CHAPTER 8
Money talks
Failed cooperation over the gold problem of the Scandinavian Monetary Union during the First World War
The Scandinavian Monetary Union (SMU) initiated in 1873 constituted one of history’s very best examples of Scandinavian cooperation. The initial agreement regarding common coins was gradually extended to include increasingly new means of payment and settlements, in the hope of establishing a basis for increased trade and integration. The cooperation of the union’s three central banks—Riksbanken in Sweden, Nationalbanken in Denmark, and Norges Bank in Norway—was exposed to a number of setbacks over the ensuing forty years. However, the SMU was by and large a well-functioning institution of cooperation by the outbreak of the First World War; it was as wartime experience uncovered important economic differences between the three countries that the union gradually dissolved.
In 1917 the leading contemporary Danish economist and central bank critic, Axel Nielsen, ended his book on the Scandinavian Monetary Union thus: ‘What is left of it, however, is difficult to see. Apparently all the economic cooperation that was built up in the nineteenth century through more than 30 years’ tireless work has been laid waste in the course of only two years, and this at a time when a new movement for a more intimate economic cooperation is growing.’1 In the midst of an incomplete development imposed by the war, Nielsen obviously was neither able to see the full picture nor know the details of what had been going on in the past couple of years; however, his defeatism was certainly pertinent.
The reasonably smooth operation of the SMU had run into a stonewall, and despite apparent efforts to maintain cooperation, each of the three central banks was basically working for their own financial or national considerations. The ‘new movement’ Nielsen was referring to was most likely the Scandinavian emergency measures of 1917, which saw trade representatives of the three countries set out to help one another out by exchanging commodities. Yet this was at a time when Scandinavian monetary cooperation was approaching its nadir. During the war, cooperation within the union continued successfully only as long as it was to all the partners’ advantage—on matters that were costly and really mattered it was virtually absent. Thus it was wartime experience that dealt the union its death blow, despite the futile attempts to restore the union that were initiated alongside the broader, international moves towards economic cooperation in the 1920s. Officially, however, the SMU existed until its termination in 1973.
This chapter will provide a short introduction to the main features and historical development of the SMU’s heyday prior to the First World War, which earned the union its reputation as ‘the most successful of all European currency unions’.2 The main focus, however, is the union’s ‘gold problem’ after gold convertibility was suspended in August 1914. Despite the fact that the value of the three currencies started to deviate on the world markets, Scandinavian gold coins still had to be bought by the central banks at the old, fixed price; gold was still legal tender at full value—par value—within the union. This implied that someone—for example, individuals, central banks, or governments—could take advantage of the situation and profit by differences between the market and the legal values. This moreover happened at a time when commodities were scarce and the role of gold as the measure of value was under attack. The point of departure in this chapter is the attempt to maintain central bank cooperation on gold policy within the SMU, with chief focus on a number of meetings—‘money talks’—between the three countries’ central banks.
The gold standard, central banks, and the SMU
During the second half of the nineteenth century, the national banks were in the midst of becoming ‘central banks’, a process which we will be returning to below. At this time the most important task of these banks was to uphold the value of money. A common way of achieving this was to fix the national currency to a precious metal such as silver or gold—or both. Coins thus had a specified metal content, and paper money was made legal tender, which meant that the central banks were obliged to sell or buy their notes for metal at a fixed price. From the 1860s gold gradually replaced both silver and bimetallism, and when the first talks about a Scandinavian union were held, there was hardly question of anything but the gold standard. Even though the Norwegian Storting rejected the 1873 convention for the Scandinavian Monetary Union, it eventually joined Sweden and Denmark in 1875.3 From its very start the SMU involved common coins containing gold at a quantity that made it easy to convert from the previous three different silver standards. The new currency and legal tender in all three countries was the Scandinavian gold krone/krona containing 0.44803 grams of gold, and the three central banks were obliged to buy and sell gold at a fixed price: 2480 kroner/kronor per kilo, against a 0.25 per cent fee covering expenses. This price was the same in 1873 as it was in 1914, and by keeping this fixed gold price the SMU was simultaneously part of the international gold standard framework of fixed interchangeable currencies.
Another task of these ‘central banks’ was to facilitate trade. This could be done either by selling or buying bills of exchange from individuals and companies, and thus acting like a normal bank, or by rediscounting bills of exchange or drafts from the expanding field of commercial banking. A bill of exchange was the most common means of payment in business transactions of the nineteenth century, a financial document by which the issuer bound himself to pay the holder an amount of money on a given date—both specified in the document—while a draft was a bill where the issuer ordered a third party to pay the holder. In international trade these bills and drafts to a large extent made the physical payment of gold or silver superfluous, and a single bill could be used in numerous business transactions and could even cross borders.4
In 1885, the SMU expanded its scope considerably when it was decided that all three central banks could sell drafts upon one another and transfer them at par value without paying a fee. This meant that, for example, Nationalbanken thus could ‘draw’ on Riksbanken by issuing a draft to a Danish importer that Riksbanken ultimately would have to pay to the Swedish exporter, which simultaneously increased Nationalbanken’s debt to Riksbanken. The decision was intended to facilitate larger settlements between the countries. In 1894, business and trade benefited further when Riksbanken and Norges Bank agreed to accept each other’s notes at par value. Nationalbanken, however, was reluctant, and did not join this part of the union until 1901, at which time the SMU was at its height. This heyday only lasted a few years, as Riksbanken in 1905 terminated the 1885 agreement. The decision was undoubtedly encouraged by the political tensions of the year, as Norway broke out of the personal union with Sweden. However, considering the development of imbalances in drawing between the banks, Riksbanken probably had good financial arguments. The 1885 agreement had made it illegal to charge a fee on drafts, although a fee might have helped to relieve Riksbanken of the demand and ultimately the need to ship gold to cover its debt. Thus bilateral replacement agreements were concluded in 1905, introducing limits for the normal interest-free debts the central banks could have on their books. Where the debt was not paid by shipping gold, it was left to the banks to decide if a fee was to be introduced for any debts exceeding these limits.
So what can be said of the status of the SMU at the outbreak of the First World War? There obviously had been setbacks, something the 1905 turmoil bears witness to: for shorter periods, fees on drafts had been introduced, although admittedly not until 1910; there is evidence that the SMU did not facilitate trade integration, as inter-Scandinavian trade actually declined; and it is difficult to distinguish between the effects of the monetary union and adherence to the gold standard. Yet all this notwithstanding, the SMU in 1914 is probably correctly described as well-functioning. When it was ‘at its best’ between 1901 and 1905 it has been labelled a ‘complete system’ of payment.5 Until this point, monetary cooperation had strengthened on an almost year-by-year basis, and the reversals of 1905 and 1910 do not alter the fact that coins and notes were still traded at par. The fees charged on drafts were temporary arrangements, and despite a fall in central bank drawing after 1905, the use and circulation of the different Scandinavian banknotes consequently expanded considerably. In addition, central bank cooperation was facilitated by, for example, the sharing of information on changes in discount rates. Recent work has also pointed to the fact that even though the SMU did not facilitate trade, it did have an important function in fostering financial integration through short-term financing.6 Finally, the SMU’s very existence and continued survival was important politically, being a major accomplishment of Scandinavianism.
By the outbreak of war, Riksbanken, Nationalbanken, and Norges Bank had all become true central banks. This first of all meant that the banks had gained a monopoly on issuing notes, something that was not that common during the nineteenth century. In fact it was not until 1897 that Riksbanken was given a note-issuing monopoly.7 Moreover, the banks had mostly withdrawn from competing with—or covering for the lack of—private commercial banks, which thus resulted in a growth in the private banking sectors of the three countries, albeit with somewhat differing outcomes. For instance, where Sweden’s financial sector saw a move towards fewer, yet bigger banks, Norway had an increasing number of small private banks. Furthermore, the three were central banks by dint of their role as lender of last resort; during times of crisis, they would come to the assistance of the financial sector—and thus the state.
This notwithstanding, there were differences in the banks’ relationships with their respective states. Where Riksbanken was state-owned, Nationalbanken and Norges Bank were privately owned, although a substantial part of their shares were held by the state and central bank legislation guided their actions. The Swedish Riksdag’s control of Riksbanken gave the governor Victor Moll (1912–29) a rather weak start. Moll, having worked his way up in private banking, was the Liberal’s chosen candidate. In the Riksdag, Moll and the Conservative’s candidate received the same number of votes, and Moll was thus elected governor by the solid foundation of drawing lots. Even though he was a Liberal MP, Moll’s counterpart at Norges Bank, the entrepreneur Karl Gether Bomhoff, was in many respects a contrast. Bomhoff had been on the Board of Directors since 1885, and in 1893 he became the first permanent, government-appointed governor of Norges Bank. It was under Bomhoff that Norges Bank had stepped in as lender of last resort in 1899 following the burst of Kristiania property bubble that year. At Nationalbanken, meanwhile, Marcus Rubin had scarcely taken the governor’s seat before war broke out, having been appointed by the government, thanks to his connections with the Liberal Party, as recently as 1913. Rubin had extensive management experience, having served as head of both the statistical bureau and the tax authorities, and his background seemed promising—he was both a statistician and an historian, and he had published several works on economic history—and although lacking formal banking experience, he seemed well equipped for the task. However, neither experience nor background could prepare the three men for what was to come in 1914.
Diverging exchange rates and the gold problem
The outbreak of the First World War had a massive impact on international business and trade. Production in the belligerent countries—the large economies of Europe—was directed towards materiel, while transportation by land and sea was made difficult by the endless trenches scarring European soil and the lurking dangers of submarine warfare. Commodities were scarce and insufficient to meet demand in the belligerent and neutral countries alike. The scarcity naturally added to the hardships of the populations of the main belligerent countries. Despite intermittently severe problems in obtaining the necessary imports, the Scandinavian countries overall experienced a favourable trade cycle up until 1917. The export sectors fared well, trading with the belligerents. During the course of the war, however, the hardships were felt increasingly strongly in neutral countries as well. Many governments ultimately found it necessary to expand the once small-scale and passive modus of the liberal, nineteenth-century state into new areas of economic affairs. Thus neutral countries, like the belligerent countries, introduced rationing in order to be able meet the demands of their beleaguered populations. And the difficulty of obtaining a vast range of commodities had an important financial knock-on effect: it laid a solid foundation for massive inflation after the suspension of the gold standard.
The war brought with it the collapse of international finance, and there was an abrupt halt to the world’s stable monetary system when most of the gold standard countries suspended gold convertibility to protect their gold reserves from a run. The Scandinavian countries were no exception. On 2 August 1914, Riksbanken and Nationalbanken were relieved of the obligation to redeem their banknotes with gold, while Norges Bank followed suit three days later. It should be noted, however, that the obligation to buy gold was still in effect; what is more, this obligation was regulated by the convention of the SMU. During the autumn of 1914 all three countries went on to ban gold exports.8 The Scandinavian countries thus followed the pattern seen across the world, where the international exchange rate stability of the gold standard soon vanished and a period of violently fluctuating exchange rates followed; the value of each country’s currency was no longer determined by the fixed price of gold but by supply and demand on the world markets.
What were the effects on the intra-Scandinavian exchange rates? Compared to the Swedish krona, the Norwegian and Danish krone soon started to lose value on the markets. The Scandinavian currencies were still traded at par well into 1915, albeit with a small, yet increasing, exchange fee charged by the banks for buying Norwegian and Danish kroner as of 1914.9 With the exception of a short period in the spring of 1916, the Swedish krona was stronger than its Norwegian and Danish equivalents for the remainder of the war— frequently far, far stronger—as can be seen clearly in Figure 1. The other two currencies had a much more uniform development.10
What were the reasons for the diverging Scandinavian exchange rates? This issue has been the subject of some debate.11 The most likely explanation is the fortuitous trade cycle, which helped the Swedish export sector and thus increased the demand for Swedish kronor compared to the other two currencies.12 Norwegian and Danish exports, mainly generated by the two countries’ primary sectors, did experience price increases; however, compared with important Swedish exports such as lumber, pulp, iron, steel, and iron ore—many of which were crucial to the belligerents’ production of war materiel—the demand for Norwegian and Danish products was simply not as great.13 Hence, greater demand for Swedish products in turn pushed up demand for the Swedish krona compared with the other Scandinavian currencies, thus resulting in a higher market price.
FIGURE 1: Riksbanken’s avista DKK and NOK exchange rates October 1915– December 1918, monthly averages. Source, Riksbanken’s yearbook 1918.
Riksbanken watched these developments with increasing concern. In 1915 Riksbanken’s holdings of Scandinavian notes increased substantially, and the surplus of notes returned came to more than 72 million kroner. The main worry was the Danish notes; whilst the quantity of notes sent back to Norway actually dipped slightly between 1914 and 1915, the quantity of Danish notes increased massively.14 As we have seen, the central banks had started charging a fee on one another’s notes that reflected their market value, so why did these developments pose a problem for the Swedish monetary authorities? Was not the normal clearing mechanism for these central bank debts—payment in gold—a satisfactory solution for Riksbanken? The problem was that the increase was not confined to Riksbanken’s holdings of Norwegian and Danish paper money. The massive inflow of gold during the last weeks of 1915 and the first weeks of 1916 was a new concern for the bank.15
Riksbanken’s problem was the combined effect of the obligation to buy gold at the full legal value set by the SMU convention and the banks’ right to settle their debts with one another using Scandinavian gold coins. This was firstly a financial matter, for when the Swedish krona rose above its gold value on the currency markets, Riksbanken’s obligation to buy gold at the convention’s fixed price thus led to a loss compared to buying gold at market value: the 2,480 kronor the bank was obliged to pay for a kilo of gold could buy substantially more gold on the market. Additionally, the Norwegian and Danish kroner had lost value against the Swedish krona, and Riksbanken’s board of directors noticed that the rate loss of private business as well as the two other central banks could be transferred to Riksbanken with the help of the convention.16
The Scandinavian obligation to buy gold at par value meant there was risk of accelerated inflation. If the market value of one of these currencies, say Swedish kronor, increased to a certain level, it would be cheaper to send gold for example from Britain or Germany than to buy this currency on the market. As the seller of gold had a right to be paid in notes, such a gold inflow could increase the money supply, and thus in turn threatened to create higher prices.17 Moreover, belligerents could use the SMU’s convention and have gold minted in Norway or Denmark, or sell gold to one of those central banks, and so obtain Danish or Norwegian banknotes, which were still legal tender in Sweden.18 The gold problem was not solely a Swedish matter—all of the Scandinavian currencies were prone to this danger. However, because of the Swedish currency’s increased value compared to the other Scandinavian currencies and the massive gold inflows, Riksbanken was rightly the most concerned of the three central banks. What is more, prominent Swedish economists had begun to criticize Riksbanken’s gold policy in addition to questioning whether gold should be the standard of value. Their criticism was strengthened as long as Sweden’s gold reserves kept piling up while commodities were scarce and prices rose.19
To sum up, the gold problem stemmed from having a monetary policy with one foot in the new world and the other left in the old. Firstly, the value of each country’s money was suddenly decided by the market, since the obligation on the central banks to sell gold for banknotes at a fixed price had been suspended. Secondly, the central banks still had to buy gold for notes at the old, fixed price. As the value of gold started to fall and the price of paper money conversely started to increase, whilst commodities were scarce due to the war, the world’s hitherto immobile gold reserves began moving. Gold was apt to go where it would be valued the most, and the Scandinavian countries—with interesting export products and an obligation to buy gold at a fixed price still in effect—were a natural haven. Of these three, Sweden stood out as the best option because of its products’ indispensability to the belligerents’ war industry. The inflow of gold risked causing massive inflation while at the same time commodities and necessities flooded out, to the extent that in Sweden there was a fear that the country would share the fate of King Midas. Hence, for the Swedes it was necessary to end the gold inflows, and if trade was to be maintained, goods thus had to be paid for with goods, not metal. To solve the gold problem it was necessary to change the legislation regulating Riksbanken, but even though Sweden could prevent the inflow of gold directly from other countries, the SMU was still in effect. This meant that without the help of its union partners, Sweden could block neither Scandinavian gold directly nor other countries from sending gold indirectly via Denmark or Norway. By the end of 1915 the SMU was obviously becoming a straitjacket for Riksbanken. But by then wheels were already in motion.
The first step – the Swedish gold blockade
In the first days of 1916 Riksbanken resumed gold convertibility and again sold gold for notes at the old, fixed price. The fear that had set off the emergency measures of August 1914 was long gone: the Swedish notes were quoted higher than their par value and a gold run was hardly a likely outcome. Faced with the option of either holding a kilo gold or 2,480 Swedish kronor, almost anyone would have preferred the latter as the market price made it more valuable. Even though the market value of the Norwegian krone in December also was higher than its gold value, Norges Bank did not think the time right to follow the Swedish initiative; on the contrary, it sought a return to more normal conditions.20 However, the Swedish response was hardly encouraging; on 8 February 1916 the Riksdag relieved Riksbanken of its obligation to buy gold at par value in an attempt to ‘prevent an uncontrolled increase of the gold reserves’, putting the so-called ‘gold blockade’ into effect.21 As Norway and Denmark had not been subjected to comparable gold inflows, Sweden’s change of policy was not followed by her two neighbours—gold was still bought at the convention’s fixed price.22
To make the Swedish gold blockade effective the convention had to be altered. The Swedish authorities thus had to make sure that Denmark and Norway would follow their lead, and Riksbanken took steps to secure their cooperation. Riksbanken sent a professor of economics, David Davidson, to Copenhagen in an attempt to convince Nationalbanken to fall into line.23 Perhaps this was also a prenegotiation attempt to obtain the cooperation of Nationalbanken prior to tackling Norges Bank on a two-against-one basis, in the hope of undermining the hard stance against altering the conditions for gold shipments that had been signalled by Norges Bank.24 A meeting between representatives of the three central banks was eventually held in Stockholm in February 1916. The outcome of these proceedings was that the right to mint gold was temporarily suspended, Norges Bank and Nationalbanken agreed to recommend to their respective governments that the obligation to buy gold at the fixed price should be suspended,25 and gold was only to be exported after ‘consulting’ the receiving central bank. Although Riksbanken thus to some extent got its way in implementing Sweden’s gold policy without terminating the convention of the SMU, the two other banks had not agreed to anything awkward.26 The manner in which Norway and Denmark implemented the ‘gold blockade’ was reluctant, and ultimately prompted by financial considerations rather than by an urge to uphold the partnership. Hence, it was not until the Norwegian gold reserves had risen to unprecedented levels a month later that Norges Bank wanted to remove the requirement to buy gold at the fixed price.27 Riksbanken had not either been able to make the gold blockade effective; as neither Nationalbanken nor Norges Bank was interested in removing the possibility of settling their debts in gold, the door was still wide open.
Second-class cooperation
The unsuccessful Swedish efforts to implement an adequate gold blockade policy would have been striking for anyone with access to Riksbanken’s vaults in the autumn of 1916; having lain more or less stable since the turn of the year, from September the gold reserve had begun to increase yet again. The loose ends of the gold problem and the working of the SMU led Riksbanken to initiate a new meeting in October 1916, this time in Gothenburg. Riksbanken’s agenda included a Danish suggestion from May—a common gold policy. Other topics included limitations on the redemption of banknotes and the question of imposing interest rates on the formerly interestfree central banks accounts. Although the exchange rates seemed to have stabilized at close to par during the spring months, developments since the summer had again contributed to divergent rates—most noticeably the value of the Swedish krona rose massively compared to the other two (see Figure 8.1). Thus, exchange rates became an important topic at this meeting, but not surprisingly it was made known to the press that one of the solutions proposed by the Norwegian and Danish representatives—settlement in gold—was not considered helpful by Riksbanken.28
Did Riksbanken come any closer to solving its problem? The three central banks were actually able to reach an agreement on some points, including that a fee could be charged when drawing on the other central banks. If the central banks were in future to buy gold on the international markets, they agreed to inform one another of price, amount, and where it was bought; similarly, information was to be shared if any were to mint gold. Nationalbanken was in addition asked to help arrange quotations of the Swedish krona on Copenhagen stock exchange, while Norges Bank was asked if interest of 3 per cent could be paid on amounts exceeding 3 million kroner/kronor when clearing the two central banks’ mutual line of credit.29 Although the Norwegians rejected this last request—which actually implied an increase in the normal interestfree debt from the 1905 agreement with Riksbanken—the other points were agreeable to both Norges Bank and Nationalbanken, and it seemed as if a common gold policy was largely established.
However, even though agreeing to share information about gold purchases and minting may seem like the intensification of Scandinavian gold policy cooperation, it might as well be seen as the opposite. The fact that the three central banks were to inform one another whenever buying or minting gold can be regarded as a rather suspect method of creating a barrier to the other countries (for which read Denmark and Norway) from speculating in gold shipments instead of paying the (Swedish) currency’s full market value. Moreover, when Norges Bank refused to countenance paying interest rates on mutual credits, it simply argued that a debt amounting to more than 3 million kroner was unlikely to be long-lived, as long as the ‘clearing could be done by gold or other means’.30 At this point the mutual lines of credit had not yet become a problem. However, under pertaining conditions there were two solutions available if further imbalances should ever develop: increase the debt, or settle it in gold. The Norwegian and Swedish central banks were on a collision course over this issue, and by its answer the Norges Bank seems simply to have ignored Riksbanken’s gold fears. Thus Riksbanken’s more acute problem had not been solved by this meeting either: Scandinavian central bank debts could still be settled by means of gold shipments.
Third time’s the harm
The effects of the unsuccessful Swedish efforts to ward off gold inflows were evident almost immediately after the October meeting, as both Norges Bank and Nationalbanken sent Riksbanken 5 million kroner in Scandinavian gold.31 Even though there were some improvements in Riksbanken’s accounts at the end of the year, including the halving of 1915’s exceptionally large number of returned Danish notes, these were paltry sums compared to the gold accounts. Scandinavian gold coin had increased by nearly 18 million kroner and foreign gold by more than 40 million kronor during 1916. Although most of the increase predated February’s gold blockade, the increase in the aftermath of the blockade was nonetheless 23 million. Thus Riksbanken explicitly stated that the gold received after 8 February was mostly Scandinavian gold coin ‘to which the provisions of the law […] are not applicable’.32
To the uninformed contemporary observer, Nationalbanken was to blame for the massive inflow of unwanted Scandinavian gold into Riksbanken’s vaults. The reason was that most of the inflow was as Danish gold coin, and in contrast to Norges Bank the Danish central bank had not stopped minting gold. This was the main reason why Swedish monetary authorities deemed it crucial to limit or cease minting when trying to solve its gold problem. Nationalbanken was exposed to harsh criticism when it was discovered that Swedish measures in effect came down to an attempt to stop Danish gold coin. The criticism was first made by Swedish economist Gustav Cassel in 1917; however, it was echoed by Cassel’s partisan in Denmark, the economist Axel Nielsen.33 In fact the outcry was so massive that Nationalbanken felt the need to defend itself publicly. However, it seems that it was not until Marcus Rubin’s second in command, Carl Th. Ussing, published his book about Nationalbanken in 1926 that the beans were spilled; the real offender had been Norges Bank.34 According to Ussing, the Norwegians wanted Nationalbanken’s debt to Norges Bank to be settled by sending Danish gold to Riksbanken in order to cover Norges Bank’s debt there. To Nationalbanken this was nothing more than a matter between Norges Bank and Riksbanken, and the Danish bank could obviously not refuse to settle its debts in the way its Norwegian creditor wanted—in legal tender.
Although Victor Moll in his New Year’s greeting to Karl Gether Bomhoff hoped to come to a ‘better understanding between the three central banks’ and with ‘good will’ overcome the differences in their cooperation, any good will the Scandinavian central bank cooperation enjoyed was running out by 1917.35 In the middle of January 1917, financial considerations once again showed its prominence over nice phrases when Norges Bank found it necessary to pay its debt to Riksbanken by sending 3-4 million kroner of gold. Riksbanken replied by admitting that it was not in a position to refuse the gold, but it appealed to Norges Bank’s sense of decency, and asked not to be forced to buy expensive gold considering the market price.36 However, as Norway did not have any commodities to spare it was necessary to cover the imbalance with gold—in Bomhoff’s words, the present situation had become ‘unbearable’.37 A few days later he forced through the gold shipment.38
The situation clearly had become unbearable to Moll as well; the need to protect Sweden from expensive gold and inflation made him recommend governmental negotiations to limit each country’s right to mint gold under the convention.39 Thus, in contrast to the two meetings of 1916, the meeting of April 1917 became an official meeting with delegates representing the three countries’ governments, with the obvious objective of altering the forty-year-old convention. Before negotiations commenced Riksbanken leaned heavily upon Professor Davidson, but as neither Nationalbanken nor Norges Bank was interested in terminating or suspending any of the convention’s paragraphs as suggested by Davidson, negotiations came to revolve around a proposal from the Norwegian delegates.40 Norway was willing to stop gold shipments on the condition that the three central banks would put credit on reasonable terms at one another’s disposal, and this standpoint was supported by the Danish delegates. As sketched out, the export ban was to be strictly maintained, although gold could be sent on agreement; if one of the bank’s debts exceeded 1 million kroner the creditor had the right to be sent gold; and a small interest rate would be charged if the deficit exceeded 3 million kroner, but if this rate were to exceed a certain level the debtor had the right to send gold. An agreement was reached only after omitting the paragraph defining interest rates—to be decided by mutual agreement between the various banks. Moreover, Moll had to urge Davidson to agree to the proposal and allowed him to state for the minutes that he did not find the agreement sufficient to achieve the aims of Sweden’s gold policy.
Just a few days before leaving Copenhagen, Marcus Rubin had written that he was heading for Stockholm, where ‘“Scandinavianism” is to be submitted to a small practical test—which I doubt it will past’.41 Did Scandinavianism pass the test, considering that an agreement actually was concluded? In the very short run one can probably regard the incident as Swedish gold policy being sacrificed on the altar of Scandinavianism. Although the preferred policy of the Swedish monetary authorities was to alter or suspend parts of the convention, the proposed changes were unacceptable to the SMU partners. To carry out its gold policy, Sweden thus would have to abandon the SMU, and the political costs of doing that were simply too high. However, Rubin was in many respects correct: ‘Scandinavianism’ barely passed the test, despite the agreement. The solution was far from sufficient as the basic problem had not changed—gold could ultimately still be sent to cover central bank debts. The only difference was that the agreement paved the way to make greater use of lines of credit. However, to the Swedish monetary authorities this was not satisfactory either; in fact, curbing international credit was precisely one of the matters that were at the centre of Swedish economic policy. Thus the economic and financial costs of upholding the SMU were almost as high as the political costs of abandoning it, and there were few indications of Scandinavian brotherhood in the interaction between the three banks. The final months of 1917, annus horribilis of Scandinavian monetary cooperation, was to bear witness to this.
1917—annus horribilis
The hardship inflicted by the war worsened in 1917, as a number of circumstances dramatically altered the economic conditions of the northern neutrals. Probably the most important was Germany’s decision to unleash unrestricted submarine warfare from 1. February, which made transportation, and thus obtaining necessities, extremely difficult. All the Scandinavian countries duly imposed rationing on foodstuffs; the extremity of the situation even led trade representatives from the three countries to meet and agree to help one another by exchanging commodities.42 To top it all, the gold problem was aggravated, as Sweden more than ever needed commodities in return for its commodities, while Norway had no commodities to spare—only gold.
The April 1917 agreement is usually seen as ending the turmoil and differences between the three central banks while solving Sweden’s gold problem.43 Nothing could be further from the truth. Although the three central banks had managed to come to terms bilaterally on the interest rates on debts, the quarrelling between Norges Bank and Riksbanken peaked this year.44 The problem was that Norges Bank’s debt to Riksbanken increased massively once they had agreed on the mutual lines of credit—and this to a degree that obviously was not expected by Riksbanken when concluding the April agreement. A central element in the bickering was the inability of Norges Bank and Riksbanken to reach a decision on which agreement should be adhered to. For the rest of the year Norges Bank solely referred to the agreement of April 1917 and the line of credit so obtained; Riksbanken felt this was unreasonable, and that the old agreement of 1905 with fees on drafts was more appropriate—particularly as Norges Bank’s debt was constantly increasing. Norges Bank’s debt to Riksbanken soon surpassed 3 million kroner—a sum that had been considered virtually impossible in October 1916—and by the end of June it had reached 7.6 million, a fact which made Moll write of his concerns to Bomhoff as this had ‘attracted some attention’.45
During the summer months the situation became most uncomfortable, as the temperature in the correspondence between the two banks increased; the banks simply could not work out their differences. ‘Our “Nordic cooperation” ’, Bomhoff complained to the deputy governor Haakon Monsen, ‘does not always seem to be particularly effective!’46 Thus by the end of July, Norges Bank’s debt had increased to 8.2 million kroner, and by the end of August more than 10 million kroner. As Riksbanken believed the debt would only increase further, it abandoned its attempts to resist gold shipments from Norway and asked to be sent 5 million kroner in Scandinavian gold in September 1917.47 This brought no relief, however; things in fact worsened. In the numerous letters between Moll and Bomhoff, the two conflicting understandings of the situation were repeated over and over again; the same arguments were made, whilst the tone deteriorated. Both banks stuck to their guns, while the Norwegian debt piled up. In the middle of October, Riksbanken yielded once more and agreed to accept another 10 million kroner in gold.48 After this the letters became increasingly poisonous, and by the end of October Moll and Bomhoff no longer sent each other personal letters with polite greetings.49 The banks could not find any common ground and Norges Bank’s debt remained large; however, Riksbanken relented yet again, and accepted another 5 million kroner in Scandinavian gold in November.50 At this point Riksbanken’s repeated concessions made Norges Bank ask if it was to be assumed that Riksbanken in the future would accept Scandinavian gold when being drawn on, in which case ‘the difficulties in essence would be at least temporarily removed’.51 The reply was short but positive; Riksbanken gave in by accepting gold beyond the last 5 million kroner as well.52
At the end of 1917 Riksbanken thus chucked its cards in and Norges Bank won the zero-sum game. The abrupt halt in gold shipments the following year did not mean that Norges Bank and Riksbanken finally had come to terms; rather, it most likely reflected the smaller fluctuations in the two currencies in relationship to changes in the world’s currency conditions and in the Scandinavian trade balance.53 Over the course of 1917—a disastrous year for Scandinavian monetary cooperation—some 34 million kroner in Scandinavian gold coin was sent to Riksbanken from Norges Bank.54 It seems likely that all of this gold was received in the teeth of Riksbanken’s wishes, and during the year the Swedish central bank’s gold reserve increased by more than 60 million kroner despite the attempts to solve the gold problem by implementing a policy that could shelter the country from gold inflows.55
Concluding remarks
The First World War effectively ended the SMU. It is often described as having been doomed from the moment gold convertibility was suspended in August 1914; however, the manner in which the union disintegrated reveals some interesting issues to do with international cooperation at times of stress. There were few obstacles to founding the union and making it work during the heyday of the international gold standard. Even the political turmoil of 1905 was overcome. However, upholding the SMU when its foundations were severely damaged—by the disbanding of the three currencies exchange rates and their relationship with gold—was a completely different matter. Thus the pursuit of national self-interest, be it financial or economic, became paramount for all three central banks when the world changed so fundamentally in 1914. The SMU was a relic of a different epoch, belonging as it did to the pre-1914, liberal era, but to the contemporaries this was not evident. The doomed efforts to maintain the partnership reflected the belief that as soon as international hostilities ended things would go back to normal.
Sweden’s wishes to change the SMU’s rules could be accommodated as long as the changes were to the satisfaction of all three parties. When desperate Swedish needs collided with desperate Norwegian needs the result was a stalemate, overcome only a slow return to ‘business as usual’. The need to do something more drastic by altering the fundamentals of the SMU seems to have been postponed by the urge to maintain the union as a beacon of Scandinavianism. This is probably the reason why the SMU survived far beyond the 1920s—when efforts were made to develop international economic cooperation beyond the narrow scope of Scandinavianism—and existed as an anachronistic survivor of the old world, until it was finally terminated in 1973. However, the lesson of discussions about collaborating to solve the Scandinavian Monetary Union’s gold problem during the First World War is painfully clear: money talks.
Notes
1 Axel Nielsen, Den skandinaviske møntunion. Et historisk rids (Copenhagen: Børsens forlag, 1917), 74, ‘Hvad der er tilbage af den, er imidlertid vanskeligt at faa Øje paa. Tilsyneladende er alt det der i det 19de Aarhundrede gennem mer end 30 Aars utrættelig Virken byggedes op af økonomisk Samarbejde, nu lagt øde i Løbet af blot to Aar, og det paa et Tidspunkt, hvor e n n y B e v æ g e l s e for en mere intim økonomisk Samvirken er i Vækst.’
2 Marcello de Cecco, ‘European monetary and financial cooperation before the First World War’, Rivista di storia economica, 9 (1992), 67.
3 The following section is based on Robert J. Bartel, ‘International Monetary Unions: the XIXth Century Experience’, Journal of European Economic History, 3 (1974), 689-704; Eli Hecskcher, ‘Sweden in the World War’, in Heckscher et. al. (eds.), Sweden, Norway, Denmark and Iceland in the World War (Carnegie Endowment for International Peace; New Haven: Yale University Press, 1930); Ingrid Henriksen & Niels Kærgård, ‘The Scandinavian Currency Union 1875–1914’, in Jaime Reis (ed.), International Monetary Systems in Historical Perspective (London: Macmillan, 1995); Nielsen, 1917; Nicolai Rygg, Norges Banks historie II (Oslo, 1954); Krim Talia, The Scandinavian Currency Union, 1873–1924 (diss.; Stockholm: Stockholm School of Economics, 2004); Lars Fredrik Øksendal, ‘The impact of the Scandinavian Monetary Union on financial market integration’, Financial History Review, 14 (2007a), 125-148; L. F. Øksendal, Essays in Norwegian Monetary History, 1869–1914 (diss.; Bergen: Norwegian School of Economics and Business Administration, 2007b).
4 See, for example, Hans Chr. Johansen, ‘Om at skrive bankhistorie’, Historisk Tidsskrift [Denmark], 90 (1990), 383-404. I am grateful to Martin Austnes for making me aware of this article.
5 Which thus for all practical purposes can be extended to 1910 (Henriksen & Kærgård 1995, 109).
6 Øksendal 2007a.
7 Gunnar Wetterberg, Pengarna & Makten (Stockholm: Sveriges Riksbank/ Atlantis, 2009), 233–4.
8 Rygg 1954, 470.
9 Ibid. 471.
10 Jan Tore Klovland, ‘Historical exchange rate data 1819–2003’, in Øivind Eitrheim, J. T. Klovland & Jan Fredrik Qvigstad (eds.), Historical Monetary Statistics for Norway 1819–2003 (Oslo: Norges Bank, 2004), Table A2.
11 Until recently, economic historians have mainly focused on the money expansion, arguing that as Norway and Denmark issued more paper money than Sweden, the value of the Swedish krona was higher simply because it was more rare. There is, however, reason to believe that this explanation is incorrect, especially in view of the years up to 1917. I am elaborating these debates in more detail in my forthcoming Ph.D. thesis.
12 See Talia 2004, 179–80.
13 Heckscher 1930, 173.
14 Riksbanken, Årsbok (Stockholm: Sveriges Riksbank, 1915).
15 After an increase between December 1914 and the last weekly report in February 1915, Riksbanken’s reserve hardly moved until mid-December 1915 (see the Riksbanken’s weekly reports 1914–1916 in Ekonomisk Tidskrift).
16 Talia 2004, 174
17 For the Norwegian experience, see Monica Værholm and L. F. Øksendal, ‘Leaving the Anchor: Monetary Policy in neutral Norway during the First World War’, International History Review, 32/4 (2010), 661-686.
18 Erling Olsen & Erik Hoffmeyer, Dansk Pengehistorie 1914–1960 (Odense/ Copenhagen: Danmarks Nationalbank, 1968), 49; Carl Th. Ussing, Nationalbanken 1914–1924 (Copenhagen: Gads Forlag, 1926), 88.
19 The severest critic was undoubtedly Gustav Cassel, for example, in his 1917 book Dyrtid och sedelöverflöd; however, the most influential of all contemporary critics was the economics professor David Davidson.
20 National Archives, Oslo (NA), Norges Bank’s archives (NB), Archive of Board of Directors I (DI), Bomhoff’s book of copies 1900–1918 (Bb0001), letter from Bomhoff to Moll, 4 January 1916.
21 NA, NB, Archive of Board of Directors II (DII), D0641, letter from Ministry of Finance to Norges Bank’s Board of Directors, 14 February 1916, ‘avværge en ukontrollert ökning av guldforraadet’; Talia 2004, 174.
22 Denmark’s gold reserve had only shown a minor increase in 1915, and even though the Norwegian reserve had increased substantially this was mostly due to events up until August. Statistics from Ekonomisk Tidskrift.
23 Ussing 1926, 88; Olsen & Hoffmeyer 1968, 49.
24 NA, NB, DI, Bb0001, Bomhoff to Moll, 16 February 1916.
25 Ussing 1926, 88–9.
26 Ibid.
27 NA, NB, DI, E0017, Norges Bank to Ministry of Finance, 30 March 1916 (draft).
28 Rygg 1954, 477.
29 Riksbanken to Nationalbanken and Norges Bank, 11 October 1916 (in Ussing, 1926, 92–3 & Rygg 1954, 477–8, respectively).
30 Rygg 1954, 478, ‘oppgjør kan skje ved gull eller på annen måte’.
31 NA, NB, DI, E0017, ‘Valutaresume pr. 30/11-16’, Norges Bank to Riksbanken, 26 October 1916, and from Norges Bank to Nationalbanken, 30 October & 10 November 1916.
32 Riksbanken, Årsbok (Stockholm: Sveriges Riksbank, 1916), ‘på vilket bestämmelserna i lagen […] icke hade någon tillämpning.’
33 Ussing 1926, 101 ff; Olsen & Hoffmeyer 1968, 52 ff.
34 Ussing 1926, 94–5.
35 NA, NB, DII, D0642, Moll to Bomhoff, 31 December 1916, ‘ökadt samförstånd de tre centralbankerna emellan’, ‘god vilja’.
36 Letter from Moll to Bomhoff, 19 January 1917, D0642-DII-NB-NA.
37 NA, NB, DII, D0642, Bomhoff to Moll, 20 January 1917, ‘Den nuværende tilstand blir snart utaalelig’.
38 NA, NB, DII, D0642, Bomhoff to Moll, 30 January 1917.
39 NA, NB, DII, D0642, Moll to Bomhoff, 22 & 30 January 1917.
40 The following is based on NA, NB, DII, D0641, the official minutes of the Stockholm meeting, 17–19 April 1917, and the undated summary by Bomhoff and Rygg of the Scandinavian conference on those dates regarding the currency convention; NA, NB, DII, D0642, David Davidson, ‘P.M. rörande den skandinaviska myntkonventionen’, enclosed with Moll to Bomhoff, 13 February 1917; Ussing 1926, 99.
41 Marcus Rubins Brevveksling, ed. Lorenz Rerup (Copenhagen: Rosenkilde og Bagger, 1962), iii. 300, Rubin to Kr. Erslev, 19 April 1917, ‘“Skandinavismen” skal underkastes en lille praktisk Prøve—som jeg tvivler om, den bestaar’.
42 Einar Cohn, ‘Denmark in the Great War’, in E. Hecscher (ed.), Sweden, Norway, Denmark and Iceland in the World War (Carnegie Endowment for International Peace; New Haven, Yale University Press, 1930), 470–1.
43 See, for example, Bartel 1974, 702; Michael Bergman, Stefan Gerlach & Lars Jonung, ‘The rise and fall of the Scandinavian Currency Union 1873–1920’, European Economic Review, 37 (1993), 515.
44 See NA, Archives of the Ministry of Finance (FIN), Finance Office (FOc), Dca0099, Riksbanken and Nationalbanken to Norges Bank, 21 & 24 April 1917 respectively, and replies from Norges Bank to Riksbanken and Nationalbanken, 24 & 26 April 1917 respectively.
45 NA, NB, DII, D0642, Moll to Bomhoff, 2 July 1917, ‘väckt en viss uppmärksamhet’; Rygg 1954, 486; the statistics here and in the following section are from Ekonomisk Tidskrift 1917.
46 NA, NB, DII, D0642, Bomhoff to Monsen, 21 July 1917, ‘Vort “samarbeide i Norden” synes ikke altid at ville bli synderlig virkningsfuldt!’
47 NA, NB, DII, D0642, Moll to Bomhoff, 11 September 1917.
48 NA, NB, DII, D0642, Moll to Bomhoff, 17 October 1917.
49 See NA, NB, DII, D0642, Norges Bank and Riksbanken correspondence from 20 October 1917.
50 NA, NB, DII, D0642, Riksbanken to Norges Bank, 12 November 1917.
51 NA, NB, DII, D0642, Norges Bank to Riksbanken, 16 November 1917, ‘vanskeligheterne, i det væsentlige, iethvertfald foreløpig være ryddet tilside’.
52 NA, NB, DII, D0642, Riksbanken to Norges Bank, 19 November 1917.
53 Rygg 1954, 489; the value of Norwegian imports from Sweden increased by 36 per cent from 1917 to 1918, while exports increased by 73 per cent (see Norges Offisielle Statistikk XII 245, (Historisk Statistikk 1968; Statistisk Sentralbyrå, 1969), Table 165).
54 Rygg 1954, 488.
55 Ekonomisk Tidskrift, 1917–1918.