Trade and Money in British West Africa, 1912–1970

Introduction

In his classic work on the economic history of West Africa, first published in 1973, Tony Hopkins argued that “the steamship and modern money had a profoundly unsettling effect on West African trade”.¹ According to Hopkins, these two changes reduced the costs for new entrants to international trade, increasing competition and extending access to international markets to a wider range of African producers. African smallholders responded to rising demand from industrializing Europe for raw materials such as groundnuts and palm oil. Cheaper consumer goods produced in Europe, particularly cotton textiles from Britain, made imported goods accessible to Africans from a wider range of income groups. European currencies introduced through this trade could be used in a wider range of transactions than precolonial currencies like cowries or manilas, which entered a slow decline.

Other scholars have disputed this characterization of West Africa’s economic history, arguing that the changes to the economic and monetary systems of West Africa were not as far-reaching or as fast as Hopkins claims. Ofonagoro argues that while European and, later, colonial currencies were used for international trade and official transactions such as paying taxes, the persistence of precolonial currencies in domestic economies raises questions about the extent of the “currency revolution”, as Hopkins had described it in an earlier article.² Ofonagoro argues that despite the best efforts of the colonial state to demonetize cowrie shells and manilas, it was not until price inflation during the 1940s reduced their value for small purchases that they were fully displaced by colonial currencies. Others emphasize the political significance of the failure of colonial currencies to displace precolonial currencies until well into the twentieth century. Saul describes the continued use of cowries in French West Africa as part of the “complex overlay of confrontations and accommodations entailed in the history of colonialism”.³ Subsequent research on a variety of cases has confirmed that the adoption—or not—of European or colonial currencies depended on a range of economic, political and pragmatic factors.⁴

These two views can be reconciled through a closer look at the connections between Africa and the international monetary system during the nineteenth and twentieth centuries, and the changing factors which influenced African demand for official colonial and post-independence currencies, as opposed to precolonial currencies. One such factor was growing consumption of imported goods, along with the widening and deepening of the African market for those goods. European traders initially continued to accept precolonial currencies—indeed some, like the Niger Company, actively resisted any proposal to switch as its agents viewed their knowledge of indigenous currencies as a source of comparative advantage.⁵ However European—and, subsequently, colonial currencies—eventually became the key to accessing imported goods. The continued use of precolonial currencies suggests that they were in demand for other types of transactions in the domestic market.

European currencies imported into West Africa were by the early twentieth century replaced by currencies issued specifically for colonial use, but which were exchanged at fixed rates for metropolitan currencies.⁶ In British Africa, this transition occurred with the establishment of the West African Currency Board (WACB) in 1912. The WACB was charged with issuing the West African pound in the four British territories in West Africa (Nigeria, the Gold Coast, Sierra Leone, and Gambia). After independence, former British colonies abandoned the West African pound in favor of their own national currencies. However, these new currencies were initially backed by sterling reserves and issued at parity with the pound sterling, thus retaining many of the essential features of the currency board system.⁷

Helleiner argues that one of the principal goals of introducing colonial currencies was reducing transaction costs in international trade, along with a reduction in transaction costs within each colony.⁸ As this article will show, the aim of streamlining international trade took precedence over the elimination of transaction costs within colonies. This article uses new data on seasonal fluctuations in the circulation of West African currencies just after independence, along with archival records on the development of colonial currency systems, to argue that demand for colonial and post-independence currencies came from, on the one hand, traders purchasing cash crops, and, on the other, from the African producers who wanted to purchase imports. The presence of large seasonal fluctuations in the demand for money has long been known from qualitative statements by officials working for the colonial government and banks operating in the colonies, but the quantitative study of such fluctuations has often been hindered by data limitations. By using monthly data on currency circulation in the first years after independence, the article provides one of the first measurements of the extent of seasonal fluctuations in demand.

The article proceeds as follows. The next section gives a brief overview of the monetary history of West Africa from the nineteenth century through independence and reviews the debates surrounding the introduction of colonial currencies. It places West Africa’s history in a global context, examining the impact of changes in the international monetary system. The section to follow presents data on seasonal cycles in the circulation of official currencies in West Africa in the years following independence. The article then presents some speculations as to why seasonal fluctuations remained so large in West Africa into the 1960s. The final section concludes by illustrating how taking seasonal fluctuations into account can help reconcile the two arguments made about the impact of colonial currencies.

Trade, Economic Growth, and Changing Monetary Systems in West Africa

The nineteenth and twentieth centuries were a period of momentous economic and political change in West Africa. At the beginning of the period, the region’s trade with Europe was still dominated by the centuries-old Atlantic Slave Trade, which reached its peak in the late eighteenth century and did not finally end until the middle of the nineteenth. The export trade in slaves was displaced by the export of agricultural produce, minerals and forest products, which expanded sharply over the course of the nineteenth century, exceeding slave exports in terms of value by the 1830s. The response of African producers to growing demand from industrializing economies in Europe combined with technological innovations like the steamship and the railway which reduced the costs of transport resulted in the rapid expansion of export production. Creeping European intervention in the region’s politics culminated in expansive territorial claims made during the “Scramble for Africa” from the 1880s, followed by just over a half-century of colonial rule. By the 1950s, colonial powers began their almost equally rapid retreat from the region, and newly independent African states grappled with the challenges of independence and development in a world complicated by Cold War dynamics and economic upheaval.

How did these changes affect African monetary systems? Though Africa’s monetary history is often siloed away from the monetary history of Europe and the US, it followed many of the same trends. Contrary to the claims of colonial officials and early historians, Europeans did not “introduce” money to West Africa. African monetary systems before colonial rule were characterized by multiple and overlapping circuits of currency usage not generally bound by the jurisdiction of specific polities. While there were state-sanctioned currencies—like the gold dust used by the Asante kingdom—they circulated alongside others. Items as diverse as cowrie shells, iron manilas, kissi pennies, cattle, and cloth—to name just a few examples in addition to the gold dust—all circulated as currencies in both domestic and international trade. The use of particular currencies was often associated with specific types of transactions—small purchases in a market, for example, might be made with cowrie shells while large payments for bridewealth would use larger units such as cattle.⁹

Such overlapping currency circuits were not unique to Africa before the twentieth century. In his global history of money, Akinobu Kuroda documents the prevalence of such circuits across Europe and Asia since the medieval period.¹⁰ Eric Helleiner also shows that national currencies as we know them today were across most of the world an innovation of the nineteenth century, when they emerged gradually from a combination of rising state capacity and political nationalism.¹¹ Prior to that, however, the circulation of multiple currencies served a variety of purposes for preindustrial economies. They allowed for transactions at a variety of different scales, and helped mitigate against potential shortages in one currency commodity or another. Guyer argues that multiple currencies were often a way buffering the effects of currency instability and providing a means of accumulation.¹²

In West Africa, this transition from multiple to national currencies was punctuated by the rapid expansion of trade with Europe and, particularly, the rise and fall of colonial rule. The first impact of the growing Euro-African trade was merely to add coins minted by European states to the mix of currencies used in the region. They did not displace cowries and other currencies but instead were often associated with particular trades—the French five franc coin, for example, became strongly linked to the groundnut trade from the Gambia and Senegal.¹³ These links were often defined by commodity rather than by European spheres of influence—the five-franc coin, for example, circulated just as easily in British colonies where groundnuts were produced as in French territory.

West Africa’s expanding trade needs to be set in the context of broader developments in the global economy. Findlay and O’Rourke argue that the nineteenth century marked a “dramatic break with the past”, in terms of the expansion of trade and the integration of global markets.¹⁴ Linked to this expansion in trade were the two factors Hopkins referenced with regard to West Africa, money and transport. New transport technology facilitated the movement of goods around the world at unprecedented speed and low cost. In addition, the adoption of the gold standard by an increasing number of countries during the 1870s stabilized exchange rates and reduced the information costs of international trade. The phrase “gold standard” has had a variety of definitions in different periods and context. The earliest metallic standards referred to the minting of coins with a certain percentage of gold content. Gold coins did not circulate during the late nineteenth and early twentieth centuries. Instead, the gold standard of this era was more properly termed a gold exchange standard in which countries committed to fixing the value of their currency to a certain amount of gold or gold-convertible foreign exchange. Under this system, governments guaranteed that their currency could be exchanged on demand for that amount. Fluctuations were managed largely through the movement of capital between major markets in anticipation of central bank intervention, often quickly enough and in large enough quantities the government intervention proved unnecessary.¹⁵

The stability and credibility of the pre-war gold standard was made possible by three conditions prevailing in Europe at the time: 1) open and flexible markets; 2) states which prioritized currency and exchange rate stability, and; 3) limited political pressure to use monetary policy for other aims.¹⁶ Such pressure did emerge from proponents of bimetallism, such as “free silver” campaigners in the United States during the 1890s, who argued that a bimetallic standard would increase the competitiveness of American exports and allow for an expanded money supply which would benefit farmers following the deflation of the 1870s.¹⁷ However, with the limited franchise which existed in Europe and North America at the time, such pressure could be contained.

Emily Rosenberg argues that the extension of the gold-exchange standard to peripheral economies was viewed in financial capitals as central to the growth of trade, economic development, and sound financial management in the periphery.¹⁸ In policy terms, “gold standard diplomacy” was put into practice through several channels. These included controlled loans, in which independent peripheral countries had to accept foreign management of their finances and monetary policy. From the perspective of European and American financiers, promotion of the gold-exchange standard in the periphery was essential for reducing the information costs of trade. Trade with countries using a silver standard or other currency systems involved considerable exchange rate risk, which adoption of the gold standard eliminated. Helleiner argues that “the most important monetary goal consistent across all colonial powers in the late nineteenth and early twentieth centuries was an economic one: to reduce currency-related transaction costs associated with trade and investment between the home country and the colony”.¹⁹ In a colonial context, the standardization of colonial currencies was part of a broader colonial development policy designed to increase growth through trade, including investments in transport infrastructure and fiscal policies concentrated on balanced budgets.²⁰

From the perspective of the periphery, the issue of monetary reform was more complex. Under a gold-exchange standard, maintaining parity with the gold-backed currency required that the supply of currency be reduced when the balance of trade was negative. This could lead to deflation, as it had in the case of the US in the 1870s, with potentially severe short-term consequences for welfare. Political voice under colonial governments was, however, extremely limited for the majority, which meant that colonial governments—like their independent counterparts—could prioritize open markets and exchange rate stability.

Despite the enthusiasm for gold standard diplomacy during the late nineteenth and early twentieth centuries, the economic and political changes during the nineteenth century had a gradual impact on the demand for money in West Africa, both in terms of its quantity and its form. During the nineteenth century, a variety of what Hopkins refers to as “transitional currencies” were used as the medium of exchange in both international and domestic commerce.²¹ This included gold (in various forms), cowrie shells, strips of cloth and copper and iron rods. Such currencies—which were not regarded as transitional to those using them—were also used as a store of value. Ofonagro argues that, in precolonial Africa, “the principal method of saving money was hoarding, and in the hinterland most homes had a treasury room where their hoards of local currencies, such as manillas, brass rods, and cowries were stored”.²²

Initially, European traders imported cowries and manilas and other commodity currencies to pay for African exports. Increasingly, however, there was a shift towards using European currencies such as British sterling or the French franc in international trade. Whether this change was primarily due to external coercion or internal demand remains the subject of debate. Hopkins argues that the change was largely demand-driven, though facilitated by the colonial state. He claims that “it was no coincidence that francs and shillings spread in areas where legitimate exports were developing most quickly, and it was no coincidence either that low denomination coins were in great demand, for they were an indication of the growing importance of small producers and traders in the new export economy”.²³

In British Africa, by 1912, the volume of imported British token coins had increased to such an extent that the British government was concerned about the possibility that the sudden repatriation of token coins from West Africa would lead to their depreciation in Britain, and cash-strapped West African governments were making repeated requests for a share of the seigniorage revenue gained by the British treasury from rising demand for the token coins.²⁴ The key change came with the introduction of the West African pound, issued by the WACB from 1913. Issues of the West African pound had to be backed by a 100 percent sterling reserve and could only be made in exchange for sterling deposits into the WACB’s account in London. This system had the advantage of maintaining a stable value for the West African pound and thereby eliminating any exchange rate risk. However, it had the disadvantages of being what Huff in reference to Southeast Asia describes as “an extreme form of the gold standard,” in that the money supply was tied primarily to the trade balance, exacerbating the effects of volatility in the demand for exports.²⁵

The WACB had many critics, who argued that the currency board system made the colonies vulnerable to fluctuations in the balance of payments, and that the investment of reserves abroad robbed colonies of much-needed development capital.²⁶ After independence, all four British colonies moved towards the establishment of new national currencies managed by central banks. In setting up these new institutions, governments were guided by divergent sets of priorities. First, they received conflicting advice from British authorities, on the one hand, and International Monetary Fund officials, on the other.²⁷ British authorities advised against the establishment of central banks in newly independent countries on the grounds that they had weak financial markets and that central banks would be subject to political interference. The IMF on the other hand argued that the greater policy space available to a central bank would allow newly independent governments to pursue more aggressive development policies and counter-cyclical policies to reduce instability.

Initially, at least, the introduction of national currencies and new central banks was less significant than the Bank of England perhaps assumed. Austen argues with regard to African institutions in general that “decolonization removed Europeans from formal control of these new structures, but did not dismantle the structures themselves”.²⁸ The capacity of monetary authorities in newly independent countries to act was limited not only by the type of institution, but by the fact that the former British West African colonies were small, open economies which remained financially dependent on Britain. Schenk argues that the establishment of new central banks served largely political purposes. In the terms under which they were issued, the new currencies differed little from the West African pound. They were issued at parity with sterling, backed by foreign reserves.²⁹

There was therefore a continuous period from 1913 to 1971, the year that marked the end of the Bretton Woods system of fixed exchange rates, in which West African currencies were issued at fixed exchange rates with sterling and therefore retained their value for the purchase of imported goods. It is this continuity across the transition to independence which makes the analysis presented in the next section—which measures seasonal fluctuations in the use of West African currencies—possible. It uses this data to offer an alternative perspective on the drivers of African demand for European and colonial currencies.

Seasonal Cycles and the Demand for Money in West Africa

This section will examine seasonal variation in the circulation of currency in West Africa and discuss its implications for understanding the demand for official currencies in West Africa. Seasonal cycles are frequently—and deliberately—ignored in recent research on monetary history. Most studies treat them as “statistical noise” and use econometric techniques to remove them from datasets prior to analysis.³⁰ In older literature, seasonal cycles featured more prominently. Jevons identified an increased demand for coin and notes during October in the middle of the nineteenth century.³¹ Kemmerer uses weekly data on a number of monetary variables to illustrate the seasonal cycles which affected the U.S. economy during the late nineteenth and early twentieth centuries.³² The aim of this literature was to identify the scale of seasonal variations in order to distinguish them from abnormal variations and panics.

While this line of research was overtaken by an interest in business cycles following the Great Depression of the 1930s, seasonal cycles have remained important features of historical financial systems. They can emerge for a variety of reasons, from the timing of annual harvests to winter construction slowdowns and the Christmas shopping season.³³ Historically, efforts to manage seasonal variations have been linked a range of institutional reforms, such as the creation of the Federal Reserve in the United States.³⁴ The 1980s and 1990s saw a revival of studies arguing that seasonal cycles are interesting in their own right.³⁵ In one study of the US, Germany, UK and Canada, Faig argues that seasonal cycles provide a “natural experiment” in assessing the demand for money using data. Seasonal purchases of consumer goods, in particular, were linked to the demand for cash.³⁶

Most work on seasonal variations in the demand for money has focused on the developed countries—where, as Moosa points out in a study of India, seasonal dynamics are likely to be different than in developing economies where agriculture is a more important sector.³⁷ This section draws on monthly data for currency circulation from the Bank of England archives as well as qualitative data from other archival sources to illustrate the importance of seasonal variations in West Africa. These data suggest that seasonal variations in the demand for currency in West Africa were strongly linked to the production and sale of cash crops. Before harvest season, traders exchanged foreign currency for local currency in order to pay producers, who then used the cash they had earned to purchase imported goods. Monthly circulation data therefore shows sharp seasonal increases reversed within a month or two following the harvest.

The quantitative data used in this article come from the first few years of independence in all four of the former British colonies in West Africa. The replacement of the West African pound with national currencies made it possible for the first time to obtain an accurate picture of currency circulation in individual countries. Prior to independence, the West African pound moved freely between countries. Such movements were not systematically recorded, rendering statistics on circulation within particular colonies almost meaningless. One report from 1934 by S.M. Jacob, the former government statistician in Nigeria, noted that “the amount of notes and coinage in circulation is not known with any precision,” and observed that reported figures were only possible if Nigeria had acquired coins and notes through inter-colonial trade (or if, has he put it, “counterfeiting is unsuspectedly large”).³⁸ Following the introduction of the Gambian pound in 1965, for example, Bank of England officials noted that currency issues in the Gambia were much lower than they had been during the colonial period, when trading companies often obtained currency in The Gambia and moved it along the coast to other territories.³⁹

Figure 1 shows seasonal variation in the four British West African countries. The same methods used for eliminating seasonal cycles from time series data can also be used to measure their scale. In this case, the data have been adjusted for seasonal fluctuations using 13–month moving averages.⁴⁰ Both the original and seasonally adjusted series for each country are given in Figure 1.

An indication of the scale of seasonal variations can be seen in Figures 2–4, which give the seasonal adjustment factors in an average year for each country. The seasonal adjustment factor is a measurement of approximately how far seasonal variations move currency circulation above or below the predicted (adjusted) trend. The influence of the end-of-year harvest season for both cocoa and groundnuts is apparent in the figures for Nigeria, Ghana and the Gambia. In all three peak circulation in most years was in December or January. Sierra Leone shows a different pattern, with peak circulation in April and May. Seasonal effects are particularly pronounced in The Gambia, where peak circulation is some 60 percent above the predicted trend.

While the quantitative data presented here is for the early post-independence period, there is substantial anecdotal evidence that there were considerable movements of money around harvest periods during the colonial period as well. Many examples of such seasonal fluctuations were instances when officials were caught short by the level of demand for currency during harvest seasons. In February 1916, for example, there was a shortage of currency in Nigeria after £200,000 had been sent to the Gold Coast to finance the cocoa crop.⁴¹ Seasonal demands for cash were sufficiently well known that they were described in the biography of Sir Alfred Lewis Jones, founder of the Bank of British West Africa. It noted that it was during the West African rainy season that the “bulk of the produce … arrives at the coast towns from the interior for sale and shipment to Europe. The Bank at such times is called upon to supply from its various coast branches very large sums of money to European mercantile houses in order to finance their purchases of produce.” It further observed that, in anticipation of this demand, the Bank “increases its normal reserves”.⁴² Nor were such fluctuations unique to West Africa. Kaminishi uses evidence from Manchuria in the 1920s to show that there was “an ongoing seasonal variation in the demand for money that was especially high in the winter period), corresponding to the harvest of soybeans, the migration of seasonal workers and the Chinese New Year.⁴³

• Download figure
• Open in new tab
• Download powerpoint

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Monthly circulation of currency in West African states: A. Nigeria, B. Ghana, C. Sierra Leone, D. The Gambia. Source: Calculated from monthly circulation data in BoE 8A218/1 using 13–month moving averages. Method for seasonal adjustment outlined in Feinstein and Thomas (2002: 28–30).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Seasonal adjustment factors for Nigeria and Ghana. Source: See Figure 1.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

Seasonal adjustment factors for Sierra Leone. Source: See Figure 1.

• Download figure
• Open in new tab
• Download powerpoint

Figure 4.

Seasonal adjustment factors for The Gambia. Source: See Figure 1.

Financial Development and Seasonal Cycles

What made the seasonal cycles in West Africa so severe? The dominance of agricultural exports provides one answer. This also explains the difference between Sierra Leone and the other countries discussed above. Sierra Leone’s mineral exports were less subject to seasonal variations than agricultural exports. But this may not explain all of the variance. Another important issue is the limited availability of financial services allowing people to undertake transactions without using coins or notes. Jonung argues that the spread of financial institutions and access to financial services reduced the velocity of money in Sweden before World War I.⁴⁴ However, the banking sector in British Africa notoriously offered limited services to Africans, which may have restricted opportunities to keep currency during the off-season.

The banking sector in British Africa during the colonial period was dominated by branches of British banks. The most prominent was the Bank of British West Africa, which became Standard Bank of West Africa following a merger with Standard Bank. The BBWA had started as an outgrowth of shipping firm Elder Dempster. It was initially granted monopoly rights over the import of British silver coin into West Africa. From 1912 it acted as the agent of the WACB, and of the four British colonial governments in West Africa.⁴⁵ During the 1920s, in particular, its reach extended beyond British Africa and into Liberia, where it also acted as banker to the government.⁴⁶

Owing to the importance of the BBWA in managing financial connections between West Africa and the world’s leading financial center in London, its papers—now housed at the London Metropolitan Archives as part of the Standard Chartered Bank collection—offer a unique perspective on the financial development of the region. This includes numbers of bank branches— which rose from 4 in 1896 to 110 by 1960—along with information their activities and their relationships with colonial governments.⁴⁷ The picture is revealing both in its scope and its limits. While there are detailed records of the BBWA’s interactions with government and merchant firms, the papers say little about the finances of African individuals. By the early 1950s, there was still limited African interaction with the formal banking sector. Figure 5 shows the division between African and non-African account holdings for the BBWA in Nigeria.

These figures were compiled at the request of the World Bank’s International Banking Mission in 1953. They are particularly striking considering that non-Africans accounted for a tiny fraction of the population of Nigeria at the time. Further, these numbers almost certainly overstate the level of engagement by individual Africans with the banking sector. In compiling these figures, the BBWA included the accounts of cooperatives and marketing boards under the ‘African’ category so that the figure given “might not be altogether unfavourable especially in number of accounts concerned”.⁴⁸ The size of these contributions can be inferred from the fact that deposits of all kinds from official or semi-official customers totaled £8,706,812, as compared with £8,338,458 for non-official customers.⁴⁹

The majority of the African population remained unbanked even in the late colonial and early post-independence periods, and cash continued to be the basis for most financial transactions. Schenk notes that “the amount of local currency that a colonial population will hold is related to the volume of cash transactions in the economy and the propensity to use banking facilities. In the West African colonies especially, the ‘banking habit’ was not widespread and most transactions were completed on cash terms. As a result, in most colonies currency in circulation made up a larger part of the money supply than bank deposits”.⁵⁰

• Download figure
• Open in new tab
• Download powerpoint

Figure 5.

BBWA deposits held by Africans and non-Africans in Nigeria, 1953. Source: Chief Accountant to District Manager, Lagos, 1 December 1953, in BBWA papers MS28760.

Limited access to banking facilities and the importance of cash for many transactions would help explain the rapid expansion in demand for money during the harvest. It does not, however, necessarily explain why the amount of currency in circulation declined again so rapidly after the harvest. This suggests that producers soon spent the proceeds of the harvest on some other asset which could serve as a store of value through the rest of the year. Reliance on annual data mean that seasonal variations—and how people coped with them—have not attracted much attention within African economic history.⁵¹

A literature on seasonal hunger and malnutrition does, however, provide some insights into how Africans leveraged a variety of currencies and goods to manage seasonal variation. Precisely which assets depended on a variety of local conditions. Hunter, in a study of Nangodi in North-East Ghana in the 1960s, concludes that cattle were used as ‘buffer stock’ to endure periods of hardship. During the hungry season, livestock are sold to provide cash crops for grain. According to Hunter, livestock are the farmer’s “mainstay in time of hunger,” to be sold “item by item as the hungry season progresses, commencing with the less valuable animals first”.⁵² This strategy was, of course, only available to those who possessed enough cattle to sell. A later study of the same venue by Jerome Destombes compared practice across the 1930s, 1960s and c. 2000.⁵³ Another study of farmers in Burkina Faso during years of famine in the early 1980s showed little evidence that cattle transactions were linked to periods of drought. Instead, the authors suggest that local credit markets and the use of goods other than livestock (including jewelry, cash holdings, cloth, and grain stocks) are used instead of livestock.⁵⁴ Outside West Africa, interviews of women in Zambia by Henrietta Moore and Megan Vaughan point to the accumulation of salt and soap to use in later exchanges.⁵⁵ Studies of other regions would no doubt yield further examples.

Some contemporary descriptions of the cash crop trade refer to the use of harvest proceeds to purchase specific types of goods. In 1919, for example, the Annual Report of the Gambia described the situation for the groundnut trade as follows:

The crop is lifted in November: the whole plant is pulled up from the ground and dried in stacks; the nuts are then beaten off with sticks. The planter keeps the nuts until the price offered by the dealers is high enough to suit his liking, unless he requires money urgently. The nuts are transported by donkey and head lots to the river towns where the trading stations are situated. After setting aside enough to pay taxes and for seed-nuts or rice bought from Government before the farming comments, the greater part of the money received is expended in buying cotton goods, etc., from the merchants—usually from those who have bought nuts.

The goods purchased after the harvest could be consumed or exchanged during the “hungry season” later in the year.

Some colonial records refer to the use of multiple currencies as reserves. The report by S.M. Jacob on taxation in Nigeria cited earlier noted “the lack of monetary reserves among the natives of Nigeria. Even in the Colony, where, if anywhere, one would have expected some accumulation of wealth, no reserves are said to exist.”⁵⁶ By contrast, the report noted, manillas were “held in readiness through the district”. Elsewhere it noted that “manillas in large quantities in the area between the Cross and Imo rivers, brass and iron rods in parts of the Ogoja, Onitsha, Benue and Kabba provinces, and a few foreign dollars along the north, constitute the main features of the average reserve”. Such reserves could also be spent in local markets. A report to the West African Currency Board by its officer in Lagos observed that “in the Owerri and Calabar Provinces large quantities of manillas are still in circulation. Brass rods continue to be used in the Ogoja and north of Calabar Provinces and in the Mamfe and Bamenda Divisions of the Cameroons Province. They are used by African middlemen for buying produce in the markets but owing to their bulk as compared with alloy and nickel their use is slowly dying out.”⁵⁷

While the use of brass rods might have slowly faded, strategies for managing seasonal fluctuations in money demands are not merely a historical problem. In his anthropological study of Zimbabwean migrant workers on South African farms, Bolt shows that they also mediate between cash and what might be termed trade goods in order to manage their resources between pay periods in a context where bank accounts are inaccessible. Paydays on large farms attract informal markets, shaping a regional commercial economy—“for many workers, payday is the time not only to pay off debts to businesspeople in the compound, but also to splash out on new items of clothing, food, and non-perishable groceries like soap. Some purchases are for use in the compound, others for remittance”.⁵⁸ A woman interviewed by Destombes in February 2000 said that she traded garden vegetables and other goods during the “hungry” season—and claimed that she had “achieved a degree of self-reliance beyond the reach of her grandmother, in whose time markets were poorly supplied and the purchasing power of people was extremely low”.⁵⁹

Conclusion

This article has used new data on the monthly circulation of West African currencies in the years immediately following independence to offer a new perspective on debates about the ‘currency revolution’ prompted by the introduction of colonial currencies. While some have argued that this introduction represented a shock to African monetary systems, others have pointed to the continued use of indigenous currencies such as manilas or cowries to suggest that this change was more gradual and less comprehensive. Taking into account seasonal variation in demand for currencies indicates that both arguments are, at least to some degree, correct.

Using data on the amounts of currency in circulation in the four former British colonies in West Africa, the article shows that demand for money was highly seasonal in nature. By paying explicit attention to seasonal variation, it builds on an older tradition in the study of monetary history in other regions in which seasonal fluctuations are used as to understand the sources of demand for money. In West Africa, the amount of currency in circulation expanded rapidly during the harvest but then declined again afterwards.

The highly seasonal nature of demand for official currencies suggests that demand was driven by their liquidity value in international trade. Limited access to financial services for the majority of the population meant that most transactions were conducted in cash. However, the decline in the currency in circulation after the harvest suggests that producers tended to convert their cash earnings into other assets, from textiles and livestock to manillas or other intermediate currencies.

The harvest was not the only source of seasonal demand. The need to pay taxes was another source of demand for official currencies. In a later account of his time as a District Officer in Kano in Nigeria in the 1930s, Morley recalled that “when the tax came in there were bags and bags of it, each of £100 in brown West African currency board shillings, bound with red tape and with a label on the neck of the bag, certifying its contents and signed by whoever had counted them”.⁶⁰ More work remains to be done on the dynamics of seasonality in individual localities. However, a focus on seasonality may help explain the persistence of multiple currency systems in the face of colonial and post-independence efforts to impose a single monetary standard.