The UK’s unpaid war debts to the United States, 1917-1980

by David James Gill (University of Nottingham)

Trenches in World War I. From <>

We all think we know the consequences of the Great War – from the millions of dead to the rise of Nazism – but the story of the UK’s war debts to the United States remains largely untold.

In 1934, the British government defaulted on these loans, leaving unpaid debts exceeding $4 billion. The UK decided to cease repayment 18 months after France had defaulted on its war debts, making one full and two token repayments prior to Congressional approval of the Johnson Act, which prohibited further partial contributions.

Economists and political scientists typically attribute such hesitation to concerns about economic reprisals or the costs of future borrowing. Historians have instead stressed that delay reflected either a desire to protect transatlantic relations or a naive hope for outright cancellation.

Archival research reveals that the British cabinet’s principal concern was that many states owing money to the UK might use its default on war loans as an excuse to cease repayment on their own debts. In addition, ministers feared that refusal to pay would profoundly shock a large section of public opinion, thereby undermining the popularity of the National government. Eighteen months of continued repayment therefore provided the British government with more time to manage these risks.

The consequences of the UK’s default have attracted curiously limited attention. Economists and political scientists tend to assume dire political costs to incumbent governments as well as significant short-term economic shocks in terms of external borrowing, international trade, and the domestic economy. None of these consequences apply to the National government or the UK in the years that followed.

Most historians consider these unpaid war debts to be largely irrelevant to the course of domestic and international politics within five years. Yet archival research reveals that they continued to play an important role in British and American policy-making for at least four more decades.

During the 1940s, the issue of the UK’s default arose on several occasions, most clearly during negotiations concerning Lend-Lease and the Anglo-American loan, fuelling Congressional resistance that limited the size and duration of American financial support.

Successive American administrations also struggled to resist growing Congressional pressure to use these unpaid debts as a diplomatic tool to address growing balance of payment deficits from the 1950s to the 1970s. In addition, British default presented a formidable legal obstacle for the UK’s return to the New York bond market in the late 1970s, threatening to undermine the efficient refinancing of the government’s recent loans from the International Monetary Fund.

The consequences of the UK’s default on its First World War debts to the United States were therefore longer lasting and more significant to policy-making on both sides of the Atlantic than widely assumed.


Decimalising the pound: a victory for the gentlemanly City against the forces of modernity?

by Andy Cook (University of Huddersfield)


1813 guinea

Some media commentators have identified the decimalisation of the UK’s currency in 1971 as the start of a submerging of British identity. For example, writing in the Daily Mail, Dominic Sandbrook characterises it as ‘marking the end of a proud history of defiant insularity and the beginning of the creeping ­Europeanisation of ­Britain’s institutions.’

This research, based on Cabinet papers, Bank of England archives, Parliamentary records and other sources, reveals that this interpretation is spurious and reflects more modern preoccupations with the arguments that dominated much of the Brexit debate, rather than the actual motivation of key players at the time.

The research examines arguments made by the proponents of alternative systems based on either decimalising the pound, or creating a new unit worth the equivalent of 10 shillings. South Africa, Australia and New Zealand had all recently adopted a 10-shilling unit, and this system was favoured by a wide range of interest groups in the UK, representing consumers, retailers, small and large businesses, and media commentators.

Virtually a lone voice in lobbying for retention of the pound was the City of London, and its arguments, articulated by the Bank of England, were based on a traditional attachment to the international status of sterling. These arguments were accepted, both by the Committee of Enquiry on Decimal currency, which reported in 1963, and, in 1966, by a Labour government headed by Harold Wilson, who shared the City’s emotional attachment to the pound.

Yet by 1960, the UK had faced the imminent prospect of being virtually the only country retaining non-decimal coinage. Most key economic players agreed that decimalisation was necessary and the only significant bone of contention was the choice of system.

Most informed opinion favoured a new major unit equivalent to 10 shillings, as reflected in evidence given by retailers and other businesses to the Committee of Enquiry on Decimal Coinage, and the formation of a Decimal Action Committee by the Consumers Association to press for such a system.

The City, represented by the Bank of England, was implacably opposed to such a system, arguing that the pound’s international prestige was crucial to underpinning the position of the City as a leading financial centre. This assertion was not evidence-based, and internal Bank documents acknowledge that their argument was ‘to some extent based on sentiment’.

This sentiment was shared by Harold Wilson, whose government announced the decision to introduce decimal currency based on the pound in 1966. Five years earlier, he had made an emotional plea to keep the pound arguing that ‘the world will lose something if the pound disappears from the markets of the world’.

Far from being the end of ‘defiant insularity’, the decision to retain a higher-value basic currency unit of any major economy, rather than adopting one closer in value either to the US dollar or the even lower-value European currencies, reflected the desire of the City and government to maintain a distinctive symbol of Britishness, the pound, overcoming opposition from interests with more practical concerns.

THE FINANCIAL POWER OF THE POWERLESS: Evidence from Ottoman Istanbul on socio-economic status, legal protection and the cost of borrowing

In Ottoman Istanbul, privileged groups such as men, Muslims and other elites paid more for credit than the under-privileged – the exact opposite of what happens in a modern economy.

New research by Professors Timur Kuran (Duke University) and Jared Rubin (Chapman University), published in the March 2018 issue of the Economic Journal, explains why: a key influence on the cost of borrowing is the rule of law and in particular the extent to which courts will enforce a credit contract.

In pre-modern Turkey, it was the wealthy who could benefit from judicial bias to evade their creditors – and who, because of this default risk, faced higher interest rates on loans. Nowadays, it is under-privileged people who face higher borrowing costs because there are various institutions through which they can escape loan repayment, including bankruptcy options and organisations that will defend poor defaulters as victims of exploitation.

In the modern world, we take it for granted that the under-privileged incur higher borrowing costs than the upper socio-economic classes. Indeed, Americans in the bottom quartile of the US income distribution usually borrow through pawnshops and payday lenders at rates of around 450% per annum, while those in the top quartile take out short-term loans through credit cards at 13-16%. Unlike the under-privileged, the wealthy also have access to long-term credit through home equity loans at rates of around 4%.

The logic connecting socio-economic status to borrowing costs will seem obvious to anyone familiar with basic economics: the higher costs of the poor reflect higher default risk, for which the lender must be compensated.

The new study sets out to test whether the classic negative correlation between socio-economic status and borrowing cost holds in a pre-modern setting outside the industrialised West. To this end, the authors built a data set of private loans issued in Ottoman Istanbul during the period from 1602 to 1799.

These data reveal the exact opposite of what happens in a modern economy: the privileged paid more for credit than the under-privileged. In a society where the average real interest rate was around 19%, men paid an interest surcharge of around 3.4 percentage points; Muslims paid a surcharge of 1.9 percentage points; and elites paid a surcharge of about 2.3 percentage points (see Figure 1).


What might explain this reversal of relative borrowing costs? Why did socially advantaged groups pay more for credit, not less?

The data led the authors to consider a second factor contributing to the price of credit, often taken for granted: the partiality of the law. Implicit in the logic that explains relative credit costs in modern lending markets is that financial contracts are enforceable impartially when the borrower is able to pay. Thus, the rich pay less for credit because they are relatively unlikely to default and because, if they do, lenders can force repayment through courts whose verdicts are more or less impartial.

But in settings where the courts are biased in favour of the wealthy, creditors will expect compensation for the risk of being unable to obtain restitution. The wealth and judicial partiality effects thus work against each other. The former lowers the credit cost for the rich; the latter raises it.

Islamic Ottoman courts served all Ottoman subjects through procedures that were manifestly biased in favour of clearly defined groups. These courts gave Muslims rights that they denied to Christians and Jews. They privileged men over women.

Moreover, because the courts lacked independence from the state, Ottoman subjects connected to the sultan enjoyed favourable treatment. Theory developed in the new study explains why their weak legal power may translate into strong financial power.

More generally, this research suggests that in a free financial market, any hindrance to the enforcement of a credit contract will raise the borrower’s credit cost. Just as judicial biases in favour of the wealthy raise their interest rates on loans, institutions that allow the poor to escape loan repayment – bankruptcy options, shielding of assets from creditors, organisations that defend poor defaulters as victims of exploitation – raise interest rates charged to the poor.

Today, wealth and credit cost are negatively correlated for multiple reasons. The rich benefit both from a higher capacity to post collateral and from better enforcement of their credit obligations relative to those of the poor.


To contact the authors:
Timur Kuran (; Jared Rubin (

How did investors view the reforms and supervisory organisations of the late nineteenth century?

by Avni Önder Hanedar (Sakarya University)

In the last couple of decades, high debt burden in emerging economies created financial crises and the low growth rate during the 2008 financial crisis led to a default problem for Greece. Some reforms were proposed, such as institutional changes and the establishment of an entity under control of the other Eurozone members to supervise the repayment of debts. These events have some similarities with the default of the Ottoman Empire and the establishment of the Ottoman Public Debt Administration (OPDA) (Düyun-u Umumiye). To deal with the inefficiencies in the Ottoman economy and political system, reforms were implemented, as supervisory organizations were established during the nineteenth century. Important ones were the adoption of the gold standard in 1880, the Administration of Six Indirect Revenues (Rüsum-u Sitte) (ASIR) in 1879, and the OPDA in 1881. It seems that many of them were not seen by investors as promising, since a British weekly magazine, Punch or The London Charivari, illustrated these events as bubbles. A paper of  Elmas Yaldız Hanedar, Avni Önder Hanedar, and Ferdi Çelikay examined how such events were perceived at the İstanbul bourse, which could shed light on today’s realities.

Cartoon of Punch or The London Charivari on 6 January 1877 about the Ottoman reforms.a caption


The paper manually collected historical data on the price of the General Debt bond traded at the İstanbul bourse between 1873 and 1883 from volumes of daily Ottoman newspapers, i.e., Basiret, Ceride-i Havadis, and Vakit. This bond was the most actively traded one at the İstanbul bourse in 1881, during the foundation of the OPDA.

A column of Vakit pointing out the values of bonds, stocks, and foreign currencies at the İstanbul bourse on 6 October 1875 (Vakit. (6 October 1875). Sarafiye, Galata piyasası, 2)

The paper is the first to measure in econometrically sophisticated manner investors’ beliefs at the İstanbul bourse in reference to the reforms and financial control organizations. Historical research does not include detailed empirical information for the effects of reforms and financial control organizations on the İstanbul bourse during the default period. Using unique data on the most actively traded Ottoman government bond, the paper extends the historical literature on the İstanbul bourse (See Hanedar et al. (2017)) and reforms (See Mauro et al. (2006), Birdal (2010), Mitchener and Weidenmier (2010) looking at bond markets in multiple developing countries, with samples that include the Ottoman Empire).

The methodology in the paper was to analyse the variance of returns (derived from the price showed in above) as a proxy of financial instabilities and risks. To model volatility, the paper estimated a GARCH model with dummy variables for reforms and financial control organizations at and after the dates of the events (i.e., short- and long-run).







The General Debt bond price (Turkish Liras) and key events. The data are derived from Vakit, Ceride-i Havadis, and Basiret, 187383.

The empirical results indicated a permanent decrease in volatility after the establishment of the OPDA and the gold standard. The foundation of a locally controlled finance commission in 1874 was correlated with a lower volatility level at the date of the event, but increased volatility in the long term. The Ottoman case is instructive for the understanding of today’s economic situation in emerging markets such as Greece, while it could be argued that long-lived and comprehensive measures with foreign creditors’ supervision on fiscal and monetary systems matter more for investors’ perceptions. Lowering government interventions on economic system and transaction costs due to bimetallism were viewed as promising. Investor beliefs that the local and short-lived reforms and supervisory organizations were ineffective could be due to several factors such as lack of measures to limit public expenditures.


Volatility changes in the General Debt bond return, 1873–83. * and *** denote statistically significant coefficients at 10% and 1%.



Vakit. (6 October 1875). Sarafiye, Galata piyasası, 2.

Birdal, M. (2010). The Political economy of Ottoman public debt, insolvency and European control in the late nineteenth century. London: I. B. Tauris and Co Ltd.

Hanedar, A. Ö., Hanedar, E. Y., Torun, E., & Ertuğrul, H. M. (2017). Dissolution of an Empire: Insights from the İstanbul Bourse and the Ottoman War Bond. Defence and Peace Economics, (Forthcoming).

Mauro, P., Sussman, N., & Yafeh, Y. (2006). Emerging markets and financial globalization: Sovereign bond spreads in 1870-1913 and today. Oxford: Oxford University press.

Mitchener, K. J. & Weidenmier, M. D. (2010). Super sanctions and sovereign debt repayment. Journal of International Money and Finance, 29(1), 19–36.

A Brief Monetary History of Ireland

by Seán Kenny (Lund University) and Jason Lennard (Lund University and National Institute of Economic and Social Research)


The Irish Famine of the 1840s is one of the great tragedies of history. Beginning with a bout of potato blight, the Irish population subsequently declined by 20 per cent between the censuses of 1841 and 1851 and has never recovered (O’Rourke, 1991). How did an agricultural shock have such devastating effects? Lynch and Vaizey (1960) argue that a lack of monetization facilitated self-dependence and barter, leaving the Irish economy vulnerable to exogenous shocks like the Famine.

In a forthcoming paper in the Economic History Review available here, we constructed new monthly estimates of the narrow money supply and annual estimates of the broad money supply between 1840 and 1921. The aggregates were constructed from a range of archival sources and contemporary publications. A major task was to reconstruct the Irish coin supply. We did this by tracking shipments of coin between the Royal Mint and Irish banks using records held at the National Archives. These flows were then added to stocks, which were either recalculated from contemporary estimates or based on recoinages.

A number of interesting results emerge from the data. First, we find that, by standard measures, Ireland was no backwater, but well monetized on the eve of the Famine. Not only was it more monetized than other European countries for which data is available, such as Norway and Sweden, it was decades ahead of others, such as Germany and the Netherlands. The new data is therefore at odds with the Lynch and Vaizey hypothesis.

A second major finding is the scale of the collapse in the money supply during the Great Famine. This monetary contraction was the largest during any event in the economic history of Ireland since 1840 and perhaps one of the deepest in economic history more generally. Currency in the hands of the public, the nation’s liquidity, collapsed by more than half, the monetary base (currency in the hands of the public plus reserves) by 48 per cent and the broad money supply (currency in the hands of the public plus net deposits) by 27 per cent.

Figure 1. The Great Famine versus the Great Contraction
Notes: Ireland indexed to 1846 = 100. US indexed to 1928 = 100. Year end.
Sources: Kenny and Lennard, ‘Monetary aggregates for Ireland’; Friedman and Schwartz, Monetary history.


Figure 1 plots the narrow (M0) and broad (M3) money supplies in Ireland during the Great Famine against equivalent measures for the United States during the Great Depression. As can be seen in this tale of two crises, the narrow money supply slumped much deeper during the Great Famine than in the Great Contraction. The broad money supply initially declined more steeply in Ireland than in the US. However, the Irish recovery was underway from 1849, while the American contraction continued until 1933.

This new data shines a light on Ireland’s statistical Dark Age, allowing us to revisit old hypotheses and others to develop new ones. On the monetary origins of the Great Famine, we found that Ireland was no less monetized than its European peer group. The Famine did, however, unleash the Great Irish Contraction, during which the money supply drastically slumped.


To contact the authors:



Friedman, M. and Schwartz, A. J., A monetary history of the United States, 1867–1960 (Princeton, NJ, 1963).

Kenny, S. and Lennard, J., ‘Monetary aggregates for Ireland, 1840–1921’, Economic History Review (2017).

Lynch, P. and Vaizey, J., Guinness’s brewery in the Irish economy, 1759–1876 (1960).

O’Rourke, K. H., ‘Did the Great Irish Famine matter?’, Journal of Economic History, 51 (1991), pp. 1–22.

EHS 2018 special: Long-term effects of financial crises

by Chenzi Xu (Harvard University)


The global financial crisis of 2008 was not unique. It had a precedent in the London banking crisis of 1866. Just as in 2008, the crisis began in the core financial market and spread to the periphery, the same happened in 1866.

The 1866 crisis has only been studied as a purely British event, but my research presents new evidence that it was a global financial crisis on the scale of 2008’s. The cities around the world that depended on British banks that happened to fail in London suffered immediate losses in exports activity. These losses took decades to recover, with the hysteresis persisting until the twentieth century.

In May 1866, Overend and Gurney, a bank’s bank and one of the most prestigious financial entities in London, declared bankruptcy. A panic erupted and almost 20% of banks headquartered in London failed. Crucially, these banks had been established in the mid-nineteenth century to globalise financial markets and trade, and they operated in cities around the world.

Figure 1. Map of British credits and failures around the world


Figure 1 shows the concentration of British banking, and the degree to which the banking crisis in London affected them. Red denotes greater losses in British financing, and the size of the circles denotes the amount of lending before the crisis.

I study the impacts of the failures of British banks in London on trade activity around the world, outside of the UK, at hundreds of ports. At the extreme, losing access to all British credit caused exports to drop 80% in the year following the crisis. The aggregate global loss in trade was 17%, which is comparable to the levels seen in the latest crisis. Given that the mid-late 19th century was otherwise a period of great expansion and growth, the counterfactual without this crisis would have been even more spectacular.

The historical context also makes it possible to study the long-run effects, and I find that countries suffering the largest drops in the supply of British credit did not recover their exports to previous partners for several decades. These persistent effects suggest that losing access to financial markets can cause substantial hysteresis.

These long-term consequences of financial market instability have yet to be established for recent crises simply because not enough time has passed. But early evidence suggests that international trade has not rebounded, even ten years after the financial crisis.


by Nuno Palma (University of Manchester)

This paper provides the first annual time series of coin and money supply estimates for about six hundred years of English history.

It presents a baseline set of estimates, but also considers a variety of alternative plausible scenarios and provide several robustness checks. It concentrates on carefully setting out the details for the data construction, rather than on analysis, but the hope is that these new estimates – the longest such series ever assembled, for any country – will open new vistas to help us understand the complex interaction between the real and the monetary sides of the English economy, at both business-cycle and long-run frequencies. Many applications are possible; for instance, O’Brien and Palma (2016) use it in their analysis of the Restriction period (1797-1821). Furthermore, the new methodology set out here may serve as a blueprint for a similar reconstruction of coin and money supply series for other economies for which the analogous required data is available.

The paper proposes two new estimation methods. The first, referred to as the “direct method”, is used to measure the value of government-provided, legal-tender coin supply only. This method does not consider broader forms of money such as banknotes, deposits, inland bills of exchange, government tallies, exchequer paper or private tokens, which became increasingly important from the seventeenth century onwards. The second method is an “indirect method,” which relies on a combination of information about nominal GDP with the value of coin supply or M2 known at certain benchmark periods. This permits estimating the volume of a broader measure of money supply over time. Figure 1 shows the main results.

Figure 1. English nominal coin supply, 1270-1870 (log scale of base 2). The periods when direct method A cannot be seen means it coincides with the baseline method (aka direct method B). Source: my calculation based on a series of sources; see text for details.

This paper is forthcoming in The Economic History Review (currently available in early view), and the underlying data has now been included by the Bank of England in their historical database

To contact the author:

The Public Works Loan Board and the growth of the state in nineteenth century England

by Ian Webster

The Public Works Loan Board was formed in 1817, when the government was faced with a stagnant economy and rising unemployment after the Napoleonic wars. It was established to lend money to finance public works like road, bridge and canal building. Later in the nineteenth century, the PWLB became a major lender to local government to finance the building of workhouses, schools, sewers and water supply facilities. The PWLB still exists in the twenty first century, and is the major provider of loans to local government.

During the nineteenth century, the PWLB survived two attempts by chancellors of the exchequer to abolish it. Sometimes its decisions were overruled by Parliament which directed the PWLB to make loans which were unlikely to be repaid. The Treasury preferred to see the PWLB as a high-cost ‘last resort’ lender when the private sector wouldn’t lend. But the prevailing view of the PWLB was as a low-cost lender to reduce the cost of national public health and education policies to local ratepayers.

These debates continue today. A recent chancellor of the exchequer sought to increase the PWLB’s interest rates closer to market rates, in order to encourage more private sector lending. He also proposed the abolition of the PWLB as a body of commissioners. There is still a debate about the merits of government borrowing to improve public infrastructure.

PWLB profits and losses 1817-76
Sums lent Profits(losses)
£M £M %
Lending decisions made independently of Parliament 37.9 3.4 9%
Lending decisions made by Parliament 4.2 (2.3) (55%)
Totals 42.1 1.1 3%


The research reached three main conclusions. First, 90 per cent of the PWLB’s lending was profitable, in spite of the fact that most loans were made at below market rates of interest. The critical factor is that lending decisions were made independently and with a prime concern about the security of the loan. The remaining 10 per cent of loans were made at the direction of Parliament. In these cases, social or economic reasons overcame the PWLB’s concern about repayment, and large losses resulted. Second, seeing the PWLB as a low-cost loan provider was a victory for local interests and national spending departments, over the Treasury desire to minimise the national debt. Third, the story of the PWLB highlights five key decisions between 1859 and 1876 that contributed to the substantial growth in government activities in the late nineteenth century. Without the PWLB’s cheap loans, it would have taken longer for elementary education and constant clean water supplies to become universal services.

To contact the author:

‘Quakers, Coercion and pre-modern Growth: Why Friends’ Formal Institutions for Contract Enforcement Did Not Matter for Early Modern Trade Expansion’

by Ester Sahle (University of Bremen)

barclays_bank_limited_signIn the wake of the Libor scandal in 2012, Barclay’s bank suffered severe reputational damage. In response, its CEO promised a return to the bank’s Quaker roots. With this he referred to Barclay’s history as a Quaker-founded bank, and the proverbial Quaker honesty. The idea of the honest Quaker businessman is part of popular culture and historians have argued that honesty in business was an inherent trait of Quakerism from its beginnings.

The Society of Friends, learned opinion would have it, disowned culpable bankrupts. Thereby, it created an incentive for Friends to be honest in their conduct of business. The empirical basis for these claims however is curiously thin. The literature cites few actual instances of disownments for business-related offences from the seventeenth and eighteenth centuries. Most known cases stem from the nineteenth century, when this was indeed common practice. The story of Quaker business honesty is thus based on a strong assumption of institutional and cultural continuity.

The Library of the Society of Friends holds records of London Quaker meetings dating back to the 1660s, when Friends first appeared in the capital. Consulting Quaker meetings’ minutes, disciplinary records, as well as journals and letters of London Quaker businessmen, I conducted the first large scale empirical study of London Quaker meeting’s attitudes towards debt and bankruptcy, c.1660 – 1800.

Surprisingly, these meetings rarely sanctioned business offenders prior to the 1750s. For about 100 years after its conception, the Society of Friends showed no particular interest in its members’ conduct of business. What is more, the letters and diaries of Quaker businessmen in this period contain no evidence that that they feared repercussions from the Society. Quaker businessmen in financial difficulties discussed their impending bankruptcy procedures, or fear of being incarcerated for debt. The possibility of disownment from the Society however, did not figure among their concerns. This indicates that the punishment of offenders was not common enough to work as a deterrence.

From the 1750s onwards, however, this changed. Numbers of disownments for business-related offences skyrocketed. The last decades of the eighteenth century saw far more disownments for business-related offences than the 100 years before.

What caused this change? The new emphasis on honesty in business was part of the Quaker reformation, a movement within Quakerism which refocused the sect’s ideals. Reform movements within religious denominations are not uncommon, what set the Quaker reformation apart was its stated emphasis on protecting the Society’s reputation, and focus on business conduct.

These priorities were a response to a political crisis of the 1750s, which took place in the Quaker-founded colony of Pennsylvania. Erupting over internal disagreements about who was to cover the expenses for the colony’s defense during the Seven Years War, it led to a public scandal which shook Quakerism across the Atlantic World. Contemporary media accused the Quakers of failing to protect the colony’s population from French soldiers and native American raiders. Quaker politicians supposed motivation, their pacifist doctrine was merely a mask for selfish greed. Pamphlets published in London attacked individual Quaker businessmen as war profiteers, who were accumulating fortunes at the expense of the lives of innocent civilians.

In other words, just like Barclay’s Bank in the 21st century, the mid-eighteenth century Quakerism suffered severe reputational damage. The sect’s new focus on honesty in business was a response to this. The Society of Friends conducted an exercise of corporate responsibility, which was a tremendous success – so successful that 250 years later, Quakerism and honesty remain inseparable in the minds of lay people and Historians alike.

Friends went on to become leaders in important ethical concerns, such as the abolition of the slave trade. Today, the Society of Friends indeed stands for an exceptional ethical approach to many areas of public life. What this story tells us is that taking action against reputational damage can lead to institutional change. And institutions shape culture. In other words, corporate social responsibility can indeed lead to a better conduct of business, to the benefit of society as a whole.


What is a market crash?

by David Le Bris (Toulouse Business School)


Despite the importance of the phenomenon, there is no clear definition of what is a market crash. Arguably, market crashes should be related to important news but it is frequently difficult to effectively match historical events with market reactions. For instance, when WWI started in July 1914, the French stock index decreased by a modest 7.14 %; a monthly drop ranked only the 105th in the French stock market history. But, a given fall in percent has a stronger impact on a stable market than it does upon a highly volatile one. A crash is not solely a given percentage decrease but represents a significant discrepancy compared to what has been previously observed.

Thus, crashes need to be identified after having taken into account the prior financial context. I propose a simple new tool to identify market crashes by measuring price variations in numbers of standard deviations of the preceding period rather than in percent. French stock market was used to a low volatility before 1914, thus the modest decrease of 7.14 % represents a fall of 6.09 standard deviations, which is the second worst case in French history. This ranking is much more consistent with history.

In a paper (forthcoming in Economic History Review), this method is applied to long term series of US and French stock prices and UK state bonds. This new tool offers a renewed story of the financial shocks. A better match between crashes and historical events is achieved than with pure price variations. Events that were financially insignificant when measured in percent become important crashes after adjustment for volatility. This improved matching brings new insights to several historical debates.

Consistent with other historical sources pointing out the severity of the 1847 crisis, this episode appears to be in the top ten crashes of the UK bond market whereas it ranks 102th in pure price variations. The start of the American Civil War caused a significant crash, supporting the cost side in the cost/advantage debate about this conflict. The Berlin conference dividing up Africa caused a considerable fall in UK bonds, as if the market took account of the future cost of African colonization for UK public finances. Pre-1914 wars (Franco-Prussian, Russo-Ottoman, Boxer Rebellion in China, Boer War, etc.) led to many crashes on both the French stock and UK bond markets, supporting the traditional narrative of the importance of these confrontations despite the weak price changes they caused in this era of low volatility.

Turning to the 20th century, the outbreak of WWI caused major crashes in both French stock and UK bond markets, mitigating the view of sleepwalking to disaster. It is not possible to distinguish more crashes before than after the creation of the Fed in 1913, whose role in stabilizing financial markets is still being questioned. Two crashes in France during the 1920s caused by monetary issues support analysis of French monetary policy as an important factor in the interwar troubles. Hot episodes of the cold war caused crashes on the US and French stock markets, which is consistent with narratives of the risk of disasters incurred at this time. There was no crash on the French stock and UK bond markets in 1929, supporting the views of a transmission of the Great Depression to Europe through other channels than financial markets. The 2008 crisis differs on this point because both French and US stock markets fell strongly.

Maybe, our understanding of financial mechanisms could be enriched thanks to this new tool.

Le Bris, David, What Is a Market Crash? (March 1, 2016). Economic History Review, Forthcoming. Available at SSRN: or