 percent, 96 percent, 93 even, rendering a conversion at par impossible. As a second best, the Estates reintroduced its old policy of selling annuities rather than bills, with marginal success.Footnote ⁷³

Thus Holland, though at the peak of its power, was trapped in a public finance system of its own making. By the 1650s the Estates depended on the market to realize its policies; consequently, the interest rate conversion succeeded, the redemption made as much headway as the international situation allowed, and the consolidation failed. Though the Estates managed to reduce debt, public finance had reached clear limits. Larger conflicts than those with Britain would require higher contributions from wealth owners. In effect, the high-level equilibrium between Holland and its creditors was really a trap for both, the sheer size of the debt tying both parties to each other hand and foot.Footnote ⁷⁴ Moreover, the Estates did not embrace the emerging secondary market as a means to lower interest costs, but actively disliked it, as their strenuous consolidation efforts show. For their part, the billholders were trapped, too, in their own savings. Interest payments boosted surplus savings, pushing down interest rates. We see here the origins of Holland as Europe's lender of first resort, offering such low interest rates that, from the late 1650s, foreign monarchs rushed in.Footnote ⁷⁵

COMING TO TERMS WITH THE MARKET, 1672–1678

The three-pronged attack by Britain, France, and two German princes in 1672 created entirely new financial parameters. Holland had begun preparing for an imminent conflict with France as early as 1670, raising troops, taxes, and loans.Footnote ⁷⁶ In October 1671 the Estates introduced a small roll-over charge on bills.Footnote ⁷⁷ Over the next four months, they boosted reserves with 9 million guilders worth of loans.Footnote ⁷⁸ By early April the Gecommitteerde Raden warned against further borrowing for fear of ruining credit, so the Estates chose to levy a 1 percent wealth tax instead.Footnote ⁷⁹ Then war broke out in earnest. Needing funds instantly, Holland resorted to forced loans. Revenues were disappointing and the Estates widened the assessment terms.Footnote ⁸⁰ Between 1672 and 1675 Holland raised 10 million guilders in free loans plus nearly 25 million guilders in forced loans, 7 percent of the estimated wealth in real estate and public securities.Footnote ⁸¹ Getting hold of any money at all proved difficult because the invasion triggered a scramble for cash.Footnote ⁸² After some deliberation, the Estates allowed receivers to accept bullion and gold and silver objects in payment, hoping to avoid investors pressing for cash through redemption.Footnote ⁸³ This happened anyway; by June the receivers were in trouble from billholders giving notice. On 2 June provincial officials intervened and raised the terms for giving notice.Footnote ⁸⁴ This failed to stop the run, so on June 13 the Estates declared a moratorium on redemptions, though not on interest payments.Footnote ⁸⁵ Even so all receivers fell in arrears.Footnote ⁸⁶ In 1675 interest payments on the forced loans of 1672 were still overdue, the Estates were also concerned about servicing the 1674 forced loans. As a result, they briefly entertained the notion of cancelling the 1672 arrears outright.Footnote ⁸⁷ Holland also paid interest due in bills rather than money and persuaded the VOC to pay dividends in public bills rather than cash.Footnote ⁸⁸

The war thus threw Holland in its deepest financial crisis since the late sixteenth century. The Estates proposed drastic measures such as renewing the existing wealth tax ledgers and widening the assessment to include all financial assets. After intense negotiations they adopted a blanket 16.6 percent increase of the estimated wealth for tax payers worth more than 3,000 guilders using updated ledgers. The Estates also decided to levy two wealth taxes of 0.5 percent each specifically applied to financial assets, thus effectively reducing nominal interest rates on Holland bills to 3 percent in 1673. The impact of this latter measure was softened somewhat by introducing a dual taxation rule. Sums due in the financial assets levy could be subtracted from sums due in personal wealth tax.Footnote ⁸⁹ This benefitted big investors to the detriment of small ones, people below the tax threshold with a redeemable or life annuity as a pension provision. The sharp drop in Gouda prices from 1673 followed by marked fluctuations and increasingly high annual turnovers shows that this caused a rush to sell redeemable annuities (Figures 5 and 6). The liquidity squeeze thus breathed life into the rudimentary local secondary markets that had begun in the 1650s and 1660s.

Why did a secondary market not emerge earlier? After all, with 130 million guilders of debt there will always be investors who need some liquidity. The receivers probably served this market, redeeming bills for later resale.Footnote ⁹⁰ In 1666 and 1668 the Amsterdam merchant Louys Trip purchased such bills.Footnote ⁹¹ Billholders could also obtain liquidity by offering their securities as loan collateral (lombarding).Footnote ⁹² The 1672 squeeze upset the balance between supply and demand, though the Gouda data do not tell us how active that market was. All the same it bears emphasis that the secondary market did not emerge on purpose, or as a consequence of diverging investors' expectations, but as a consequence of the friction between aggregate supply and demand, just as the secondary market in VOC shares had originated in sustained demand after the closure of the company's subscription books.Footnote ⁹³

The Gouda data also show that investors discounted the securities tax in prices, leading private and public rates to diverge. At the start of the war, private interest rates in Amsterdam shot up, but once the immediate danger had disappeared, they declined again (Figure 1). A small portfolio of 67 loans extended by one Amsterdam merchant banker on collateral of VOC shares between 1673 and 1683 shows a sharply declining interest rate, the lowest one dropping from 3.5 percent in 1677 to as little as 2.5 percent in 1683.Footnote ⁹⁴ Public rates followed a quite different trajectory. During the summer of 1672 the yield on Holland's securities also peaked as sellers could find no buyers. The available bill data for 1672 suggest that prices plummeted to 30 percent, recovering to the 70–85 range during the second half of the year, then dropping again to 50–55 in late December before returning to 76 early in 1673.Footnote ⁹⁵ Low securities prices directly translated into higher borrowing costs. In 1673 the cities had to borrow their usual advance on the forced loans at 6 percent per year.Footnote ⁹⁶ The market situation then eased somewhat, but with bill prices below 80 Holland had to offer at least 5 percent for free loans. Considering this too high, the Estates continued issuing forced loans at 4 percent. That proved difficult and in 1674 receivers were once again warned not to sell bills below par, which they obviously were doing.Footnote ⁹⁷As late as 1678 the Rotterdam receiver could raise money only by offering 5 percent.Footnote ⁹⁸ By now players in the secondary market held the whip hand. During the mid-1650s bill prices had performed no more than a signaling function for delegates to the Estates, helping them to decide whether or not to lower interest rates; from the 1670s, however, the Estates had to follow the market rate, whether they wanted to or not.

THE DOUBLE-EDGED SCALPEL, 1678–1713

The 1678 Peace of Nijmegen enabled Holland to restore its finances. With a debt-to-tax rate of fourteen, the situation was worse than before De Witt's efforts of the 1650s and 1660s (Figure 3). The circumstances of the 1650s returned, but with a difference. The province could not redeem large sums, so instead it raised the rate of both the excises and the real estate tax. The extra revenues helped to push the debt-to-tax rate to 8.5 by 1690, its lowest level for more than 40 years (Figure 3). Holland also reduced expenses by forcing an interest rate cut in the form of a withholding tax. In 1680 the Estates imposed a 0.5 percent wealth levy. Public securities were taxed at 0.5 percent of the face value, deducted from interest payments by the receivers.Footnote ⁹⁹ A similar withholding tax on interest payments had been used between 1596 and 1607, but opposition had prevented subsequent attempts to levy it. In 1634, for instance, the Estates had rejected a proposal for taxing interest payments on bills and on other loans because credit would collapse if private administrations were opened for inspection.Footnote ¹⁰⁰ The Estates considered a withholding tax in 1655, but abandoned it in favor of an interest rate cut. In December 1671, only months before Louis XIV's invasion, De Witt had great difficulty in persuading the Estates to enact a temporary 5 percent levy on interest payments.Footnote ¹⁰¹

The 1680 withholding tax was also discontinued, presumably because of its immediate impact on credit. After a steady recovery during the late 1670s, Gouda and Leiden annuities prices halted in 1681, drifted for a year or two as if testing the Estates' intentions, then resumed their rise back to par (Figure 6). Clearly investors wanted compensation for their sacrifices during the war before being amenable to an interest rate cut. With the swell of interest payments not absorbed by new debt issues running at some 7 million guilders a year, that moment would come more or less automatically (Figure 4). Modest redemptions helped to prepare the ground; fiscal reforms succeeded in generating the surplus needed for about 2.6 million guilders of redemptions during the 1680s.Footnote ¹⁰² In 1687 the Estates reintroduced the 0.5 percent withholding tax on public debt interest payments. The rate was doubled to 1 percent the following year; again the dual taxation rule cushioned big investors. Compared to the blunt knife of the ledger-based wealth levies with their built-in obsolescence, the withholding tax was a scalpel, easier to impose, more efficient, and cheaper to apply. Moreover, the Estates could use the withholding tax rate to test the market's readiness for new issues, lowering or raising the interest differential between existing bills and new issues as required. The adoption of this tax also signaled that the Estates considered bills a fair proxy for overall wealth, i.e., ownership was spread sufficiently wide over wealth owners for the tax to result in an equitable assessment. The moment proved well chosen; Gouda-Leiden prices hardly moved, so the market accepted the interest rate cut (Figure 6).

FIGURE 6

THE ANNUAL AVERAGE PRICE OF HOLLAND'S LOSRENTEN IN GOUDA AND LEIDEN AND THE VALUE OF LOSRENTEN SOLD AS A PERCENTAGE OF THE TOTAL DEBT OUTSTANDING THERE, 1670–1720

Source: Gouda City Archives, Oud-Rechterlijk Archief, Inv. Nrs. 482–493; and Leyden City Archives ORA 71–73.

The withholding tax was also instrumental in funding the quarter century of almost continuous warfare which followed William of Orange's ascension of the English throne in 1689. During this period, Holland's debt nearly doubled from 160 million guilders in 1690 to almost 300 million at the Peace of Utrecht in 1713, pushing the debt-to-tax ratio back to over fourteen. To fund the wars, the Estates resorted to their customary round of substantial tax raises, including the real estate tax, the excises, and a doubling of the withholding tax on interest payments to 2 percent. Consequently, from 1690 investors suffered an annual haircut of a similar magnitude to the forced loans of the 1670s. Billholders are said to have borne this stoically because countering the French threat required sacrifices, but they did not have much of an option. Home alternatives such as buying real estate suffered a similar tax pressure on yields and foreign alternatives like British securities were slow to materialize.Footnote ¹⁰³

Moreover, the annuity prices from Gouda and Leiden show an immediate impact of taxation on the price of public debt (Figure 7). The market accepted the initial reductions of 1687–1690, but prices then dropped to 80 percent in 1695. The Estates responded by halving the withholding tax to 1 percent following the end of the Nine Years' War in 1698, causing annuities prices to veer back. The reimposition of a 2 percent levy at the start of the War of the Spanish Succession in 1702 clearly disappointed investors and prices fell gradually to reach 75 percent in 1712, giving a 3 percent yield. By that time, even a 0.5 percent tax cut failed to restore prices.

FIGURE 7

THE NOMINAL INTEREST RATE AND THE YIELD ON HOLLAND'S LOSRENTEN, 1650–1720

Source: Liesker and Fritschy, Gewestelijke financiën, pp. 368–71; Tax incidence 1687–1698 based on NA 3.01.29, Inv. Nr. 802, “Memorie van de Extraordinaris Consenten in Holland”; Gouda City Archives, Oud-Rechterlijk Archief, Inv. Nrs. 482–493; and Leyden City Archives ORA 71–73.

The price drop shows Holland caught in a vicious circle of its own making. Its scalpel cut both ways; to raise the very large loans needed, the Estates had to offer incentives which drove down the price of old debt to the same degree and then necessitated new incentives. For example, at first Holland sold new issues at a tax-exempt 4 percent for the duration of the war. This period had to be extended to ten years after the war for later issues.Footnote ¹⁰⁴ When the market for 4 percent bills appeared exhausted, the Estates adopted a successful British public debt innovation, lottery loans, issuing lottery bills totaling 34 million guilders in 1711, 1712, and 1713.Footnote ¹⁰⁵ During 1710 and 1711 Holland also issued twenty-year bills at 9 percent, of which 4 percent interest and 5 percent amortization.Footnote ¹⁰⁶ The entire volume issued during 1690–1712 consisted of loans incentivized in one way or another (Figure 2). In 1709 sluggish home demand even led the Estates to take a 1.5 million 4 percent loan from the Swiss canton of Bern.Footnote ¹⁰⁷

Under these circumstances, the receivers needed all their skills to coax local markets into accepting new issues while holding on to old, lower-yielding bills. They probably did this by pretending to make exclusive offers to selected customers, which they could because the primary market remained opaque. The Estates published no financial information whatsoever; only the receivers knew how much money was to be raised or still needed to be subscribed, so they could play investors against each other. Moreover, despite negative transfers during the early 1700s (Figure 4), the mismatch between investment opportunities and surplus savings continued. Though old bills sold under par on the secondary market, there was no rush to redeem them at par with the receivers because investors feared being excluding from new issues at higher rates that could raise the overall yield of their portfolios. This gave the receivers considerable leverage. New loans were probably fed to investors according to their status with the receiver, favorites being able to buy new, untaxed issues, perhaps even be allowed to swap old, taxed bills for new. Less privileged investors would have needed to bargain, guarded threats of reimbursement serving to coax them into accepting whatever it pleased the receiver to offer them. In 1706 the French diplomat Adrianus Helvétius reported admiringly from The Hague on the skills with which the Receiver-General Receiver of the United Provinces, Cornelis de Jonge van Ellemeet, exploited his market to get new issues subscribed.Footnote ¹⁰⁸

Consequently, by the Peace of Utrecht in 1713 Holland's bill debt, homogeneous in 1687, had become a patchwork of different issues. Rather than preventing the rise of a secondary market, as Larry Neal surmised, this was actually the consequence of pressure from the secondary market.Footnote ¹⁰⁹ At this moment in time, Britain struggled for a secondary market to absorb its short-term debt overhang and ended up creating liquidity by way of securitization, with the 1720 boom and bust as a direct consequence.Footnote ¹¹⁰ By contrast, Holland had no such need and thus avoided the worst of the 1720 excesses.Footnote ¹¹¹ The key difference between the two countries would seem to have resided in Holland's fiscal decentralization, the system of feeding debt directly to local markets in proportion to their tax yield, rather than issuing everything from the center.Footnote ¹¹² In this way, investors could be targeted, tax receipts matched with interest payments, debt issues tailored to available savings, and trust maintained by leaving local people in charge. Britain's public finance trajectory was a particular one, determined by its particular circumstances, and not by any general law of public finance development. This particularity explains why public finance best practices could not easily be transplanted, why the “Dutch Finance” imported by William III into Britain did not really work there, and why so few European countries adopted British fiscal practices even during the nineteenth century.Footnote ¹¹³ In public finance, each country needed to develop its own wheel, no doubt inspired by foreign examples, but always adapted to local circulation.