Beyond Miracle and Malaise. Social Capability in Côte d’Ivoire and Senegal during the Development Era 1930–1980

This paper investigates the outcome of the efforts to economically catch up during the so-called development era in French-speaking West Africa. An attempt is made to measure and discuss key elements of social capability over the period 1930–1980 in Côte d’Ivoire and Senegal following Moses Abramovitz’ interpretation of social capability. The paper distinguishes between four elements of social capability: degree of structural transformation, social and economic inclusion, the state’s autonomy, and its accountability. We find that there was significant but uneven progress in social capability in both countries during the development era. Despite their differences in economic performance, both countries confronted fundamental shared challenges. Most notably, our analysis highlights how persistent lack of broad-based access to economic opportunities played a significant role in disrupting sustained economic and social progress in the two countries. This gives an opportunity to reflect on similarities and differences between the development era and the recent African growth phase.


Introduction
This paper goes beyond aggregate growth performance to investigate the quality of social and economic development in Côte d'Ivoire and Senegal during the so-called development era . Strong growth performances in many parts of Sub-Saharan Africa over the last decades have led observers to paint a cautiously optimistic picture of the current state, and future, of the continent (McMillan and Harttgen 2014;Rodrik 2014;Af DB et al. 2016; Thorbecke and Ouyang 2016) and pose the question whether we are watching an economic miracle in the making (Rodrik 2014). The present growth episode, however, is not the first such period in the modern economic history of Africa. Looking back to the 1950-1960s, in the middle of the development era, we find not only a similar air of optimism in the discussion about prospects for African growth but also vocal critics (Dumont 1962;Amin 1971;Easterly and Levine 1997;Cooper 2002, pp. 85-86). Starting in the interwar period, the development era marks a time of conscious steps to modernize and develop; first within the realm of colonial reign and from the 1960s as independent nations. However, although per capita growth during the zenith of this period was almost as high as the current period, growth deteriorated in the second half of the 1970s and Sub-Saharan Africa continued to diverge Bbig time^from the rest of the world ( Fig. 1; Pritchett 1997).
The development era was a crucial period for Sub-Saharan Africa that was marked by the transition from extractive colonialism to independent nation building. While there is a large literature that analyses the causes of lagged development in Africa since independence, lack of data-including reliable measures of GDP, economic sectors, household income, poverty, and inequality-has obstructed scholars from analyzing the evolution and drivers of economic growth over the long run, stretching into the colonial period. This knowledge gap has made it difficult to explain the fabric of economic growth and why growth during the development era did not convert into sustained development or modern economic growth. By implication, possibilities to unveil the dynamic and multi-dimensional nature of the current growth process in historical light has been greatly circumscribed.
We attempt to capture part of the multi-dimensionality of the development process by adopting a revised version of Moses Abramovitz' social capability framework, which builds on the proposition that a country's potential for rapid growth is strong when Bit is technologically backward but socially advanced0  (Abramovitz 1986, p. 388). Drawing on the literature of long-term economic growth and recent work on the role of the state, inequality, and structural transformation in development, we distinguish between four elements of social capability: degree of structural transformation, social and economic inclusion, and the state's autonomy and its accountability. Based on a dataset compiled from a variety of colonial and other public sources, we quantify elements of social capability for two West African countries-Côte d'Ivoire and Senegal-over the long run. Our aim is to review how social capability in Côte d'Ivoire and Senegal developed during the development era and to increase our understanding of why Côte d'Ivoire and Senegal did not embark on sustained economic growth despite relatively favorable contexts. A main theoretical contribution of the paper is the development of measurable indicators aligned with the concept of social capabilities. Several questions underpin our approach. How can the development progress be conceptualized and understood beyond aggregate growth figures? What was the nature of the deeper social and economic transformations that occurred in Côte d'Ivoire and Senegal during the development era? What mechanisms held back the transition to more sustained economic growth? In what way can performance during the development era inform our understanding of current growth patterns? Since we are aiming at generating, on the one hand, a more full-fledged understanding of why the development era did not lead the way into a process of sustainable growth and at the same time, on the other hand, highlight the relative change and occasional progress in the pattern of the underlying capabilities in two major economies in French West Africa, our approach does not lend itself to conventional labeling of comparative research designs, such as the classical dichotomy of either Bmost similar^or Bmost different^research methodologies. Our research method can rather be likened to comparative country analytical narratives through the application of an elaborate theoretical framework to two country cases, with the aim of using thick descriptions to increase our understanding of the growth process (Rodrik 2003). By using two cases, we can relate developments between the two countries and tease out more clearly common and unique features. To our knowledge, this is the first attempt to empirically assess the evolution of social capability in West Africa spanning both the colonial and independent periods. 1 Our choice of countries is primarily motivated by the fact that both Côte d'Ivoire and Senegal were relatively prosperous during the late colonial period-Senegal was the center of French colonial administration and industry in French West Africa and Côte d'Ivoire was endowed with high unexploited potential for cash crop production-but failed to achieve sustained economic growth and converge with the rest of the world after independence (Fig. 1). To understand this apparent contradiction more fully, we need to take a historical perspective, since economic, social, and political structures established during the colonial period have been so influential on the countries' development trajectories after independence. Another important motivation for focusing on Côte d'Ivoire and Senegal is the continuity and commonalities they display in terms of macro-institutional framework (Senegal was historically more developed and diversified, while the real expansion in Côte d'Ivoire started in the 1940s, see Fig. 2). The two countries were leading members of the federation of colonial French West Africa and after independence kept strong political, economic and monetary links to France (Hayter 1966;Amin 1971;Milburn 2004).
We argue that our approach to social capability better captures the dynamic and cumulative process of social and economic development to address these questions compared to more narrow or cross-sectional frameworks. Indeed, we find that there was considerable, but uneven, progress in social capability in both countries during the development era, despite the significant differences in growth trajectories. This rapid shift from extractive colonies to independent nation states within 50 years has not been properly recognized in the current development literature, which is often focused on contemporary social and economic challenges or cover shorter time periods. However, our analysis highlights how persistent lack of broad-based access to economic opportunities played a significant role in disrupting economic and social development in the two countries. We argue that the relationship between social capability and growth is complex and non-linear and that the social capability approach is useful in bringing out underlying growth constraints. These constraints are captured in the analysis of each of our four elements: transformation, inclusion, autonomy, and accountability. To varying degrees, both countries made some progress on all accounts but fell short of developing capabilities to put the countries on a path towards modern economic growth. In particular, our analysis highlights how persistent lack of broad-based access to economic opportunities played a significant role in disrupting economic and social development in the two countries. Our findings also give an opportunity to reflect on similarities and differences between the development era and the current African growth phase.  1925 1927 1929 1931 1933 1935 1937 1939 1941 1943 1945 1947 1949 1951 1953 1955 1957 1959 1961 1963 1965 1967 1969 1971 1973 1975 1977 1979 Côte d'Ivoire Senegal

Understanding and Measuring Social Capability in a Data Scarce Context
Scholarly attention to social capability grew out of efforts to explain why poorer countries so infrequently have been able to catch up with richer ones. The aim of the social capability approach was to develop and extend the so-called theory of convergence, which is based on the premise that a country's productivity growth rate varies inversely with its initial productivity level (Barro and Sala-i-Martin 1992). This is itself a corollary of the standard neo-classical Solow model that predicts inter-country convergence of income per capita over the long term (to the Bsteady state^) based on the assumption of diminishing returns to capital. Given the diverging income gap between rich and poor countries in the world economy since World War II, however, this prediction has received scant empirical support. Pursuing alternative lines of thinking, the concept of Bsocial capability^was conceived in studies of the exceptional Japanese growth miracle as an explanation for its ability to make efficient use of imported advanced technology (Ohkawa and Rosovsky 1973). Hence, the social capability approach should be seen as a critique of the conventional and excessively simple theory of the potential for growth, which according to Abramovitz, rests on the assumption that backward economies quite easily can make large technological leaps, more rapid accumulation of capital and effortlessly reallocate redundant rural labor to higher value added occupations. A major complication with the social capability concept is that it has not been defined with any rigor, which reduces its applicability and constitutes a plausible reason why it has been largely ignored for the past few decades. Abramovitz (1986) identified a country's social capability with technical competence, measured by years of education and its political, commercial, industrial, and financial institutions, but did not specify more precisely the relevant variables or the mechanisms involved. In later work, Abramovitz (1995, p. 29) more explicitly made the case for the importance of the deep-seated causes of growth, which provide the fundamental institutional and attitudinal set up of society that governs opportunities and incentives. Other authors emphasized the various components of social capability, including human capital, political system, distribution of income and wealth, political system, government policy, and financial systems (Perkins and Koo 1995). More recently, social capability has increasingly become equated with institutions and/or social capital (see, e.g., Temple and Johnson 1998;Snowdon and Vane 2005, p. 634;Putterman 2013). Another complication in assessing the importance of social capability in the development process is its cumulative evolution: BSocial capability develops in a process of interaction between the development a country can achieve, given its state of technology, capital and social capability, and the further effect of that development on its social capability, which is one of the conditions for further development.^ (Abramovitz 1995, pp. 39-40).
In an elaboration of these ideas, we propose to go beyond mere years of schooling, which is an excessively reductionist indicator, to measure social capability by assessing social and economic structures of a society that are central for entering and sustaining a process of modern economic growth. Given the relative elusiveness in making the concept of social capability empirically operational, there is a need for a more specific and tangible selection of indicators. We propose to do this through four elements of social capability: transformation, inclusion, autonomy, and accountability. We take as our point of departure the Kuznets-inspired discussion of Abramovitz (1995), which asserts that social capability basically consists of two classes of elements relating to (1) Bpeople's basic social attitudes and political institutions^and (2) Bthe ability to exploit modern technology^. The first class of elements (1), contains, according to Abramovitz, social arrangements for effective incentives. In this instance Abramovitz pays tribute to Kuznets (1966) and the fundamental role of long term growth in egalitarianism, which Abramovitz understands as Bthe social outlook that opens the way to talent and sanctions rewards for its accomplishments^(p. 32). We regard this in a broad sense as Binclusion^providing both incentives and sense of belonging; for instance, the extent of individual access to productive resources and whether the market is open to entry. As regards effective political institutions, we believe this is reflected in and better captured by the autonomy and accountability of the state, given the difficulties in finding a strong empirical relationship between democracy and growth (Rodrik 2014). While we recognize the importance of formal institutions, we argue that they are mainly influential to the extent that there is capacity to enforce these institutions. The second class of elements (2), i.e., the ability to exploit modern technology, is at the social level effectively synonymous with transformation, which is an indicator of the ability of an economy to allocate productive resources to economic activities with higher value added (see Hausmann et al. 2013). Hence, we derive from Abramovitz' two classes of elements, four distinct but inter-related and measurable capabilities: transformation, inclusion, autonomy and accountability.
Our selection of elements shares the insight by Adelman and Morris (1965) that the socio-political setting conditions economic performance. While Adelman and Morris (1967), before the advent of the discussion of social capabilities, did a pioneering and comprehensive attempt to relate growth to sociopolitical structure through factor analysis (over 40 indicators for 74 countries), our choice of method is appropriate for deeper small-sample macro-level investigations, when factor analysis, requiring large-N sample, is impossible. The four core elements are primarily justified by prevailing theoretical discussions rather than based on empirical correlates as in factor analysis. As such, we follow more closely the approach of Abramovitz, whose discussion on social capabilities was not based or even linked to the works of Adelman and Morris. We now proceed to defining the four elements of our framework.
Transformation relates to the degree of structural change in the economy, building on the idea that a transfer from low-to high-productivity activities is needed for sustainable economic growth (Kuznets 1966;Lin 2012;Rodrik 2014). Such as transfer does not seem to have occurred in Africa, in what Austin et al. (2017) call a pattern of Binterrupted industrial growth,^resulting from constraints imposed by factor endowments, global economic relations, and government policy. Indeed, recent evidence shows that structural change weighed on labor productivity in Africa in the 1990s, but turned positive from around 2000 . Data on sector shares are limited for the colonial period, but it is possible to give a historical account of the evolution of production and employment in industry and agriculture, in addition to related government policies.
Inclusion refers to the distribution of productive capabilities among the population and the access to economic opportunities, factors found to be strongly associated with economic growth in Africa by Adelman and Morris (1967). Theoretically, high inequality is potentially detrimental to growth in a number of ways; for instance, by not making productive use of human capacities, by constituting the source of growth inhibiting social conflict and policies, and by shortening growth spells (e.g., Alesina and Rodrik 1994;Persson and Tabellini 1994;Ostry et al. 2015). In the absence of good poverty and inequality data, we study economic inclusion through relative wages and incomes and social inclusion through the evolution of social structures and access to education. Education is generally considered a fundamental driver of economic growth through its effect on human capital accumulation (Schultz 1962;Easterlin 1981;Abramovitz 1986;Barro 1996). This is an area in which Africa did see important gains during the development era (Berg 1964;World Bank 1981;Cooper 2002; Prados de la Escosura 2013). Political inclusion is considered briefly under accountability.
Autonomy is the ability of the state to keep vested interests at bay, while at the same time being connected to society to ensure the relevance of goals and policies, resulting in what has been labeled Bembedded autonomy^ (Evans 1995). Embedded autonomy implies the simultaneous use of both reprimands and commendations in the interaction with influential actors and the possibilities for Brevenue bargaining^ (Bräutigam 2008). Autonomy can be measured by the ability of the state to tax the non-poor while maintaining support for universal social development to minimize tax evasion. This builds on the assumption that state capacity is a function of the ability to extract taxes and that such resources give the state freedom to implement policies (Fukuyama 2013;Besley and Persson 2013). The historical record shows that both colonial and subsequent independent governments had low capacity to implement policies, which posed a threat to their autonomy (Cooper 2002;Herbst 2014).
Accountability refers to the quality of governance and provision of public goods. The important role of the state in providing law and order, manage social conflicts, enforce contracts, and support markets has been widely emphasized in recent decades (World Bank 1997;Rodrik 1999;North et al. 2009;Lin 2012;Bardhan 2016). A fundamental aspect of accountability is legitimacy among local populations, which was low among extractive and ethnically diverse states conceived by the colonizing powers (Badie 1992;Young 1994;Englebert 2000). In the empirical section, we focus the discussion on how independent African government labored to ensure legitimacy among their populations. We also study their investments in education and health, since the latter is a common measure of the state's Bcollective^capacity (Besley and Persson 2014).
While linking up to and specifying Abramovitz original discussion on social capability, our approach attempts to capture the multi-dimensionality of the development process by including factors that can be considered fundamental for sustained economic development based on the literature on long-term economic growth (Kuznets 1966;Maddison 1988;Adelman 2000), and the recent literature on the role of the state, inequality, and structural transformation in development (Soifer 2008;Rodrik 2008;North et al. 2009;Fukuyama 2013;Besley and Persson 2014;Rodrik 2014). We aim to capture distinct and central elements of social capability that are at least in principle measurable. The assumption is that without progress on these fundamental variables growth driven by mere capital formation, international trade, or aid risks being one-off and unsustainable (Adelman 1983;Abramovitz 1995). Our approach is based on the recognition that economic and non-economic variables at macro level tend to move together or Bcluster^ (Adelman and Morris 1967;Besley and Persson 2014). Inherent in the capability approach is the two-way direction of causality between capabilities and sustained growth. Development is a very complex process, what we need are concepts to deal with the Bmess^ (Mann 2012, p. 4).
Our rationale for the use of the social capability concept is that it with greater success than other development theories can address why long-term growth in developing countries is or is not sustained by unpacking the process of growth and capture the potential for catching up. As such, the approach may in principle and with further advancement serve as Btroubleshooting^for countries not able to sustain progress in economic growth. The approach makes possible the comprehension of economic and socio-political development underlying growth estimates, particularly in settings where GDP accounts are unreliable or missing. This allows for an assessment of the catching up potential of poor countries by grasping changes in the elements over time, allowing for a deeper understanding of growth dynamics rather than establishing correlates of growth.

The Development of Social Capabilities in Côte d'Ivoire and Senegal
We now move on to the empirical part of the paper, which consists of an application of our four proposed elements of social capability on the Côte d'Ivoire and Senegal during the development era.

Transformation
The historical evidence indicates that notable structural transformation did take place in both Côte d'Ivoire and Senegal during the development era, although it is difficult to say how much for lack of data and the significant methodological challenges involved in assessing the size of the subsistence and informal sectors. Before the Second World War, the Federation of French West Africa was basically a large trading colony controlled by French trading firms (Boone 1992;Suret-Canale 1977). By the end of the 1970s, the industry had grown to represent 17% of value added in both countries (Table 1). In parallel, Côte d'Ivoire experienced a remarkable export boom in agricultural commodities and timber, while there was more modest growth and later even stagnation of peanut production in Senegal (Fig. 3). As a result, Côte d'Ivoire remained highly dependent on three successful exports-coffee, cocoa beans, and wood-that made up almost 70% of the country's exports at the end of the era. Peanuts remained important in Senegal, but represented only around a third of exports, and had been joined by exports of petroleum products, phosphate, and fish and shellfish, including fish conserves. Senegal's exports thus had become more diversified than those of Côte d'Ivoire despite the more dynamic economic performance of the latter.
However, the evolution of formal employment shows that the structural transformation was short-lived and limited in both countries. While the share of wage employment grew over the full period, it fell or stagnated after independence in both countries (Fig. 4). Additionally, the share of private employment in Senegal was very low, showing the tininess of the private sector. In the end, the transformation of the development era proved insufficient in both countries to generate sustained economic growth and create jobs for growing populations.
Both countries were thus unable to break fundamentally with a colonial legacy of dependence on agriculture and inadequate industrialization. French colonial authorities encouraged the expansion of cash crops, but mainly through investments in infrastructure, and discouraged the establishment of local production facilities. Local factor endowments determined the type and expansion patterns of cash crops. In Senegal, peanut production was suitable because of the semi-arid climate and light soils of western Senegal and the way its cultivation could complement traditional family food crop production and be based on existing production techniques (Pélissier 1966, pp. 33-38).
After the Second World War, there was a shift in colonial economic policy as colonial authorities began to encourage investment in manufacturing, but the outcome was limited by high cost of production, domination of French trading firms, and low local purchasing power (Boone 1992, p. 65). Dakar, which was the manufacturing Cote d'Ivoire Senegal 1 9 5 0 1 9 6 0 1 9 7 0 1 9 7 9 1 9 6 0 1 9 7 0 1 9 7 9   1930 1937 1946-50 1951-54 1956-60 1961-65 1966-70 1971-75 1976-80 Cacao -CI (left axis) Coffee -CI (left axis) Peanuts -SEN (right axis) Fig. 3 Production of cacao, coffee, and peanuts ('000 t  (Boone 1992, p. 68). Even with this narrow industrial base, the Senegalese economy was considerably more diversified than Côte d'Ivoire as shown by the earliest structural data produced by the colonial authorities (Fig. 5). Independent governments in both countries thus inherited narrow economic bases dominated by agriculture, but managed that legacy in different ways. In Senegal, the government continued to focus on the peanut sector, despite the low potential of the sector to continue to be a growth engine. The state took complete control over the sector through marketing agents, credits and subsidies, top-down cooperative structures, and price controls to break the previous dominance of French trading firms. These policies developed into a system of patronage that favored public officials and selected large-scale farmers and agro-industry, while taxing smallholder farmers and consumers (Delgado and Jammeh 1991, pp. 6-8). Still, there was less blatant extraction from peanut farmers than in other African countries, since the government relied on political support from the peanut-producing Muslim leaders, the Mourides ( 1986, p. 214). There were few public efforts to diversify agricultural production and support local food production, why food imports came to represent a quarter of total imports in the 1970s (World Bank 1979a, 1979b. In the 1960s and 1970s, peanut production volumes stagnated and declined due to misguided policies, the exhaustion of virgin land, stagnating productivity, climatic variability, and the end of French preferential prices in 1967-1968 (Vanhaeverbeke 1970;Suret-Canale 1977, p. 207;World Bank 1979a;Boone 1992) (Fig. 4). The situation generated rural discontent (malaise) and at the end of the 1970s, peanut farmers helped bringing down the state peanut monopoly-the ONCAD-that collapsed in 1980 with a debt corresponding to 15% of Senegal's GDP (Boone 1992, p. 204). Senegal's manufacturing sector was hard hit by the disappearance of Senegal's favored access to the French West African common market (de Benoist 1979). To counter these effects, the Senegalese government encouraged import-substitution industries through measures such as import protection, investment promotion, price controls and export promotion. The industrial sector remained vulnerable as it consisted mainly of resource-based export-oriented industries (peanut processing, phosphate mining, and fish processing) and import-substituting light industries (World Bank 1984, pp. 64-68). Manufacturing was mainly for the domestic market, while processed primary products exported mainly to Europe, especially France, continued to dominate exports. Production mostly involved processing imported materials, with protection giving little incentive to use local materials.
The independent Ivorian government also continued to focus on agricultural development, but with greater success than in Senegal. Uniquely for African countries, the Ivoirian government's agricultural policies were considered sound, based on reasonable and stable producer prices, low agricultural wages, efficient organizational structures, and immigration of foreign unskilled and skilled labor (World Bank 1978, p. 37f). Agriculture was taxed, but also received substantial support in return. There were serious efforts to diversify agricultural production into bananas, pineapple, palm oil, and rubber for export, and rice and cotton for local use, why Côte d'Ivoire remained largely self-sufficient in food in contrast to Senegal (Stryker 1972). At the same time, the agricultural sector suffered from some important weaknesses. Despite some improvements in productivity, the expansion of production had basically been based on cultivation of new land and most producers remained small (Amin 1967;Hecht 1983). The sector depended on migrant labor from regions north of Côte d'Ivoire, which contributed to immigrants amounting to around a million people or a quarter of the population in 1965 (Amin 1967, p. 43). Low agricultural wages fueled internal migration to urban areas (World Bank 1978, p. 40). Lastly, in an attempt to modernize the Northern part of the country, where cocoa and coffee could not be grown, the government spent significant resources on misguided investments in expensive and wasteful large sugar estates (Bamba 2016, ch 6).
The break-up of the federation and a dynamic agricultural sector created opportunities for the Ivorian industry to expand and replace imports from Senegal. Government policies to this end were described by external observers in positive terms as liberal, pragmatic, and gradualist (World Bank 1978). The value-added in manufacturingmainly the production of consumer goods for the growing local market and the processing of agricultural raw materials for export-more than doubled in real terms between 1960(World Bank 1967. The share of industry in the economy doubled between 1950 and 1970, even though agriculture was booming (Table 1). These results were lauded by contemporary economists. Elliot Berg (1971, p. 228) wrote at the time: BGradualism in the Ivory Coast has […] been associated with genuine economic transformation^. Structuralist analysts were more circumspect, noting persistent large foreign ownership (68% in 1975, World Bank 1978, emerging social tensions and dependence on vent-for-surplus expansion of cash crop production (Amin 1967). It was noted that Côte d'Ivoire's relatively liberal policies encouraged large-scale capital-intensive production, which had few linkages to the domestic economy and did not generate employment (World Bank 1978;Monson 1980;Alfthan 1982).
To conclude, there was a degree of structural transformation in both countries during the development era, most notably in Côte d'Ivoire because of the export boom, but the dependence on agriculture and foreign ownership largely persisted.

Inclusion
During the development era, there was a shift from highly exclusionary policies and practices to higher degrees of social and economic inclusion. French colonial rule had led to a leveling out of social status by a disempowerment of local aristocracies and the abolition-at least in principle-of slavery (Gellar 2005, p. 7). At the same time, new segmentations had been introduced based on race and access to French language, education, and public employment, which was only gradually Africanized. The local entrepreneurs who had thrived in both Côte d'Ivoire and Senegal in the early colonial period had been wiped out by colonial authorities around the turn of the nineteenth/ twentieth century and production and trade had fallen into foreign (mainly French and Lebanese) hands (Amin 1967;Amin 1971;O'Brien 1979).
After the Second World War, there were some corrections in labor market discrimination, as African demands forced the colonial authorities to adopt a string of labor market reforms that pushed up African wages, spread European benefits to Africans, and increased job security (Fall 2011). These reforms mainly benefitted the few Africans in formal employment and contributed to urban wages outgrowing rural incomes and widening the urbanrural divide so characteristic of developing countries (Fig. 6). After independence, these reforms grew to a halt, as unions were coopted by the independent governments. The state and banks continued to favor foreign companies and experts. The state also absorbed much of the surplus generated in the cash crop sectors, while local peanut, coffee, and cocoa cultivators mainly invested in small business and construction with limited value added or links to the rest of the economy (Amin 1967(Amin , 1969. Local Senegalese capital remained confined to commerce throughout the development era and private Senegalese entrepreneurs owned merely 3% of industry and the state around a quarter at the end of the 1970s (Fieldhouse 1986, p. 224;Boone 1992, p. 129). In contrast, in Côte d'Ivoire, the Ivoirization policies of the 1970s favored an increase of Ivorian participation in industry, with the state owning more than 50% and Ivorian private capital around 12% by the late 1970s (Secher Marcussen and Torp 1982, p. 101;Rapley 1993, p. 80).
The differences in economic structures between the two countries influenced the distribution of revenue and income. Being the center of former French West Africa, it is unsurprising that Senegal entered independence as a more unequal country than Côte d'Ivoire, due to a significantly higher urban-rural gap in average incomes (Morrisson 1968, p. 155), but by 1970, the two countries appear to have been at similar levels of inequality (Table 2). After independence, the urban-rural gap leveled out in both countries (an urban-rural income ratio of around 5 in Senegal compared to 4 in Côte d'Ivoire) (Lecaillon et al. 1984), because of the departure of well-paid foreign experts and migration from rural areas (Bonnefond and Couty 1988), while urban incomes continued to outgrow rural incomes in Côte d'Ivoire (Hecht 1983). The income distribution within the rural sector was more skewed in Côte d'Ivoire, because of the profitable cash-crop production (the cash-crop-subsistence income ratio was estimated to 3.4 in Côte d'Ivoire, compared to 3 in Senegal) (Lecaillon et al. 1984). There were considerable spatial income disparities between cash crop producing and food producing areas in both countries (see, e.g., Amin 1967, p. 82).
A key means for independent governments to achieve wider social inclusion was by promoting primary education. Until the Second World War, the colonial educational systems remained severely underdeveloped in both colonies. Enrollment was low and  1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 Peanuts -SEN Cacao -CI Coffee -CI  most students were given only partial elementary education, often in village type of schools with poor instruction and high dropout rates. Secondary and vocational education was largely an urban phenomenon. After the war, the French authorities established ambitious plans for the expansion of education and the curriculum and educational structure were increasingly reformed to resemble the French educational system (Bolibaugh 1972). The result was an acceleration of enrollment, but from very low levels and far below the original plans (Fig. 7). Less than 25% of the age group received elementary education and a meager 2% received secondary education in both Côte d'Ivoire and Senegal by 1957. Only 69 students from Côte d'Ivoire and 172 from Senegal (including Europeans) were awarded the French baccalaureate that year (Bolibaugh 1972). There was no great rupture with the colonial educational system in either Côte d'Ivoire or Senegal at independence. Both governments invested heavily to expand education to promote social and economic development and meet popular demand. By the end of the development era, primary and secondary enrollment rates were considerably higher in Côte d'Ivoire than in Senegal irrespective of gender and educational level-gross enrollment rates in primary schooling as share of the age group was 58% for girls and 90% for boys in Côte d'Ivoire compared to 35%/52% in Senegal, and for secondary schooling. 12%/27% in Côte d'Ivoire and 8%/15% in Senegal. 2 In both countries, the cost to the public budget of these investments was high and there were concerns about the relevance and usefulness of the schooling (Bolibaugh 1972;World Bank 1978, p. 276ff;World Bank 1979c). More worryingly, the lack of job creation in the economy and the fact that many qualified jobs were occupied by foreigners and French technical expertise meant that the many school leavers eventually ended up in 0% 2% 4% 6% 8% 10% 12% 14% 16% 1929 1931 1933 1935 1937 1939 1941 1943 1945 1947 1949 1951 1953 1955 1957 1959 1961 1963 1965 1967 1969 1971 1973 1975 1977 1979 Côte d'Ivoire Senegal  1964, 1967, 1969; 1970-1980: World Development Indicators the public sector or unemployed (Fig. 4). There were thus limits to how much the educational expansion could compensate for the persistent economic and social exclusion of large swathes of the population.

Autonomy
Our discussion of the autonomy of the state focuses on its revenue generating capacity. The fiscal systems of the two Côte d'Ivoire and Senegal were established early on in the colonial period to finance the colonial administration as the French government adopted a policy of self-sufficiency for the colonies (Huillery 2014). Tax revenues were largely stagnant until the Second World War, but increased noticeably in both countries during the post-war boom (Andersson 2017) (Fig. 8). Around 1960, the tax burden was high by developing country standards-in Côte d'Ivoire tax revenue as a share of GDP reached around 20% compared to 15% in Senegal, despite the fact that tax collection in both countries was constrained by overly complicated tax structures, inefficient tax administration and tax breaks to foreign investors (Abdel-Rahman 1965;Boone 1992, pp. 113-119). After 1960, there was significant divergence between the countries in terms of tax revenue, as the commodity boom gave Côte d'Ivoire significantly higher fiscal autonomy than stagnating Senegal. In contrast, both countries remained highly dependent on trade taxes throughout the development era, making state finances subjugated to world market conditions for the main cash-crops. This hit Côte d'Ivoire particularly hard as the country went into crisis at the end of the 1970s and its tax revenues fell markedly, while those of Senegal remained more stable. Alternative non-fiscal sources of revenue in the form of direct investments, financial aid, and debt non-fiscal emerged after the Second World War as the colonial principle of financial self-sufficiency was abandoned and the French colonies started having access to public funds from France, mainly through the so-called FIDES program. These initial transfers made up a large share of public investment in the post-war period. After independence both Côte d'Ivoire and Senegal continued to receive substantial French support under various cooperation agreements. The main vehicle was the Fonds d'aide de cooperation (FAC), which financed technical assistance (there were, e.g., more than 1300 French technical assistance personnel in both countries in 1966), budgetary aid, and various investments (Hayter 1966, p. 155). Eventually, these external flows contributed to reducing state autonomy by creating dependency on external aid in both countries. With a booming economy, this became less of a concern for Côte d'Ivoire, while Senegal remained highly dependent on aid, never falling below 40% of capital formation and reaching almost 10% of GDP at the end of the development era (Fig. 9).
Autonomy was also negatively affected by the accumulation of debt that started already in the post-war period as some of the French support to the colonies was given as loans (Suret-Canale 1977). As the balance of payment problems in both countries worsened, Senegal became the first African country to participate in a structural adjustment program with IMF in 1979, while the government of Côte d'Ivoire agreed to its first program only in 1989. In sum, there were notable gains in autonomy as the domestic fiscal systems developed, but this was not sufficient to meet the great investment needs of two developing economies, ultimately revealing the dependency on France and other external financing in both countries.

Accountability
We end the empirical part by reviewing aspects of the legitimacy of the state and the provision of public goods as indicators of accountability. The colonial state that was West Africa was small, coercive, and wedded to metropolitan priorities, which gave it low popular legitimacy among local populations. The notable exception to the authoritarian rule was the so-called Quatre Communes in Senegal where the inhabitants for historical reasons were guaranteed citizen rights and could send members to parliament. After the Second World War, accountability gradually increased as extractive colonialism gradually gave way to greater political and associative participation (Cooper 1996). The first leaders of independent Côte d'Ivoire and Senegal-Félix Houphouët-Boigny and Léopold Senghor-both rode to prominence during this time. A key stepping-stone to independence was the so-called loi-cadre adopted in 1956 that transferred significant powers to the colonies and introduced universal suffrage. After independence, Presidents Houphouët-Boigny and Senghor were both confronted with the challenge of governing their ethnically diverse nations and building the legitimacy of the new states. In Senegal, President Senghor pursued a home-grown variant of African socialism and actively sought to develop ties between the central government and regional elites, such as the Mourides, purportedly building on a tradition of ethnic tolerance (Boone 1992;Gellar 2005). In Côte d'Ivoire, President Houphouët-Boigny refuted ideology and banked on maintaining support from the politico-economic elite and the broader population by delivering economic prosperity under a formula labeled Bpragmatic capitalism ( Zolberg 1971, p. 12f;Zartman AND Delgado 1984;Waldner et al. 2017). Despite these differences, there were some significant similarities between the two regimes. They were one-party states that managed to maintain high degrees of stability throughout the development era thanks to charismatic leaders, the electoral support that emerged at the end of the colonial period, successful coalition building, and the French support embraced by both countries (Collier 1982, p. 152f;Zartman and Delgado 1984;Hesseling 1985, p. 364).
The establishment of these authoritarian and patrimonial systems undermined the development of alternative mechanisms for accountability that could be used by citizens to demand public services (Cooper 2002, pp. 156-160). In Senegal, a significant share of public spending went to salaries and benefits as a legacy of the colonial period (Fig. 10). With the colonial federal bureaucracy based in Dakar, much of local tax revenue was used to bear the cost of French staff that was considerably more expensive than local personnel (Huillery 2014). Senegal thus gained independence with a massive public staff bill amounting to almost 60% of total government expenditures (Morrisson 2006). While these levels fell somewhat over time, they remained high. In contrast, the share of public expenditure paying for staff remained lower in Côte d'Ivoire from the 1950s onwards. Uniquely in Sub-Saharan Africa, as the development era ended, Senegal went through a partial democratization process, largely unmatched by Côte d'Ivoire.
One way to increase legitimacy and accountability is to increase social spending. As already indicated, this is what happened during the colonial period, particularly for education (Table 3). The independent governments of the two countries invested heavily in education to respond to popular demand. By 1975, education absorbed a fifth of public expenditures in both countries, which forced both governments to curb the expansion of schooling to reign in expenditures. The focus on education stands in stark contrast to the negligence of health care, for which the shares of expenditure fell after independence to approach pre-war levels. The World Bank summarized the state of Senegal's national health care system in a 1979 report in the following way: BThe shortage of public funds can be directly related to sluggish economic growth. As a result, health care remains insufficient, in particular outside Dakar where about 80 percent of the population has no access to health services.^(p. 22). Clearly, education was a higher priority when independent governments were building their nation-states, as opposed to colonial governments that early on invested in health care to support the economic exploitation of the colonies and stop the spread of epidemics (Diop 1997).  1930 1936 1947 1954 1961-65 1966-70 1971-75 1976-80 Côte d'Ivoire Senegal Fig. 10 Share of salaries and benefits in total public spending 1930-1980. Source: 1930-19541961-1965and 1976-1980: BCEAO public finance statistics. Note: Federal budget expenditures not included in 1930-1954 Table 3 Social spending as share of total government expenditure Education (%) Health (%) 1930 1936 1947 1954 1965 1975 1930 1936 1947 1954 1964 1975 Federal budget expenditures not included in  Share of current expenditure

Conclusions
By using recently compiled data to study two countries of geographical and institutional proximity, we have shown the diversity of outcomes that were produced in Côte d'Ivoire and Senegal during the period 1930-1980. The social capability approach allows us to go beyond one-sided growth patterns to study underlying transformations to, in our case, go beyond simple representations of miracle and malaise. We show that there was considerable progress in social capability in both countries during the development era. This rapid and broad-based transformation from extractive colonies to independent nation states within 50 years has not been properly recognized in the literature, which is often focused on contemporary social and economic challenges or cover shorter time periods. The high growth rates of Côte d'Ivoire allowed the country to rapidly expand social capabilities in all areas from a low level, but Senegal could also achieve significant gains despite its much less dynamic economy. This shows that development outcomes cannot be attributed to mono-causal or fundamental factors, such as factor endowment or social structures. More complex relationships are at play.
We argue that the social capability approach gives us a framework for disentangling the processes involved in the spirit of Abramovitz (1986Abramovitz ( , 1995. Most notably, our analysis highlights how persistent lack of broad-based access to economic opportunities played a significant role in disrupting sustained economic and social progress in the two countries. This lack of inclusion also gives us an indication for where to look to understand whether contemporary African growth experiences are different and more likely to be sustained. The process of transformation gathered some strength but was insufficient to create momentum for long-term change. Despite some productivity gains in agriculture, particularly in Côte d'Ivoire, the manufacturing sector remained weak in both countries. On autonomy and notwithstanding some progress in fiscal capacity in both countries, they both relied overwhelmingly on external sources of revenue, reflecting relatively weak development of taxable domestic economic dynamics. In terms of accountability in both countries, investment in schooling was relatively extensive, but the neglect of investment in other key public goods and a bias towards public spending on the administrative apparatuses, meant that people at large did not benefit from public investments.
Two main implications for today come out from this study. First, our results suggest that even a strongly growing agricultural sector as in the case of Côte d'Ivoire is not sufficient to drive long-term growth unless combined with social and structural transformation that create productive employment opportunities in other sectors. This calls for broader consideration of social capability and political economy to assess the potential for less successful countries in catching-up. Further research is needed to understand the long-term dynamics involved. Second, this study gives an opportunity to reflect on similarities and differences between the development era and the current African growth phase. Although it may be tempting to draw parallels between the two periods, the context today is quite different with more integrated world markets and a multi-polar world. Even so, just as the development of social capability did not suffice to generate sustained growth in the post-development era, a closer investigation into contemporary social capability may yield clues as to where Sub-Saharan Africa might be heading in the future. In this way, knowledge about the accumulation of social capability from the past should provide a better understanding of today.