I

Malmö diskont was founded in 1803. King Gustav IV Adolf introduced a monetary reform that year, linking the official currency (the riksdaler) to specie. Simultaneously with this reform, it was considered desirable to encourage the establishment of regional banks so as to fulfil two purposes. The primary purpose was to stimulate regional economic development, and the secondary purpose was to help promote the use of the ‘new’ currency.Footnote ¹ Three banks were chartered: one in Gothenburg on the west coast of Sweden, one in Malmö in the southern Swedish province of Scania and one in Åbo in Finland, which was then part of Sweden. Åbo diskont was closed as a consequence of the Swedish–Russian War of 1808–9. The share capital in the diskont banks was subscribed by private individuals. In exchange for providing a committed line of credit at a favourable rate (3 per cent per annum), Riksens Ständers Bank (a parliament-controlled bank located in Stockholm and essentially the central bank of Sweden) would share in the profits of the diskont banks. In 1810, Göta Kanaldiskont, a new diskont bank, was established to assist in financing the construction of Göta canal, a canal connecting the Baltic Sea with the North Sea. By 1810, the banks in Gothenburg and Malmö had become well established and very profitable and had distributed generous dividends to their shareholders (Andersson Reference Andersson1985, p. 20, table 4).

These banks were called ‘diskont banks’ (discount banks) because they engaged in the lending practice of discounting and deducting the interest on loans up front. At diskont banks, the maximum loan period was six months and surety would normally be given by two reputable citizens or by other security in shares, bonds or merchandise of the highest quality.Footnote ² Whereas Riksens Ständers Bank had a monopoly on lending in Stockholm, each of the three diskont banks almost exclusively conducted their lending business in their respective regions. Because Riksens Ständers Bank did not accept deposits, the regional diskont banks were able to fund themselves in Stockholm. Bonds with a 3 per cent interest rate, known as ‘3 per cent bonds’ or ‘three percenters’, were issued against deposits at Malmö diskont.

When Malmö diskont was founded, it had three directors. Two of them, Frans Suell (a merchant and industrialist) and Cornelius Faxe (a merchant) hailed from well-respected families in Malmö. The third director, Carl Magnus Nordlindh, who had been born out of wedlock in a city near Stockholm, began his meteoric career in Malmö by forging connections with influential people in the city. He managed to become an established merchant, member of the town eldest (the governing body of the burghers) and, in 1805, mayor of Malmö. Nordlindh also achieved high-ranking positions in guilds, merchant societies and order societies. Following Suell's and Faxe's resignations from Malmö diskont in 1812 and 1813, respectively, their positions were filled by Suell's half-brother Henrik Falkman and Samuel Christian Björkman, both of whom were merchants.

Until they folded in the banking crisis of 1817, all three diskont banks helped develop the financial sector in their own regions in the early nineteenth century. They played a role in financing infrastructural changes (e.g. Malmö diskont, among other things, supported the enclosure movement in Scania) and in increasing their loan volume and means of payment in circulation. Loans from the three banks were generally not issued in specie or government money but rather in certificates of deposits (assignations) with Riksens Ständers Bank. In everyday trade, the public accepted assignations, as well as 3 per cent bonds, as means of payment. In so doing, the diskont banks increased not only their loan volume but also the means of payment in circulation. In the decade leading up to the crash of 1817, the paper money issued by the diskont banks accounted for about one-fifth of the money supply (Brisman Reference Brisman1908, p. 187; Sveriges Riksbank 1668–1924, part V, 1931; Andreen Reference Andreen1958, p. 614). The remainder was issued by Riksens Ständers Bank and the National Debt Office.

II

This article attempts to answer the following research question: why did Malmö diskont go under during the bank crisis of 1817? Institutional theory is used here as a tool to describe the interaction between institutions and actors. According to conventional wisdom, the banking crisis of 1817 resulted from the combined effects of the ending of the Napoleonic Wars and criminal behaviour among the directors of Malmö diskont. I argue that these are valid reasons for Malmö diskont's problems but do not adequately explain the directors' actions. When the bank was liquidated, the bank operation was sound, apart from the three major borrowers (Brisman Reference Brisman1924, p. 56).

An institutional analysis is conducted in this article. Institutions influence the way in which actors behave and interact. North (Reference North1990) states that ‘institutions are the rules of the game in society or, more formally, are the humanly devised constraints that shape human interaction’ (p. 3). For this process to be understood, the character of the institutional framework must also be analysed. Interaction between institutions and actors in the financial market consists of a chain of events, which influences the effects of rules and regulations. This chain of events begins with the need for new or modified rules (institutions) and continues with their design and wording, their communication to those that are affected, actors' perception of the rules and underlying intentions and the follow-up and enforcement of the rules that are established. This chain of events determines how institutions work. The way in which actors behave can, in an interactive way, give rise to the need for further institutional change. Although this chain of events primarily refers to formal institutions, it can apply to informal institutions as well. Commons (Reference Commons1957), who also encourages the use of an institutional approach when studying informal institutions, describes the role of unwritten rules in society as follows:

In contrast to today's elaborate formal banking laws and rules on audit and bank inspection, a majority of business rules were informal during Malmö diskont's existence.

There is a lack of universal consent on the influence of institutions. According to North (Reference North1990), ‘a mixture of informal norms, rules, and enforcement characteristics together defines the choice set and results in outcomes’ (p. 53). In a different context, North (Reference North1993) claims that ‘Institutions form the incentive structure of a society and the political and economic institutions, in consequence, are the underlying determinant of economic performance’ (p. 1). Similarly, Hodgson (Reference Hodgson1988) states, albeit more explicitly, the following: ‘Thus in a subtle way, through the operation of market conventions, routines and rules, the individual in the market is to some extent “coerced” into specific types of behavior’ (p. 179). In contrast to the deterministic view expressed by North and Hodgson, Sjöstrand (Reference Sjöstrand, Groenewegen, Pitelis and Sjöstrand1995) writes: ‘It has been suggested above that institutions influence human (inter)-actions, but this is not synonymous with a declaration of environmental determinism’ (p. 28). In his book, Kindleberger mentions Coase's more extreme view of the situation, who contends that ‘demand and supply call the tune and that institutions dance to it’ (Reference Kindleberger2006, p. 4). As I have already stated above, both viewpoints are too simplistic, and the connection between institutions and actors is complex and, thus, cannot be reduced to an either-or situation.

Information about existing rules is needed as a basis for decision-making. Ambiguous, missing or faulty information can lead to incorrect decisions being made. However, in the case of Malmö diskont, inaction was not an option. Those in charge had to make decisions even if information about the rules was ambiguous, missing or faulty. As a consequence, some of the decisions that were made were indeed incorrect. Institutional theory can also be used to discuss the personal conduct of the directors and their reaction to institutional change. Disturbances in the regulatory system may have occurred as a consequence of new institutions failing to conform to the norms of society, which Sjögren and Knutsen (Reference Sjögren, Knutsen and Ögren2010) call an ‘institutional clash’ (p. 188). The roles that the actors have assumed for themselves are a part of this. Scott (Reference Scott1995) notes: ‘All institutional conduct involves roles. Thus roles share in the controlling character of institutionalization’ (p. 38). If a conflict arises between formal institutions and norms in society, or between formal institutions and roles assumed by actors, it alters the effect of institutional change.

III

The diskont banks thrived during the period from 1803 to 1812. Malmö diskont opened its doors for business at a very auspicious time. The enclosure movement had made progress, and credits to agricultural investments in connection with the transformation of farming in Scania were in high demand. As a consequence of achieving success during this period, the bank provided shareholders with generous dividends (Andersson Reference Andersson1985, p. 20, table 4). Crises occurred, however. For instance, when Åbo diskont was closed, runs struck the other two diskont banks. They were stemmed with liquidity support from Riksens Ständers Bank.

Sweden had spent a large amount of money fighting the war with Russia. Consequently, the monetary base expanded immensely during the period up to 1810. The official money supply (money issued by Riksens Ständers Bank and the National Debt Office) increased by 50 per cent (Andreen Reference Andreen1958, p. 65). Attempts to maintain the convertibility of the riksdaler into specie became futile, resulting in its abandonment in March 1810 without any formal decision being made (Skogman Reference Skogman1846, p. 27). The monetary expansion coincided with a dramatic rise in inflation, and speculation became rampant in real estate (Brisman Reference Brisman1924, p. 35).

Turbulence marked the first two decades of the nineteenth century in Sweden. For instance, Sweden had to cede one-third of its area (Finland) to Russia following the Swedish–Russian War of 1808–9. A coup d'état occurred in 1809, resulting in the dethronement of King Gustav IV Adolf and the institution of a new royal dynasty in 1810 and a new constitution. The head of the new dynasty was Jean Bernadotte, an experienced military officer and administrator who had been one of Napoleon's marshals. Bernadotte was officially adopted by the newly crowned King Karl XIII and was bestowed with the name of Karl Johan and the title of crown prince.

During the period from 1812 to 1814, interest rates in the unofficial credit market were much higher than in the diskont banks; hence, it became difficult for these banks to attract deposits and to lend at a time characterised by a scarcity of credit (Brisman Reference Brisman1924, p. 41). In an attempt to continue to generate profits – and, in so doing, allow for dividends to be paid to its shareholders – Malmö diskont devised new and innovative lending methods. However, these methods were criticised by the auditors of the bank, and at the annual general meeting of shareholders in 1814, the continuation of these lending methods was forbidden (Report to the extraordinary meeting of shareholders on 8 January 1818). Although the directors committed themselves to ending this behaviour, it did not happen. In 1815, Parliament reiterated that the lending methods were not permitted; the directors continued to use these methods anyway, without receiving sanctions from the authorities (Halling Reference Halling1878–80, pp. 143–5; Brisman Reference Brisman1924, p. 44; Sjövall Reference Sjövall1940, p. 15). During the final years of the Napoleonic Wars, illicit trade occurred as a means by which to bypass the Continental Blockade. Although lending became increasingly risky, it did not stop. In the case of Malmö diskont, three merchants in particular were engaged in risky lending practices: Jonas Ståhle, Jacob Kock and Haqvin Malmros. As a consequence of becoming involved in several failed transactions, Ståhle experienced a liquidity squeeze (Sjövall Reference Sjövall1940, pp. 10-11).

The end of the continental wars in 1815 represented a turning point for Malmö diskont. An extraordinary parliamentary session was held that year. During the session, the possibility of extending the charters for two of the three diskont banks was debated. The 15-year charters for the diskont banks in Gothenburg and Malmö were to expire in 1818. The decision to extend the charter for Malmö diskont by two years to 1820 was made under the condition that it would adhere to strict new rules from BancoUtskottet (these rules are hereafter referred to as the ‘1815 BU rules’), a permanent subcommittee of Parliament responsible for handling financial system issues.Footnote ³

Sweden actually benefited economically from the Napoleonic Wars in that transit trade through several ports in the west and south of Sweden (among them Malmö) flourished as a means by which to bypass the Continental Blockade. Following the wars, Sweden's economy languished. Poor crop yields from 1816 to 1818 also contributed to the downturn (Historisk statistik för Sverige 1959, p. 44). In the aftermath of the continental wars, many merchant bankruptcies occurred in Stockholm, Gothenburg and Malmö (Brisman Reference Brisman1924, p. 51). These bankruptcies had a drastic effect on Malmö diskont's depositors in Stockholm. In November of 1815, after the parliamentary session had ended, bank runs on Malmö diskont and Gothenburg diskont threatened to bring both banks down. Acting Finance Minister Hans Järta, working with Crown Prince Karl Johan, managed to thwart the threat by using state funds to provide liquidity to Malmö diskont. About 700,000 riksdaler was extended to the bank to meet redemption of assignations and 3 per cent bonds. Thus, the bank run subsided in a matter of a few weeks. After the crisis, the government formulated draconian rules (I will refer to them as the ‘1816 government rules’) for the diskont bank to follow in its future activities.Footnote ⁴

Meanwhile, without the government's knowledge, Ståhle, Kock and Malmros had accumulated enormous loans (Halling Reference Halling1878–80, p. 146). In terms of the size of the loans provided to these three merchants, they far exceeded the authority given in the Articles of Association. These credits were rolled over, which was also against the rules of maximum maturity.

In 1816, Malmö diskont edged closer to a financial catastrophe. Although the government loan was amortised in good order, several unscrupulous activities were being carried out. Early in 1816, Ståhle confided to Nordlindh about his intention to admit his perilous financial position and go into bankruptcy; however, Nordlindh believed that Ståhle's financial issues would eventually dissipate and, therefore, persuaded him to refrain from divulging this information (Falkman Reference Falkman1986 [1891], p. 461; Sjövall Reference Sjövall1940, p. 21). Because Ståhle was one of Malmö's most prominent merchants, Nordlindh feared that the effects on Malmö diskont would be cataclysmic if Ståhle's financial woes were revealed. Upon discovery of Malmö diskont's failure to adhere to the Articles of Association, an auditor, who had been appointed by Riksens Ständers Bank, requested at the annual general meeting that the Chancellor of Justice investigate and possibly initiate a legal course of action against the directors of the diskont. The government turned down this request.Footnote ⁵ Guarantors learned of the unstable financial positions of the three major borrowers (Ståhle, Kock and Malmros) and began withdrawing their sureties.Footnote ⁶ Ståhle, Kock and Malmros reacted by issuing guarantees among themselves and even falsifying signatures on guarantees.

By 1817, the government became wary of the financial condition of Malmö diskont. On 21 May, it called upon the county governor in Malmö, Wilhelm af Klinteberg, to conduct a thorough investigation of the financial condition of the bank. In an attempt not to rock the financial system, the government ordered the bank's auditors to issue a ‘clean’ audit report for the activities of 1816.Footnote ⁷ In June, the annual general meeting of Malmö diskont was held. In September, the county governor finalised his report on the financial condition of Malmö diskont, in which he stated that the portion of the three main borrowers' loans that was unsecured amounted to around 640,000 riksdaler. This amount represented about one-third of the diskont's total outstanding loans and exceeded the share capital by more than six times.Footnote ⁸ In late September, the government decided that the next course of action would be to demand that the directors of the diskont banks resign. Anticipating another bank run, Parliament made available funds amounting to 500,000 riksdaler to protect the bank. However, the depositors at Malmö diskont reacted so quickly and strongly that the bank was subjected to a massive run. The government never used the full available amount to protect Malmö diskont but only a fraction thereof.Footnote ⁹

Instead, in early October, the government decided that Malmö diskont would be taken over by the state. This decision coincided with runs on the other two diskont banks, which resulted in their closing.

An extraordinary session of Parliament was convened in November, the sole purpose of which was to determine the fate of the diskont banks. Parliamentary deliberations continued into 1818 until it was decided that the National Debt Office would assume the assets and liabilities of Malmö diskont and shut down the bank. As a result of closing Malmö diskont, the government lost about 430,000 riksdaler; however, the loss associated with the closure of Göta Kanaldiskont (914,000 riksdaler) was more than twice as much (Brisman Reference Brisman1924, p. 57; Andersson Reference Andersson1985, p. 44). Unlike the other two diskont banks, the diskont in Gothenburg distributed a surplus to its shareholders. Holders of assignations issued by Malmö diskont were paid in full, and holders of three percenters were also paid in full but only after a lapse of five years (with interest). In the aftermath of the crash, Nordlindh, Björkman and Falkman, as well as two of the three major borrowers, stood trial and received prison sentences, which they served. None of them returned to business. The province of Scania did not recover from its economic downturn until about 1830. In the intervening years, several new banks – a branch of the Riksens Ständers Bank and a couple of savings banks – were set up in Scania. In 1831, a privately owned commercial bank was established in the region. It was located in Ystad, about 100 km from Malmö. Establishing a bank in Malmö was considered out of the question because the unpleasant memories of the failure of Malmö diskont had not yet faded in the public mind. The dismantling of Malmö diskont was a lengthy process, requiring about 30 years to be completed (Andersson Reference Andersson1985, pp. 43–6).

IV

Sections V and VI describe how the institutional framework was constituted during Malmö diskont's existence. The institutions are divided into formal (Section V) and informal (Section VI) arrangements. Formal institutions originated from directives from authorities such as laws and royal ordinances (including the charter for Malmö diskont, which also comprised the Articles of Association for the bank) as well as the 1816 government rules.Footnote ¹⁰ Among formal institutions, audits and bank inspections are discussed as well as the question of which government or parliamentary entity was responsible for the structure of the financial system in Sweden.

From a methodological perspective, categorising institutions in the early nineteenth century poses some challenges – not so much with royal proclamations or laws but rather with ambiguously or narrowly defined formal institutions. These comprise, for example, banking inspections and audits as well as responsibility for the banking system. This situation raises the question as to whether it is appropriate to apply concepts of institutions of the present day to the past. This article demonstrates that these concepts did actually exist in the past but not in a shape as formalised as today.

The institution of lender of last resort is included among the formal institutions. Because the concept of lender of last resort is usually considered to be a phenomenon of a later date, it merits a special discussion. Kindleberger's modern definition of lender of last resort states that ‘the lender of last resort stands ready to halt a run out of real and illiquid financial assets into money by making more money available’ (Reference Kindleberger2000, p. 161). Although the concept as such was formulated much later than the events surrounding the rise and fall of Malmö diskont, the principles were first set out by Henry Thornton in 1802, just before the diskont banks in Sweden were founded. In chapter 7, entitled ‘Of country banks – their advantages and disadvantages’, in An Enquiry into the Nature and Effects of the Paper Credit of Great Britain, Thornton (Reference Thornton2010 [1802]) states: ‘If any one bank fails, a general run upon the neighboring ones is apt to take place, which, if not checked in the beginning by pouring into the circulation a large quantity of gold, leads to very extensive mischief’ (p. 173). Thornton emphasises that immediate action must be taken in certain situations and that the lender of last resort has to be prepared to infuse large quantities of liquidity. He then provides a real-life example of when actions had been too weak and ineffective, causing the crisis to grow to such proportions that the consequences became impossible to manage (Thornton Reference Thornton2010 [1802], p. 187). The crisis to which Thornton refers occurred in 1793 and involved canal construction (Kindleberger Reference Kindleberger2000, p. 225). Seventy years later, Walter Bagehot stated in his book, Lombard Street: A Description of the Money Market (Reference Bagehot1999 [1873]), the following rules to which a lender of last resort should adhere:

Thornton and Bagehot highlight the need for the action to be rapid and decisive and of sufficient size. Both authors stress that the purpose is to infuse capital into a bank that is illiquid, not one that is insolvent. Illiquidity is a natural occurrence for banks because they are normally mismatched – that is, deposits are more quickly withdrawn than loans recalled. Thornton and Bagehot also prescribed that assistance should be provided at penalty rates for the assisted bank so that it does not create moral hazard. This type of situation can arise when banks, emboldened with the knowledge that the authorities would intervene on their behalf if they become illiquid, take excessive risks. Neither Thornton nor Bagehot suggests that insolvent banks should be saved by a ‘bailout’ (to save banks with a negative equity); they argue that assistance should be given only to banks when short-term deposits are withdrawn rapidly (e.g. caused by a bank run), thus making it impossible for them to meet these withdrawals by liquidating short-term assets. As Ögren (Reference ögren2003) points out, lenders of last resort face the dilemma of distinguishing between illiquidity and insolvency (p. 242).

In this article, I argue that Riksens Ständers Bank acted as a lender of last resort on numerous occasions. Nor were they the only lenders of last resort around this time.

In France, the newly established Banque de France intervened in financial crises as early as 1810 (Wood Reference Wood2000, p. 208). In England, the Bank of England was privately owned and the largest bank in the country, but had no explicit responsibility for the banking system. However, during the crisis of 1825 – brought about by easy monetary conditions and a wave of purchases of Latin American bonds, which then defaulted – the Bank of England had to intervene as a lender of last resort and protect the country banks in particular (Bordo Reference Bordo1998; Neal Reference Neal1998). In several other crises and bank panics during the last part of the eighteenth century and early nineteenth century, the Bank had acted as a quasi-central bank, providing liquidity to troubled banks (James Reference James2012).

In the United States, Alexander Hamilton, the secretary of the treasury from 1789 to 1795, had reorganised public debt and founded a bank (Bank of the United States), which was intended to act as a central bank. Together with a growing network of banks, a financial system was in the making. However, the bursting of a speculative bubble in 1792 threatened to ruin Hamilton's ambitions. By injecting substantial amounts of money into the bond market, increasing credit facilities and acting decisively, he managed to handle the crisis successfully, thus allowing confidence to be restored in the financial system (Sylla et al. Reference Sylla, Wright and Cowen2009).

The ultimate motive behind the actions of a lender of last resort is to protect the financial system in a crisis, which, if not contained, may disrupt the entire economy. When private banks issued notes that were used as means of payment, the risk that banks would default threatened the availability of money. If a bank defaulted, its notes became worthless, the money supply decreased and the negative effects on economic growth of scarce money (and vice versa) were known very early (Hume Reference Hume1953). In the aforementioned examples, it was the state of the financial system that was in peril, and the rescue efforts were primarily designed to protect the system, not individual banks.

V

The formal institutional framework of the period of time under study is discussed in this section. The main legal document was the comprehensive Law of Sweden of 1734. It comprised, among other things, Acts dealing with marriage, inheritance, trade, real estate, ill deeds and legal procedures. Following the collapse of Malmö diskont in 1817, its directors were tried under this law. The rules for conducting business were specified in the Trade Act in the law.

The Articles of Association for Malmö diskont were included in its royal charter. Operationally, these articles were similar to the Articles of Association for diskont banks that had been established during the late eighteenth century. Concerning creditworthiness, the Articles of Association stipulated that credit could be extended (or surety accepted) only by those having a recognised fortune or generally well-accepted credit (at that time, credit meant ‘trust’ or ‘good name’). Essentially, banks adopted this merchant credit custom as banking practice.

Malmö diskont was greatly influenced by the 1815 BU rules and the 1816 government rules, two important sets of institutions. The former, which had been issued in response to one of the reasons for the parliamentary session of 1815, concerned the extension of Malmö diskont's charter and outlined several special operating rules for Malmö diskont to follow, thus limiting its freedom of action. The latter, which were introduced in response to the November crisis of 1815, were meant to have a ‘straitjacketing’ effect by setting forth several operational rules: only one-third of loans repaid was allowed to be re-lent; they also set a borrowing limit of 1,000 riksdaler (compared with 5,000 riksdaler in the Articles of Association) for each borrower unless collateralised. All loans were to be unconditionally repaid at maturity, with a possible extension of six months; if the loan was not repaid, the collateral would be collected by the bank and sold. Those loans exceeding the prescribed maximum were to be decreased to the maximum amount and not renewed. In addition, Malmö diskont was to repay as soon as possible the cash support it had received from the state, amounting to 700,000 riksdaler. Before the crash of 1817, Malmö diskont did indeed repay the cash support in full.Footnote ¹¹

When diskont banks existed in the early nineteenth century, responsibility for the banking system was unclear. On the one hand, Parliament was responsible for Riksens Ständers Bank and the National Debt Office. On the other hand, the government was responsible for economic affairs. A conflict between Parliament and the government with its roots in the eighteenth century was carried over into the nineteenth century when both official institutions endeavoured to gain greater control of the policy regarding financial system management (Rexius Reference Rexius1917, p. 23; Fritz Reference Fritz1967, p. 143; Nilsson Reference Nilsson1988, p. 5; Larsson and Lindgren Reference Larsson, Lindgren and Cassis1992, p. 340). For instance, the government issued the Articles of Association for the diskont banks, and Parliament issued the 1815 rules for the diskont banks. Following the November crisis of 1815, the government seemed to possess greater control of the situation than did Parliament. When the diskont collapsed, the government assumed responsibility for directing the process. However, in the crisis that ensued after the fall of Malmö diskont, the government immediately called another extraordinary parliamentary session so, thus shifting responsibility back to Parliament.

In Sweden, the Royal Bank Inspection was formally established in 1907. However, from 1868 onwards, there existed a bureau in the Finance Ministry that was assigned to monitor developments in the financial system (Sommarin Reference Sommarin1927, p. 6). Although conventional wisdom is that bank inspection did not exist before its establishment in 1907, it can be argued that bank inspection functionally (albeit not formally) existed in the 1810s. For instance, special investigators appointed by the government and Riksens Ständers Bank carried out three ad hoc inspections of Malmö diskont between 1815 and its collapse.

Malmö diskont's operations were monitored by four auditors. Of these four auditors, three were elected among its shareholders, and one was appointed by Riksens Ständers Bank. Typically, the auditors elected among the diskont bank's shareholders were both shareholders and borrowers in the diskont bank. From 1812 onwards, the auditors were responsible for disclosing the bank's practice of issuing non-covered (without a corresponding deposit) 3 per cent bonds. The authorities responded by stating that management should refrain from performing this bank practice. Although management failed to comply, the authorities did not impose any sanctions.

Despite the fact that a lender of last resort did not formally exist in Sweden during the nineteenth century, the state carried out actions on a couple of occasions, notably in 1857 and 1879, which have been typified as those of a lender of last resort. I argue in this article that as early as 1815, there is evidence of last-resort lending. The government, exhibiting strength and possessing access to seemingly vast sources of money, loaned money to two banks besieged by depositors and holders of assignations. The government demonstrated remarkable strength and seemingly bottomless sources of money. The state had also aided the diskont banks several times before 1815 by providing cash and support in liquidity crises and for the Göta kanal project (Brisman Reference Brisman1924; Andreen Reference Andreen1958). Therefore, even without a formal structure, the diskont banks were accustomed to receiving support from the state (a lender of last resort).

VI

The informal institutions are discussed in this section. There are two main groups: best practice and trust/reputation. During the early nineteenth century, best practice in banking was not formalised in Sweden. Most of the lending occurred outside the banking system (Lindgren Reference Lindgren2002). Lending was predominantly conducted by and between merchants, mostly in connection with trade. These merchants also lent to the general public – mainly to other business partners and to relatives. Credit was provided under what can be described as best practice among merchants. Although loans were often collateralised by merchandise, the decision to lend was also commonly based on a borrower's reputation – that is, borrowers who had exhibited good business behaviour and established a good reputation for themselves were more likely to be granted loans. At Malmö diskont, the Articles of Association stipulated that loans could be provided to individuals who were ‘considered’ to be wealthy or of good standing. Financial analysis as a basis for credit granting was not introduced in Sweden until the beginning of the twentieth century. Marcus Wallenberg of Stockholms Enskilda Bank is said to have been the first banker to require financial statements from companies that wanted credit from the bank (Gårdlund Reference Gårdlund1947, p. 138). The merchant credit system was also founded on social relations. Before engaging in risky lending practices at Malmö diskont, each of the three merchants – Ståhle, Kock and Malmros – were well connected and held in good repute. Ståhle was considered one of the leading merchants in Malmö, Kock was married to Nordlindh's sister and Malmros hailed from a prominent family.

Social relations and reputation were important aspects of conducting business at Malmö diskont. All three directors of Malmö diskont belonged to or held high positions in Malmö's professional groups such as the Merchant Society (Handelssocieteten) and the Agricultural Society (Hushållningssällskapet). Society life revolved around, in particular, the Order of the Amaranth and the Guild of St Knut. The shareholders of Malmö diskont even formed their own social group. Because most of these shareholders were well-connected, prominent citizens of Scania, their membership in the same group helped promote social relations. Membership in the groups comprised merchants, estate owners (mostly nobility) and craftsmen who very often belonged to more than one group. This sense of belonging to a group instilled trust among members, which in turn enhanced their respective reputations.

VII

The next step in the analysis is to assess qualitatively the interrelations between the institutional framework and the actors concerned. According to the legal requirements clearly stated in the law of 1734, the directors of Malmö diskont committed many crimes, including falsifying documents and withholding information. Although the law also explicitly stated that it was the responsibility of shareholders to fulfil all obligations of the bank, there was no evidence of any attempts to do so after the crash. A special committee, appointed by Parliament, instead decided that the state should handle the dismantling of Malmö diskont and that bankruptcy proceedings would not be carried out. The motives for this procedure are a matter of speculation. The most likely explanation is that if a bankruptcy proceeding was initiated, the ensuing events might be beyond the control of the government, and severe disturbances might occur in the financial system. Lundh (Reference Lundh and Bring1930, p. 300) writes about the danger posed by the possible disruption of the orderly proceedings by individual creditors. Another explanation is that the crown prince was worried about placing too much pressure on the elite of Scania, a politically, strategically and economically important region of Sweden. A large portion of the elite of Scania were also shareholders in Malmö diskontFootnote ¹² (Shareholders' register). At that time, there existed a recent precedent on the effects of unlimited liability by the owners of a bank in a bank crash. In 1772, Ayr Bank, under the designation of Douglas Heron & Co., crashed (Rait Reference Rait1930; Hamilton Reference Hamilton1956; Checkland Reference Checkland1975). The other banks in Scotland, under the leadership of the Bank of Scotland and the Royal Bank of Scotland, acted to avert a systemic crisis for the banking system. Following the bankruptcy, the losses were reclaimed from the shareholders. The dismantling of Ayr Bank took several decades to complete. The effects on Scotland were disastrous. The owner of a share in Ayr Bank, bought for £500, had an average loss of not only the £500 but also £2,200 plus interest up to 1788 (Checkland Reference Checkland1975, p. 133). By 1775, 114 out of a total of 226 proprietors at the time of the crash had suffered bankruptcy (Rait Reference Rait1930, p. 167). The crash also meant ruin for many bankers in Scotland as well as in London. However, depositors and lenders to the bank were repaid in full – as would formally be the case with Malmö diskont. If this outcome of the 1772 crisis were known to the crown prince, and if it were true that he did not want to upset the people of Scania, the example of Ayr Bank would not promote the initiation of a bankruptcy proceeding.

The Articles of Association not only represented a source of mismanagement but also facilitated it – particularly in terms of credit extension, which should be granted to individuals of ‘good standing’ and ‘assumed fortune’. Admittedly, this practice had its roots in merchant business, and, on the whole, it had worked well. Similar clauses had been present in the Articles of Association of previously established diskont banks during the eighteenth century and had not led to any major calamities. However, the rules did not work in the troubled times during and after the Napoleonic Wars. Another tension-building clause in the Articles of Association refers to the fact that big loans should not be extended if doing so would crowd out smaller and better borrowers.Footnote ¹³ Although prudent, this rule was impossible to follow. For example, when credit was tight in 1816, three big borrowers at Malmö diskont were already crowding out smaller ones.

If rules can be easily circumvented and are not properly enforced, individuals will solve a problem the wrong way. The 1815 BU and 1816 government rules were unambiguous in their design but tough on the diskont banks, particularly Malmö diskont. If these directives had been strictly adhered to at Malmö diskont, with Ståhle as a major borrower, the consequences for the other borrowers and the province of Scania would have been dire. Many borrowers had made long-term investments that would yield enough to cover the debt service of the loans. The loans, however, could not be recalled on short notice. Hence, the rules were not followed. This situation made it possible for the rules to be circumvented. When the directors broke rules and were discovered, they were not sanctioned. They were instead given clean audit reports. No sanctions were imposed on directors who promised to correct their behaviour but did not necessarily do so. Because the enforcement of rules was lax, the rules were perceived as less strict than they actually were.

Another area of ambiguous governance was whether the government or Parliament was responsible for the banks. Before the crisis of 1815, the diskont bank directors had been given the impression that they were subordinate to Parliament and Riksens Ständers Bank. When the government assumed firm control of affairs for the diskont banks in early 1816, the identity of the bank's principal became unclear, inspections occurred only ad hoc and audits were directed by the government.

Institutions had provided support to diskont banks several times during their existence in much the same way that lenders of last resort would act. Whenever the diskont banks had liquidity problems, repayments were quickly made (Brisman Reference Brisman1924, pp. 30–4). Because they had always received help in the past, the perception was that they would always get help when conditions were dire. This can be described as a moral hazard, which contributed to the directors' actions.

Before the collapse, the loans to the three major borrowers at Malmö diskont were estimated to be about six times the share capital and were generally considered unsecured. In retrospect, the rest of the bank was, however, sound (Brisman Reference Brisman1924, p. 56). Thus, if the government had wanted, it could have continued the business. The directors may have come to the conclusion that they would be punished for their personal behaviour but that the government would maintain Malmö diskont because of its importance as a financial institution. This was indeed the opinion of the county governor, Wilhelm af Klinteberg, who assumed that Malmö diskont would continue its business.Footnote ¹⁴

Banking practice at Malmö diskont was, in many respects, based on merchant practice, which is not surprising considering that all three directors were merchants. Lending policies among merchants, such as those involving longer than permitted maturities, and the generous dividend policy of Malmö diskont were consistent with merchant practice at the time but unfitting for a diskont bank (Samuelsson Reference Samuelsson1951, p. 221). The former behaviour increased the credit risk, and the latter led to a deterioration of the capital base.

Social relations have an impact on the behaviour of bank directors. Without the trust of depositors, banks cannot fund their assets. Building public trust attracts depositors, creditors and shareholders. Trust is also an important aspect of building a good reputation. The public's trust and perception of Malmö diskont was derived from the directors' extensive social relations and positions. However, the public's trust would have disappeared in the event of disclosure of the bank's severe financial problems. The public had complete trust in Malmö diskont up to the day that it was announced in Malmö that the government would assume control of the bank and that the directors would be prosecuted (Halling Reference Halling1878–80, pp. 165 ff.). The directors' failure to maintain the public's trust resulted in their loss of social position. Without their social position, they could not conduct business and participate in the social life of Malmö. The imperativeness of maintaining trust and conserving reputation in their business prompted members of management at Malmö to cover up their misdeeds.

VIII

The question of why Malmö diskont collapsed in the autumn of 1817 can be divided into two parts: why did Malmö diskont not collapse earlier? Why did Malmö diskont collapse in October 1817? The period before 1817 was rather bleak for merchants and banks because it coincided with the end of the Napoleonic Wars and a weakened economy. However, there are several reasons why Malmö diskont did not collapse before 1817. First, several groups of people (including diskont managers, shareholders in Malmö diskont, borrowers and depositors) benefited if the diskont carried on its business. Second, the crown prince did not want to risk upsetting the citizens of Scania, who were from one of the most important provinces (from both a political and an economic standpoint) in Sweden, by allowing Malmö diskont to fall. Third, the state might have lost some of the 700,000 riksdaler it had lent to Malmö diskont if it collapsed. Fourth, Riksens Ständers Bank would have lost its share of the profits of Malmö diskont if it closed. Last, the risk of domestic contagion was apparent to the government. If the Göta Kanaldiskont bank, which was responsible for financing the Göta kanal project, also collapsed, this responsibility would then be transferred to the government. Together, these factors contributed to the postponement of Malmö diskont's fall.

Although the interaction between institutions and actors set up the collapse of Malmö diskont in October of 1817, the death blow was dealt by an actor. The institutional framework was overridden by the government, led by the crown prince. It would have been possible to save Malmö diskont for the costs were not overwhelming. As previously mentioned in this article, the government had 500,000 riksdaler at its disposal from Parliament. The latest estimate of unsecured assets before the crash was 640,000 riksdaler, which turned out to have been 430,000 riksdaler in actuality. The difference was not large. However, continuing business would have entailed reforming the institutional structure. The most immediate reforms should have involved changing the Articles of Association to become more oriented towards consolidation of the bank. The enforcement of rules in the Articles of Association also needed to be more effective. A revision of the 1815 BU and 1816 government rules would have been necessary to prevent a reduction of operations from occurring too rapidly. However, Malmö diskont was not saved, and the reasons for letting the bank fall in 1817 can be summarised as follows: first, the crown prince suspected the diskont banks of financing currency speculators, who worked against the crown prince's ambitions to strengthen the Swedish currency. The crown prince thought Denmark was acting against his currency policy of strengthening the value of the riksdaler and viewed the neighbouring province of Scania as an accomplice, which partly involved increasing smuggling into Sweden, thereby weakening the currency.Footnote ¹⁵ Second, the crown prince was of the opinion that the root of the currency problem was the excessively large money supply, of which the diskont banks' assignations and 3 per cent bonds were a substantial part (Brisman Reference Brisman1908, p. 187; Sveriges Riksbank 1668–1924, part V, 1931; Andreen Reference Andreen1958, p. 614). Third, the crown prince carried with him from France his experiences with the bursting of the Mississippi bubble in 1720 and the erosion of the value of the assignats during the 1790s as well as the advantages of a centralised banking system practised by Napoleon. In addition, the audit (presented in late September) showed without doubt that Malmö diskont was insolvent. The crown prince was thereby in bad faith. Any indulgence of misbehaviour placed him in a very sensitive position because the crown prince was not yet king of Sweden (King Karl XIII died on 5 February 1818, and Karl Johan became king on the following day). Last, the initial liquidity injection produced the exact opposite of the desired effect. There was a run on not only Malmö diskont but also the other two diskont banks. The crown prince and the government then decided to refrain from ‘throwing good money after bad’ and to stop the rescue mission.

IX

In conclusion, the interaction between institutions and the actors concerned in connection with the fall of Malmö diskont functioned badly. This is the ‘chain of events’ described in Section II. The joint responsibility of shareholders for the obligations of the bank was not exercised. The Articles of Association of the diskont were ambiguous – although in many respects conforming to the then prevailing best practice in the unofficial merchant credit market. There were no limits on total lending or borrowing from the public. Breaches against the Articles of Association were permitted after the fact, as long as the diskont management promised to correct its behaviour. The new rules for the conduct of business issued by Parliament and the government in 1815 and 1816, respectively, were unambiguous but extremely restrictive. Although these rules were so restrictive that they were impossible to follow, they could be easily circumvented. It was unclear who the principal for the banks was; the inspection was ad hoc and, in retrospect, shallow; and auditors of the bank were permissive and reticent to impose sanctions. The enforcement of the rules was also weak. Up until the crash occurred, the lender of last resort function was clear and effective and instilled confidence in the diskont management that help would be forthcoming. The management of Malmö diskont lacked banking experience, which was in contrast to the merchants of Gothenburg, where several banks had existed in the latter part of the eighteenth century. The tight social circle that existed in Malmö that was necessary for creating trust and, thereby, creditworthiness, deterred early disclosure and transparency in the bank's business. The incentive structure was conducive to continuing business expansion instead of maintaining a more prudent banking policy. Taken together, the interaction between institutions and actors led the diskont management in the wrong direction. With the financial condition of Malmö diskont and its institutional structure, a collapse was imminent. However, a crash could have been avoided with a capital injection and restructuring of the regulatory framework. To summarise, the interaction between institutions and actors undermined the banking system, but it took an individual action to topple the bank.

In-depth empirical case studies of individual bank crises in the nineteenth century merit investigation because, as Reinhart and Rogoff (Reference Reinhart and Rogoff2009) point out, ‘We have been here before’ (p. xxv). According to Reinhart and Rogoff, there exist common factors in financial crises that can be traced back for centuries. In today's financial world clamouring for new regulations and rules for the banking game, the case of Malmö diskont is illustrative of what can happen if the rules are ambiguous, contradictory to other rules, badly written, inadequately communicated or poorly enforced.