Comparisons between Liberia and Japan during the 1950s appear frequently in histories of Liberia. They first originated in the Monthly Bulletin of Statistics published by the United Nations in May 1962, which noted that Liberia’s rate of economic growth during the previous decade was the second fastest in the world after that of Japan. The same fact was then repeated in a study of Liberia’s economy published in 1966 by a team of Northwestern University economists led by Robert Clower, quoted here. Other histories have also taken it up. US diplomat Charles Morrow Wilson, who spent part of his career working in Liberia, observed that “by 1954 and continuing through 1960, Liberia, with the single exception of Japan, led the entire world in the growth rate of what census statisticians like to call the real gross national product.”Footnote ¹ Journalist James Ciment writes in his 2013 book, “Liberia in the 1950s and early 1960s boasted the second-highest economic growth rate of any nation in the world, behind only Japan.”Footnote ² Beyond history books, it has a popular resonance: I first heard it in a conversation with P. Bloh Sayeh, then director of the Liberian national archives, during my 2017 visit to Monrovia.

From the vantage point of the twenty-first century, this seems a strange juxtaposition. Japan is one of the wealthiest countries in the world, and Liberia one of the poorest. According to the World Bank’s World Development Indicators for 2017, a child born in Japan can expect to live more than twenty years longer than a child born in Liberia, and while 100 percent of the Japanese population has access to electricity, less than a quarter of Liberians do. Similarly, almost every Japanese adult is literate, but fewer than half of Liberian adults are. Visitors to Tokyo arrive at one of the two major international airports, take a direct train that glides between skyscrapers into the city center before transferring onto the city’s expansive and efficient metro system. In contrast, the limited number of flights landing at Roberts Field near Monrovia in that year had a bumpy taxi over a runway first constructed during World War II and frequently shut down due to maintenance problems.Footnote ³ The only transport into the city is by car, over largely dirt roads overlooked by buildings half-destroyed by war and never fully reconstructed.

Japan and Liberia did not start in the same place when they began their rapid postwar ascents. Before wartime destruction delivered a sharp shock to the Japanese economy, it had been the first country in Asia to achieve modern economic growth after the institutional reforms of the Meiji Restoration in 1868.Footnote ⁴ In 1850, Japan’s GDP per capita was $901, while Liberia’s was $441. The unit is 1990 international dollars, which is frequently used to make comparisons in historical national accounting. Subsistence income on this metric is defined as $400, or a dollar a day for most of the population plus a small elite. Both countries were independent nations increasingly isolated by the growing influence of European empires in the regions around them, and both struggled to define their position alongside the growing economic and military might of Europe and the United States. Over the decades that followed, however, Japan was able to carve a path for Asian industrialization, build its own empire, and marshal enough resources to engage in a global war. In contrast, Liberia’s economy remained stagnant across much of the late nineteenth and early twentieth centuries. It was not until the 1930s that a rapid influx of foreign investment and growing demand for rubber and iron ore produced the growth that would prompt the comparisons between the two countries in the 1950s.

Recent quantitative research in African economic history has made it easier to compare African countries both to each other and to those in other regions over time.Footnote ⁵ This work has challenged a number of standard narratives about Africa and its role in the global economy over the long run. By various metrics, African countries were not necessarily substantially poorer in the nineteenth century than those of Asia and elsewhere. They also experienced periods of rapid growth, contrary to common narratives about African economic stagnation.Footnote ⁶ Unfortunately, this work has not been distributed evenly across African countries. Differences in the availability of statistics on population, government finances, production, and trade have led to significant inequalities in the volume of research addressing the period before 1950. British colonies have received the most attention, in large part thanks to a standard set of statistical reports – known as the Blue Books – which all British colonial governments submitted to London each year. French colonies follow close behind, allowing for comparisons between British and French colonialism. Portuguese and Belgian colonies occasionally feature, though much less frequently. By contrast, Liberia and Ethiopia have been almost entirely absent, leaving a significant gap in our understanding of how their forms of governance and interactions with the world economy shaped their economic performance over the long run.

This chapter helps fill this gap by providing the first annual reconstruction of Liberian economic performance since the middle of the nineteenth century. Since there is no equivalent of the British Blue Books for Liberia, this involves the compilation of data from various, mainly qualitative, sources. Reconstructing the story of Liberia’s economic history in this way is not unlike Anne Carson’s reconstruction of the life of the monster Geryon in her Autobiography of Red; it requires piecing together a tale from the scraps of information that have survived over the centuries. With such a fragmentary record there will certainly be grounds to quibble over details and interpretation. The aim of the chapter is to document the broad trends of Liberian development over the nineteenth and twentieth centuries and provide a foundation for the more detailed thematic chapters to follow. It provides a transparent accounting of how these statistics have been assembled to provide a basis for future improvements and in order to counter broad claims about the particular unreliability of African statistics.Footnote ⁷ This will, I hope, offer a blueprint which people researching other neglected countries – notably Ethiopia – might follow in order to build a more rounded and inclusive picture of Africa’s long-run economic development.

Reassembling the Fragments

Struggles to obtain data on Liberia are no new problem. In 1901, a report on Liberia’s trade and commerce submitted by the British consul to the Foreign Office stated: “Authoritative statistics being unavailable, all others which are obtained from business houses and steamship agencies are apt to be biased by nationality or business relations, and are consequently more or less unreliable. I prefer, therefore, not to give any.”Footnote ⁸ True to his word, the consul duly submitted a trade report that lacked any statistics on trade. The position improved in the reports that followed, but in 1910, the report noted that “financial considerations have necessitated the suspension of the work of the statistical department of the government, and practically no information in respect to the trade and commerce of the country has been procurable for the last two years.”Footnote ⁹

One of the first comprehensive economic histories of Liberia, initially written as a London School of Economics (LSE) PhD dissertation in 1936, had much the same complaint. The author was an African American named George Brown, who had grown up in Kentucky and previously studied at Howard University in Washington, DC. In 1934, Brown enrolled in a PhD program at the LSE, where he took courses in anthropology, economics, and history and was particularly influenced by economic anthropologist Bronislaw Malinowski.Footnote ¹⁰ In his dissertation, and in the published book based on it, Brown lamented that “as official sources and accredited public documents on Liberian history are far from numerous, the research problems of selectivity give way almost wholly to the more difficult task of critically assessing such documents as do exist.” In particular, he observed that “the absence of accurate and creditable records of national wealth, business surveys, investment values and production statistics, agricultural, industrial employment and population censuses is a serious handicap.” Also missing was information on “wages earned by indigenous persons working abroad, and philanthropic contributions from abroad.” Without official sources, these had to be calculated from the “private, scattered, fragmentary and, in some cases, simply estimated records.”Footnote ¹¹ The fact that Brown thanks the Liberian president for use of his yacht during the period of his PhD research suggests that these problems were not a matter of access to government resources but rather indicate that such records were never produced to begin with. Data remained scarce even in 1962 after Liberia had enjoyed the period of economic expansion discussed in the previous section and the government had grown significantly in size. Robert Clower, leading the Northwestern University project, noted during his time in Monrovia that “a major bottleneck to effective planning in Liberia is likely to be the absence of any systematic records of statistics collected by government departments and central compilation of governmental regulations and activities.”Footnote ¹²

The successive difficulties faced by Cromie, Brown, and the Northwestern team have been compounded in more recent decades. Both environmental conditions and looting during the civil wars have taken their toll, particularly on records from the nineteenth century.Footnote ¹³ A recent survey of surviving archival holdings yielded just two boxes of material from before 1900.Footnote ¹⁴ The vast bulk of what remains dates from the presidency of William Tubman (1944–71) and beyond. Additional materials from these early periods can be found in the Liberia collections at Indiana University Library, and earlier material about Liberia survives in the national archives of its major trading partners and international organizations such as the League of Nations. However, even with this archival triangulation, much of Liberia’s early economic history is the subject of only limited surviving documentation.

This lack of easily accessible data has made it difficult to compare Liberia with other countries in terms of the patterns of its development, particularly for the period before World War II. However, it remains possible to construct data series, which can be used for comparative work, by harvesting quantitative information from largely qualitative sources. For example, the annual statements of the president of Liberia to the legislature frequently contain data on government finances and, less frequently, trade, which can be compiled into a mostly continuous data set. These are supplemented or checked by figures given in diplomatic correspondence or the media. Sporadic trade reports provide additional data on exports. The rest of the chapter presents annual data on population, government finances, and trade. It concludes by using these data to construct measures of GDP per capita, which are compared with other countries in Africa and around the world.

Population

Population data are one of the most basic categories of data needed to track patterns of development over time. Unfortunately, there are few reliable direct measurements of the population of African countries before the second half of the twentieth century, and almost none at all before the nineteenth. Gareth Austin writes that “all aggregate figures for the population of pre-colonial sub-Saharan Africa, or its major sub-regions are ‘guesstimates’ based on backward projection from colonial census reports.”Footnote ¹⁵ More recently, Sarah Walters has noted that “despite considerable research, especially on the demographic impact of slavery and colonialism, knowledge of Africa’s past population trends remains sparse compared with other world regions.”Footnote ¹⁶ There are some isolated exceptions, such as John Thornton’s study of the Kingdom of the Kongo from the sixteenth to eighteenth centuries, which relies on baptismal registers kept by missionaries in the region.Footnote ¹⁷ Colonial governments made some effort to calculate the population of their territories. However, the limited capacity of colonial states meant colonial administrations had to make do with uncertain estimates by district officers for most regions. Resistance to colonial taxation and forced labor recruitment also made many Africans reluctant to be counted.Footnote ¹⁸ The first censuses judged to be reliable were not produced until after World War II, and, in many countries, not until after independence.

Even by this fairly dismal standard, however, data for Liberia are limited. Only the early population of migrants from abroad is relatively well documented. From the 1820s, censuses were taken of the migrant population, giving details of not only their numbers but also their ages and the number of deaths that year.Footnote ¹⁹ Unfortunately, these censuses did not continue after 1843, and there was no attempt to extend these counts to include the indigenous population. The next census was not for more than a century, in 1962. As a result, there was considerable uncertainty as to the population of Liberia for much of its history as a state. In 1912, the American consul wrote in response to a request for a Liberian census that “there has never been an official census taken of Liberia as a whole, neither has there ever been, so far as I am able to ascertain, any enumeration in the cities and towns of the country or in the political subdivisions which are counties. Statements with reference to the population of Liberia are merely estimates.”Footnote ²⁰ Writing in the 1930s, George Brown noted that “the total population of Liberia is unknown. Estimates vary widely and apparently in the interests of the occasion.”Footnote ²¹ In 1956, Philip Porter gave the list of estimates over time, reproduced in Table 1.1.

The most systematic estimate of the population prior to the 1962 census was a PhD dissertation by Porter, a geographer, who used aerial photographs taken earlier in the 1950s to build a count of “huts” around Liberia and then field research to estimate the number of occupants per “hut.” By this method, his estimate was 744,340, plus or minus 50,000. This was not far off some of the lower estimates of the 1940s but below the total given by the 1962 census of 1.02 million, which was in itself significantly lower than some earlier estimates. The lack of previous complete censuses made the calculation of population growth rates impossible. To remedy this, the Liberian government began a population growth survey of 70,000 people in 1969.Footnote ²²

Without reliable historical censuses, estimates of the population of Liberia and other African countries before the middle of the twentieth century are made primarily by assuming a reasonable growth rate and projecting backward from the first reliable count. Patrick Manning uses the growth rate of India to develop new estimates for African countries.Footnote ²³ The use of the Indian growth rate has been critiqued by Frankema and Jerven, who argue that India was far more densely populated than much of sub-Saharan Africa.Footnote ²⁴ They argue that the growth rates of land-abundant countries in Southeast Asia are more appropriate. Figure 1.1 shows estimates of the Liberian population using these two different growth rates. In the Manning estimates, the early population was higher but growth was less rapid. In contrast, the higher growth rate assumed by Frankema and Jerven results in a lower estimate for Liberia’s population in 1850.

Figure 1.1 Alternative historical population estimates for Liberia.

Source: Frankema and Jerven, “Writing history backwards and sideways,” Manning, “African population”

However, one problem for both these Liberia estimates is the starting point from which backward projections are made. They rely on population estimates by the UN for 1950 and 1960, which were made before Liberia’s first census, and were actually higher than the census total. For example, the 1960 UN estimate was 1,052,000, as compared with a total population in 1962 based on the census of 1,016,443. While the gap between these is relatively small (around 3.5 percent), using the 1962 figure for calculating the population by the same method used in Frankema and Jerven reduces the overall population throughout the period. In 1850, for example, the revised estimate of the population within the modern boundaries of Liberia was 295,768 instead of 330,073. This implies that using the Frankema and Jerven figures for Liberia would tend to bias per capita figures downward.

At the same time, it is also worth asking whether the growth rates assumed by Frankema and Jerven for the West African region as a whole are plausible for Liberia. Both Manning and Frankema and Jerven rely on what they refer to as “situational” revisions to growth rates, which attempt to take into account local events which may have affected fertility or mortality rates even if no direct evidence of the impact is available. Frankema and Jerven motivate their modification of Manning’s situational adjustments largely with reference to the particular history of commercial expansion and colonial conquest in the region. For example, they argue that the West African growth rate likely differed from that of East Africa because the “integration of West Africa into the Atlantic economy through the so-called cash-crop revolution had been much further advanced in 1900 than the integration of other parts of tropical Africa in either the Atlantic or Indian Ocean trade networks.”Footnote ²⁵ Furthermore, they argue for adjustment in Manning’s growth rates based on the fact that “colonial ‘pacification’ came earlier in West Africa,” which “stabilized political relations and reduced investment risks.” It was also “associated with growing domestic commercialization of rural and urban economies.” These developments, along with colonial medical interventions such as smallpox inoculation, “created more favorable conditions for populations in West Africa to flourish.”

Frankema and Jerven rely heavily on the experience of Ghana (colonial Gold Coast) in framing these estimates. However, there is considerable evidence to suggest that population growth rates in the Gold Coast may not fit the experience of Liberia well. As data presented later in this chapter show, the half-century from c. 1880 was, overall, a period of economic stagnation for Liberia, with frequent and often severe periods of shrinking. This had significant consequences for government finances, and government penury left little scope for the kinds of medical interventions which may have reduced mortality in Ghana. A report by the US Public Health Service, commissioned in response to an outbreak of yellow fever in Monrovia in 1929, noted, for example, that “Monrovia has no modern method of sewage disposal.” Furthermore, it

Colonial governments could hardly be described as generous in terms of the provision of urban sanitation networks, but anecdotal evidence suggests that the position in Accra was better than in Monrovia. In response to an outbreak of plague, which killed 127 in 1908, construction began on the first pipe-borne water system – the Weija Water Works, which transported water to Accra from the Densu River – in 1914.Footnote ²⁷ The availability of clean water led to declines in mortality from a variety of water-borne diseases, including dysentery, diarrhea, typhoid, and guinea worm. This is not to overstate the infrastructure of colonial Accra. There was, for example, no parallel provision of sewage disposal.Footnote ²⁸ Still, studies of cities elsewhere in the world suggest that the provision of clean water had a dramatic impact on mortality rates, particularly through the decline in mortality not only from typhoid but also from the long-term health effects of surviving typhoid.Footnote ²⁹

Without direct demographic data, it is difficult to know precisely how much these problems affected population growth, though there are suggestions from around 1960 that mortality remained high. In 1958, a report on public health by the United States Operations Mission in Liberia (USOM/L) stated that “the life expectancy and the infant mortality are not known in Liberia nor are there any other statistics that are worth talking about. It is, however, known that the infant death rate is quite high (estimated to be as high as 70%). It is also known that the maternal death rate is high, but nobody knows just how high.” Disease burdens were also significant, according to the report.

A memorandum by a member of the Northwestern team noted that “although malaria is the largest single disease in Liberia, malnutrition and dysentery (in addition to malaria) are the causes of the absence of a ‘population explosion’ in Liberia. Another cause is the high infant mortality rate due to the ignorance of midwives.”Footnote ³¹

Some conclusions can be drawn from data collected from the 1960s onwards. In 1960, for example, life expectancy was higher in Ghana than in Liberia (forty-five as opposed to thirty-five), while mortality rates were higher in Liberia. Turning to fertility, Ghana’s fertility rate (births per woman) was initially higher than Liberia’s but declined faster. In studies of demography, trends over time are often more informative than comparisons of levels at a single moment.Footnote ³² The comparison of crude birth rates shown in Figure 1.2, for example, highlights the different trends between the two countries. In Ghana, fertility was higher initially but reached its peak around 1970. In Liberia, by contrast, fertility continued to increase until it peaked in 1981. An increase in fertility followed by an “irreversible” decline is one of the key stages in what demographers refer to as the demographic transition. In the first stage, high fertility rates and high mortality rates mean there is little or no population growth over the long run. When mortality rates fall, population expands rapidly as fertility rates remain high and often increase. Owing to a variety of economic and social changes, fertility rates then begin to fall. In the final stage, low fertility rates and low mortality rates mean there is once again little or no population growth.Footnote ³³ Research in historical demography has provided empirical evidence of this transition in a wide range of countries, from industrializing countries in eighteenth-century Europe to developing countries in the twentieth century. Placing Ghana and Liberia within this wider context suggests that Ghana began the demographic transition earlier than Liberia, making it difficult to believe that they would have shared the same growth rates across the twentieth century.

Figure 1.2 Crude birth rates per 1,000, 1960–2015.

Source: World Development Indicators

Figure 1.3 provides a reestimate of Liberia’s population statistics, which reflect its later path in the fertility transition. The estimates are calculated by revising the starting point used in Frankema and Jerven’s estimates, and by shifting the growth rate trend forward by ten years. This has the effect of dating the beginning of more rapid population growth during the twentieth century to the 1930s rather than earlier, which for Liberia reflects the beginnings of the period of economic growth discussed in the introduction to this chapter. This method produces figures for the nineteenth century which are somewhere between the two series in Figure 1.1: in 1850, 379,000 as compared to 330,000 (Frankema and Jerven) and 453,000 (Manning). To be sure, these estimates remain uncertain and there is much work to be done in the field of African historical demography to revise the “guesstimates” produced by backward projection under whatever assumptions. The role of internal migration, for one, remains imperfectly understood.Footnote ³⁴ One aim of this book, however, is to broaden our understanding of long-run growth in Africa in order to sharpen the situational adjustments made for individual countries.

Figure 1.3 Recalculation of Liberia’s population statistics.

Sources: For figures up to 1962, see text. After 1962, see Liberia, Statistical Handbook

Public Revenue and Expenditure

Government finances are the next indicator considered in this chapter. Fiscal history has been an important strand in the quantitative economic history of Africa. What is taxed and how the proceeds are spent reflect wider political and economic characteristics of a country. Joseph Schumpeter argued that “the fiscal history of a people is above all an essential part of its general history. An enormous influence on the fate of nations emanates from the economic bleeding which the needs of the state necessitates, and from the use to which the results are put.”Footnote ³⁵ In global economic histories, the expansion of the public treasury is associated with the broader increase of government and legal capacity, itself related to economic development.Footnote ³⁶

In Africa, low population densities could make the collection of tax more difficult.Footnote ³⁷ However, some African states did develop elaborate systems of taxation and tribute, often linked to the taxation of trade rather than land or income. More centralized states emerged particularly in areas where ecological diversity presented opportunities for specialization.Footnote ³⁸ During the colonial period, the pressure to build local revenue sources came from the unwillingness of most metropolitan governments to sanction large financial transfers to their colonies.Footnote ³⁹ How they did this depended on a combination of local political and environmental conditions.Footnote ⁴⁰

The Liberian government faced the same challenges as other states, colonized and independent. The prevailing theme in the historiography of the Liberian fiscal state is its almost constant insecurity, from the declaration of independence through the first half of the twentieth century. Monday Akpan, writing about the Liberian economy of the nineteenth century, noted that

He supports this statement with fragmentary data on revenue and expenditures given in nominal terms for a series of benchmark years (see Table 1.2). This table hints at some of the main fiscal issues which dominate a fuller history. There is a steady but slow increase through the nineteenth century, but with setbacks in individual years, including deficits in 1878 and 1908. However, the table is fragmented, with large gaps between observations across crucial periods.

Figure 1.4 expands on this work by giving annual estimates of public revenue in constant prices from 1845 to 1980. For the nineteenth century, these estimates are based primarily on the annual statement of the Liberian president to the legislature.Footnote ⁴² Each year this statement would include reports on the statistics of Liberia’s finances and trade for the year. Unfortunately, the original reports from this period have mostly not been preserved. However, the presidential statements themselves often give totals for revenue and expenditure, as well as sometimes more disaggregated figures. These data are supplemented by occasional figures from diplomatic correspondence or other archival sources. Where no direct observation has been found, linear interpolation is used to fill the gaps. Gray bars indicate observations from historical records. Figure 1.5 does the same for public spending.

Figure 1.4 Real government revenue, 1845–1980

Constant 1950 US$

Figure 1.5 Real government spending, 1845–1980

Constant 1950 US$

The data are presented in constant US dollars. What currency to use for long-run figures in Liberia is not necessarily straightforward, as Chapter 5 documents in greater detail. After declaring independence, the Liberian government issued its own currency, the Liberian dollar, which was at least in theory issued at par with the US dollar. In practice, however, the Liberian dollar was a paper currency issued without the backing of gold reserves or other assets. Through the nineteenth century, it was vulnerable to depreciation relative to gold-backed currencies. This led to the gradual substitution of British sterling, in an early example of “dollarization.” In 1943, with assistance from the US military, Liberia changed from the British West African shilling to the US dollar. In the decades since, Liberia has continued to use the US dollar alongside the Liberian dollar. Presenting these figures in British sterling would be equally appropriate for some periods. The deflation of the series also presents difficulties. No systematic price data exist for Liberia across this period. Reflecting the change of currencies, price indices for British government spending are used up to 1943, after which US price series are spliced on. The two series are similar so in practice it makes little difference which is used.

The two figures show that the size of the government treasury expanded slowly in the first half-century following Liberia’s declaration of independence, then more rapidly thereafter. However, it also shows the volatility of both revenue and spending. During the early period there were brief periods of rapid growth, in the 1860s and 1870s, but these were followed by sharp declines. A more substantial increase came around the turn of the century and lasted until the beginning of World War I. This was also followed by a sharp contraction, as was an increase of revenue in the 1920s. These fluctuations were small compared to the growth of government revenue during the postwar period. As the next section will show, this increase in revenue was linked to an expansion in foreign trade and investment.

Bringing the revenue and expenditure series together shows that more comprehensive data do not change the story of fiscal instability told by Akpan and others. Figure 1.6 shows the budget balance, measured as the surplus or deficit as a share of total revenue. The volatility of both revenue and expenditure resulted in frequent and substantial deficits. Even during phases when revenue was increasing rapidly, such as during the 1920s or the 1950s, spending kept pace with and often exceeded revenue. The Liberian government was thus frequently in financial trouble, even as it expanded in terms of both its share of the overall economy and its capacity.

Figure 1.6 Liberia’s budget balance, 1845–1980

Why were the public finances so volatile? In part, this is explained by the sources of revenue. Liberia, like other West African economies in the nineteenth century, relied heavily on the taxation of trade.Footnote ⁴³ Tariffs charged on imported goods were the largest single source of revenue in Liberia until the introduction of an income tax in 1951, though efforts to diversify the sources of revenue – discussed in more detail in Chapter 7 – meant their significance declined over time. Still, since international trade was a key source of income with which Liberians paid direct taxes, these too tended to be more difficult to collect during periods of global economic crisis such as World War I or the Great Depression. The Liberian government had few means – peaceful or otherwise – by which to compel people to pay taxes, and its limited legitimacy with much of the population meant voluntary compliance was difficult to build. Foreign companies operating in Liberia under concession agreements were offered generous tax exemptions as an enticement to investing in Liberia, so their contribution to the treasury was often less than it might have been.

Looking beyond the tax system, the Liberian government had few other means with which to mitigate sudden declines in revenue. It had no central bank until the 1970s. In its early decades, it would often turn to printing additional notes to satisfy short-run financial obligations, but this resulted in the depreciation of the Liberian dollar documented in Chapter 5. Once it began to use a foreign currency, this was no longer an option. It also accumulated substantial debts, as outlined in Chapter 6, but the introduction of foreign financial controls from 1906 also made this more difficult. In any case, neither of these mechanisms was sufficient to cope with the scale of the deficits accrued, and often the only real option was to cut spending, with difficult political consequences.

International Trade

Much attention is paid in African economic history to patterns of foreign trade, the third indicator reconstructed here. Trade links between Africa and the rest of the world have existed since antiquity. However, the volume of trade between Africa and Europe accelerated rapidly in the nineteenth century.Footnote ⁴⁴ The abolition of the slave trade by most European countries in the early nineteenth century did not immediately end the trade. However, exports of palm oil, groundnuts, and gum arabic gradually displaced the shipment of slaves across the Atlantic, as African merchants and elites responded to growing global demand.Footnote ⁴⁵

Since trade was a key source of both government revenue and commercialization in Africa, the level of exports has been interpreted by economic historians as a broad indicator of economic expansion.Footnote ⁴⁶ While data on domestic agricultural production tend to be scarce, international trade is much better documented for most countries. For Liberia, however, no annual statistics of imports and exports survive before the early twentieth century. Instead, these need to be pieced together from sporadic reporting in qualitative sources, which for trade often leaves large gaps.

In the absence of regular annual statistics, exports can be estimated indirectly. The simplest method is through linear interpolation, as in the series used by Banks presented in Figure 1.7. Before 1908, the Banks’ estimates show a steady increase in exports from 1850. This seems improbable given the turbulence of Liberia’s economic history at the time. The period covered by the interpolation included a coup d’état in 1871, changing trade policies, and disputes with neighboring territories. It also included considerable volatility in the global economy, most notably the global depression of the 1870s and 1880s. Annual data reported for similar countries across the same period show significant annual variation rather than steady increases. Figure 1.8 gives nineteenth-century trade data for Liberia’s neighbor, Sierra Leone, as an example. A more accurate way of filling in missing values builds on the close relationship between public revenue and the value of trade. As noted earlier, trade taxes were the most important source of revenue, and thus the two tended to move together. Where annual data are available, the correlation for the nineteenth century is very high: 0.93 for Sierra Leone for the period before World War I, for example.

Figure 1.7 Interpolated estimates of Liberian trade, 1870–1914 (nominal values).

Source: A. S. Banks, “Cross-national time-series data archive,” http://www.databanksinternational.com

Figure 1.8 Sierra Leone exports, 1850–1913.

Source: Sierra Leone, Blue Books

Figure 1.9 gives estimates of the value of exports derived using this method. First, linear interpolation links the sporadic nineteenth-century estimates. Second, the ratio between exports and revenue is used to adjust the interpolation to take account of annual volatility in the data. These estimates provide the first overall picture of Liberian exports over this long period. There are two phases of rapid expansion: one in the middle of the nineteenth century and the second a century later.

Figure 1.9 Real exports, 1845–1980 (in thousands of 1950 US$).

Source: See text

During the first such boom, some Americo-Liberians built considerable fortunes as middlemen in the growing Atlantic trade. In between, there was a long period of stagnation punctuated by considerable volatility in the real value of exports. This period of stagnation began with the global depression of the 1870s, a period of economic retrenchment worldwide. The expansion of global trade resumed in the 1880s, and Africa was among the fastest-growing regions in terms of export production. However, these data suggest that Liberia was seemingly unable to participate in this growth as it had in the middle decades of the century. As argued briefly in the next section and in more detail in Chapter 6, this was due primarily to the lack of investment in infrastructure comparable to other parts of West Africa. By contrast, the resumption of growth in the value of exports from the 1930s was linked to the rapid influx of foreign investment and production of exports by foreign concessionaires. During this period, as discussed in Chapter 9, the Liberian government aggressively courted foreign companies through the granting of large and generous concession agreements.

Figure 1.10 Export value shares by decade, 1850–1980

Note: Data are scarce for the 1870s and 1880s, but qualitative records suggest that it is likely that palm products and timber made up most of the remaining export value.

The primary exports in the early phase of growth were palm products and coffee. Figure 1.10 gives the value shares of specific exports by decade. While palm oil and palm kernels were exported across West Africa and thus do not generate much comment, there was great enthusiasm for Liberian coffee during this period: samples were even sent to the Great Exhibition of 1851. Data on coffee show a substantial increase through the nineteenth century. In the twentieth century, coffee exports were uneven, falling in the years leading up to World War I, then rising again during the interwar period, and falling to zero during World War II, before reviving during the second phase of growth after 1945. By the middle of the twentieth century, coffee was a minor contributor to Liberia’s exports. During the 1930s, rubber became by far the most important export, comprising 82 percent of the total value of Liberia’s exports on the eve of World War II. The export of rubber grew rapidly in this period due to the establishment of the Firestone Rubber Plantation in 1927. Soon after World War II, however, rubber was overtaken by iron ore, which became the engine of the rapid growth that would prompt the comparisons between Liberia and Japan. Exports of iron ore grew from nothing in 1952 to 25 million tons twenty years later. Like coffee and rubber, however, the success of the Liberian iron ore industry was short-lived. By the early 1990s, the mines had ceased production entirely.

Real imports, shown in Figure 1.11, followed a similar pattern to exports. Data on the composition of imports is more fragmentary, but what data exists suggests that the nature of the goods imported shifted over time. During the nineteenth century, and particularly during the first decades after migrants arrived, imports were dominated by two classes of goods: those which could be traded with indigenous communities, particularly cotton textiles and tobacco, and goods largely consumed by the communities of migrants themselves. These included clothing, furniture and other household goods, and some foodstuffs including flour, salted fish, and meat.Footnote ⁴⁷

Figure 1.11 Real imports, 1845–1980 (in thousands of 1950 US$)

Some historians have interpreted the latter as an indication that emigrants scorned indigenous foods and subsisted on imports instead. William Allen, however, has shown that this was not the case, pointing to evidence from letters and other records suggesting widespread consumption of cassava and other African products.Footnote ⁴⁸ Rather, food imports reflected what was to become a persistent problem in Liberia’s economic history, namely the limited integration of local markets due to the high cost of transport from the interior to the coast. As a result, urban populations depended on imports of rice through most of the period, despite the fact that rice was widely produced locally. Rice imports increased rapidly with the expansion of rubber plantations and mining after World War II, as shown in Chapter 9.

Foreign trade was thus a key ingredient of Liberian economic development over the nineteenth and twentieth centuries. It was the main source of both economic growth and periods of shrinking, and on top of that underpinned government finances and the private incomes of the elite. The next section incorporates these figures with estimates of the shares of different sectors in the economy to build the first annual estimates of Liberian GDP per capita since 1847.

Liberian Historical National Accounts, 1845–2008

The three data series developed in this chapter can be used to construct the first estimates of Liberia’s GDP per capita built from original archival data from 1845 to 1980. These estimates allow Liberia’s economic performance since the middle of the nineteenth century to be compared to other countries, both in the region and farther afield. They show that Liberia’s pattern of economic development over this period was characterized by considerable volatility, with the gains achieved during periods of rapid growth erased by economic shocks, both internal and external. While Liberia shared this characteristic with other West African countries, the pattern of growth in Liberia differed from that of the rest of the region.

Official GDP per capita data produced by the Liberian government begins only in 1964. Prior to that, there were two efforts to calculate Liberia’s national income during the 1940s and 1950s. The first was an unpublished series compiled by a US government survey conducted in Liberia in 1959.Footnote ⁴⁹ The second was by the Northwestern University team for the 1950s.Footnote ⁵⁰ Subsequently, there has been only one attempt by an economic historian to estimate GDP per capita for Liberia before 1945. Leandro Prados de la Escosura’s pioneering figures for a set of benchmark years from 1870 are based on extrapolating the relationship between terms of trade data and per capita income from the post-1945 period backward, using the largely interpolated trade data from Banks, replicated in Figure 1.7.Footnote ⁵¹ The estimates presented here use the archival data presented in previous sections of this chapter to calculate GDP per capita from the output side, dividing the economy into three sectors: the traditional sector (which includes agriculture and domestic industry and services), government, and the export sector. The method used is based on Broadberry and Gardner’s estimates for eight other African countries, with some modifications due to data constraints.Footnote ⁵² These are outlined in more detail in Appendix 2.

In Liberia, as in other African economies, the domestic production of food and other goods for home consumption is the least well-documented of the three sectors discussed here, especially before the mid-twentieth century. In 1962, Northwestern economist Robert Clower described the agricultural sector as being divided among (1) foreign-owned plantations; (2) commercial farms; and (3) “tribal” farms producing for subsistence. Subsistence farmers made up the vast majority of those employed in agriculture, and of the labor force as a whole. He estimated, based on aerial surveys conducted during the 1950s, that there were some 150,000 “tribal farms” in production at that point. The plantation and commercial sectors had grown more recently and produced exports crops, primarily rubber.Footnote ⁵³ The 1971 agricultural census painted a similar picture. It reported 128,076 agricultural holdings, with a total farm population of 681,702 (out of a total population of 1,362,000). Of these, 110,839 holdings produced rice, with an average size of 2.1 acres, and 73,103 produced cassava with an average size of 0.9 acres. These were the two main staple crops. There were 9,744 farms producing rubber, with an average size of 43.1 acres.Footnote ⁵⁴

The growth of plantation and commercial agriculture had prompted what Clower described as a slow process of commercialization in the subsistence sector, with the production of some traditional cash crops like cocoa, coffee, and piassava.Footnote ⁵⁵ This had also been true in the first era of export expansion when indigenous farmers began to produce their own coffee and cocoa during the nineteenth century following the cultivation of both crops on settler farms near the coast.Footnote ⁵⁶ However, this process was slowed by two factors: the high cost of transport from the interior, discussed in greater detail in Chapter 6, and in the twentieth century an institutional structure that gave the indigenous population little incentive to invest in agricultural productivity. This is discussed in Chapter 7.

Owing to the lack of systematic data on production in the subsistence sector, the estimates presented here follow the method used in early historical national accounts and assume that this sector grows in line with population.Footnote ⁵⁷ In more recent work, the assumption that per capita consumption remains constant even as wider economies change has been challenged. To get a more accurate picture of how demand for food develops over time, newer estimates combine population data with estimates of income elasticity.Footnote ⁵⁸ In estimates for other African countries, this is done using real wages.Footnote ⁵⁹ Unfortunately, very little price and wage data survive for Liberia before 1945, and thus it is not possible to apply the same approach here. Output for the government sector is measured by real government spending. As shown in the section on government finances, real government spending per capita grew rapidly in the first decades after Liberia’s declaration of independence in 1847. Growth then slowed across the period from the middle of the 1870s until the early 1940s, when growth accelerated again. This follows broadly the pattern of real exports, used to measure output in the export sector. That these two should move in parallel is consistent with the importance of trade for government revenue.

To calculate GDP from these series requires applying the appropriate weights for the three sectors. Differences in definition mean the weights vary slightly between the US government, Northwestern University, and official Liberian government series, but all show a rapid growth of the export sector as a share of the overall economy. The time series of aggregate GDP is constructed from real output indices for the three main sectors, using the 1964 weights from Table 1.3.Footnote ⁶⁰ This is then divided by population to produce estimates of GDP per capita.

Table 1.4 shows that in 1845, Liberian GDP per capita was $430, just above bare-bones subsistence, which is usually taken to be $400. This represents most people living at $1 a day, which was the World Bank’s definition of poverty in 1990, plus an elite who were much richer, but were too few in number to have a significant impact on the average GDP per capita. It rose to around $500 in the two decades following the declaration of independence but then stagnated for a half-century before a period of rapid growth from the late 1930s. This growth was driven largely by expansion in the export sector, though the government sector also grew during this period. It continued until the 1970s, after which it stagnated then declined by 1980.

For reference, Table 1.4 also gives GDP per capita in Ghana and Japan, the two comparator countries mentioned so far in this book. Appendix 2 expands the range of comparisons. It shows that Liberia and Ghana had similar levels of GDP per capita in 1884, but Liberia fell behind through the next few decades until the interwar period. At its lowest point in the early 1930s, Liberia’s GDP per capita was less than 40 percent of Ghana’s. In the 1940s, Liberia’s rapid growth brought it back into parity with Ghana, and during the 1950s and 1960s, its GDP per capita was substantially higher than Ghana’s.

This divergence is striking and signals a wider difference between Liberia and other West African countries. During the late nineteenth and early twentieth centuries, Liberia stagnated even as Ghana (then the Gold Coast) and most other West African countries enjoyed rapid growth based on the production of export commodities – in Ghana’s case, cocoa. This growth was facilitated in no small part by the construction of railways. Construction of the Gold Coast railway began in 1898, and by the turn of the century it had reached the main cocoa-growing areas. For reasons discussed in Chapter 6, Liberia was unable to raise sufficient capital for railway construction and thus could not capitalize on export growth during the decades of globalization that preceded World War I.

Equally striking is Liberia’s sudden growth from the 1930s, which allowed it to not only catch up with Ghana but actually exceed it by a substantial margin. This can be attributed to two factors investigated in greater detail later in the book. First was the establishment of the Firestone Rubber Plantation. While the plantation did not actually begin production until the 1930s, it also laid the groundwork for the rapid influx of foreign concession companies producing rubber, iron ore and timber during the two decades after 1945. It was these concession companies, and to a lesser extent the commercial farms related to them, which were largely responsible for the growth in exports, which in turn fueled the expansion of the government sector. This growth, and the uneven distribution of its proceeds, is the subject of Chapter 9. The second factor was foreign aid, examined in Chapter 8. Beginning during World War II, Liberia was a key recipient of foreign aid from the United States which contributed to improvements in road and port facilities.

Though Japan’s per capita income was higher than that of the two African countries, comparison with Japan tells a similar story up to the 1930s, with Liberia falling steadily behind. The impact of the war on Japan’s economy along with the acceleration of Liberian growth led to a short period of convergence in which Liberia regained its relative position of the late nineteenth century. The two grew at roughly the same pace during the 1950s before Liberia fell decisively behind from 1960. Its level of development was substantially lower and, as the rest of the book will explore, it did not have the same level of human capital and state capacity, leaving it vulnerable to periods of shrinking. During the civil war of the 1990s, Liberian GDP per capita dipped below the level of the nineteenth century, and subsequent growth has failed to regain the highs of the 1970s.

Conclusions

By bringing together the scraps of data on Liberia’s economic history which have survived in a variety of qualitative sources, this chapter has provided a quantitative overview of Liberia’s economic history since the middle of the nineteenth century. While the figures here can certainly be improved – and hopefully will be in future research – they tell a story of economic turbulence which is consistent with much of the existing historiography. However, they also allow Liberia’s story to be compared with that of its neighbors in a systematic way for the first time.

From the data in Table 1.4 and the underlying series discussed earlier in the chapter, it is possible to separate Liberia’s economic history into four periods. The first, dating from the 1820s until the 1870s, was a period of growth and expansion, as it was across much of the West African coast. Rising demand for African crops and forest products like palm oil paved the way for a rapid growth in the export of such goods. Americo-Liberian elites competed with indigenous merchants for a place in that growing trade, and some achieved considerable success in the coastal shipping of palm oil and subsequently coffee. Beginning with the global depression of the 1870s, Liberia’s fortunes took a turn, both in absolute and relative terms. Like other such mercantile communities along the coast, Liberia’s merchants began to lose their position of comparative advantage as steamships and larger European firms began to dominate. The construction of railways in neighboring countries saw more and more Liberian goods be exported out over land borders, with no benefit to the Liberian treasury. This downward trajectory was not reversed until the 1930s when the Firestone Rubber Plantation came into production. From then, exports of rubber and iron ore fueled rapid growth which saw Liberia catch up to Ghana and other African countries in terms of its per capita GDP.

The proceeds of this growth were unevenly distributed and accrued mostly to an urban elite in and around Monrovia. Inequality between regions and groups was not new to Liberia, as subsequent chapters will show, but it was certainly exacerbated by the specific pattern of growth during the middle of the twentieth century. The strains this placed on the political system began to emerge in the middle of the century, and in 1980 the unified rule of the True Whig Party finally fractured, ultimately leading to the civil wars which cost hundreds of thousands of lives and erased the economic gains of the period after World War II.

Liberia’s story over the nineteenth and twentieth centuries is thus an extreme example of the pattern of growth observed in most countries before industrialization, characterized by phases of growth, which are then reversed by periods of shrinking.Footnote ⁶¹ The scale of the shrinking is due to a combination of limited economic diversification and institutional weaknesses. Liberia’s dependence on a small number of primary exports left it vulnerable to changing global demand for these products, and some of its biggest periods of shrinking came during global economic crises or the two World Wars. The most significant downturn, however, was during the civil wars of the 1990s, resulting from the failure of the Liberian political elite to redistribute economic gains more widely and allow a greater share of the population a voice in political decisions.

Subsequent chapters of the book explore how these two vulnerabilities emerged and developed over time. Part I of the book examines their foundations. Chapter 2 uses anthropological and archeological evidence to examine what we know about the economy of the “Grain Coast” before the arrival of migrants from the United States. Chapter 3 focuses on the migrants themselves, drawing together various collections of data to ask who came to Liberia and how the selection of migrants influenced its later history. Part II focuses on the Liberian government’s struggles to exercise its new economic sovereignty in a region increasingly dominated by European influence. The dilemmas it faced and the solutions it found helped shape the impact of the long period of stagnation on the shape of Liberia’s economy and society. In Part III, three chapters examine how the Liberian government under Americo-Liberian rule used strategic alienation of its sovereignty to help sustain the pace of economic growth and, in consequence, its own survival.