Revista de Historia Interdisciplinaria, XLVII:2 (Otoño, 2016), 193–212.
Johan Fourie
The Data Revolution in African Economic
History Improvements in computing power, increased connec-
actividad, and more advanced analytical techniques herald the era of
Big Data in fields as diverse as astronomy, economics, biology,
and management. Todavía, in the history profession, this data revolution
has gained only limited traction. One obvious reason is the lack of
Big Data from the distant past. As Eric Schmidt, the former CEO of
Google, remarked in 2010, from the dawn of civilization through
2003, five exabytes of information were created. Only seven years
más tarde, that much information was being created every two days. Como
the saying goes, sin embargo, not all of these data are useful, and not
everything that is useful is captured in these data. En efecto, historians
have to distinguish what is meaningful from what is not. Too much
information can be as problematical as too little.1
Johan Fourie is Associate Professor of Economics, Stellenbosch University. He is the author of
“The Remarkable Wealth of the Dutch Cape Colony: Measurements from Eighteenth-
Century Probate Inventories,” Economic History Review, LXVI (2013), 419–448; with Erik Green,
“The Missing People: Accounting for the Productivity of Indigenous Populations in Cape
Colonial History,” Journal of African History, LVI (2015), 195–215.
The author thanks Gareth Austin, Angus Dalrymple-Smith, Pim de Zwart, Felix Meier
zu Selhausen, Ewout Frankema, Leigh Gardner, Erik Green, Kris Inwood, Martine Mariotti,
Alex Moradi, Nonso Obikili, Auke Rijpma, Jan Luiten van Zanden, the members of the LEAP
(Laboratory for the Economics of Africa’s Past) team at Stellenbosch, and two anonymous
referees for valuable input and fruitful discussion during the writing process. He received
financial support from the South African National Research Foundation, the Elite Project
Fund of the Faculty of Economic and Management Sciences at Stellenbosch University,
the Oppenheimer Memorial Trust Sabbatical Study Fund, the HB & MJ Thom Travel Fund,
and Economic Research Southern Africa, which issued an earlier version of this paper as
ERSA Working Paper #555. Jan Luiten van Zanden was a most generous sabbatical-study
host in Utrecht.
© 2016 por el Instituto de Tecnología de Massachusetts y The Journal of Interdisciplinary
Historia, Cª, doi:10.1162/JINH_a_00977
1 Eric D. Feigelson and G. Jogesh Babu, “Big Data in Astronomy,” Significance, IX (2012),
22–25; Liran Einav and Jonathan Levin, “The Data Revolution and Economic Analysis,"
NBER Working Paper No. 19035 (2013); Vivien Marx, “Biology: The Big Challenges of
Big Data," Naturaleza, CDXCVIII (2013), 255–260; ANUNCIO. Howe et al., “Big Data: The Future of
Biocuration,” ibid., CDLV (2008), 47–50; Andrew McAfee and Erik Brynjolfsson. “Big Data:
The Management Revolution,” Harvard Business Review, XC (2012), 60–66. Schmidt’s figures
have been disputed (http://readwrite.com/2011/02/07/are-we-really-creating-as-much), pero
the point is that the quantity of data is increasing exponentially.
yo
D
oh
w
norte
oh
a
d
mi
d
F
r
oh
metro
h
t
t
pag
:
/
/
d
i
r
mi
C
t
.
metro
i
t
.
mi
d
tu
/
j
i
/
norte
h
a
r
t
i
C
mi
–
pag
d
yo
F
/
/
/
/
4
7
2
1
9
3
1
7
0
6
5
9
7
/
j
i
norte
h
_
a
_
0
0
9
7
7
pag
d
.
F
b
y
gramo
tu
mi
s
t
t
oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3
194
| J O H A N F O U R IE
The purpose of this Research Note is to show the extent to
which the data revolution is consequential for the field of history.
The surge in computing power and access to data-processing software
and online resources have enabled historians during the past two de-
cades to capture historical statistics on a much larger scale than before.
The statistical records stored in the archives of the imperial powers, como
well as in those of the former colonies, are now available for analysis
on an unprecedented scale. The data revolution is especially valuable
when applied to regions where written records are scarce, como
sub-Saharan Africa. This Research Note documents how a new
generation of economists, geographers, and historians is rewriting
African history using archival records (colonial sources written for
purposes unrelated to current research questions) in conjunction
with geographical, climatic, and demographic studies of the distant
past to gather surprising new information.2
A NEW AFRICAN ECONOMIC HISTORY African history has always suf-
fered from a limited number of written records. Not until the
1960s, when most African countries had gained independence, did
historians in Africa and elsewhere begin to borrow methods from ar-
cheology, linguistics, and oral history to investigate the continent’s
rich and varied past. Information about Africa’s economic history at
this point was in high demand. Africa’s rapid economic growth rates
during the late colonial and early postcolonial periods and the eupho-
ria immediately following independence raised new questions about
the effects of imperialism and colonialism and the possibilities of future
prosperity. Economic-history departments were established in univer-
sities across the continent. Debates fostered within these newly estab-
lished departments could be fierce, informed by the ideological battles
between Marxist and liberal economics. Evidence in support of the
various positions came most often from qualitative sources that orig-
inated during the colonial era, although the abundant statistical records
were not completely neglected. Many pioneers of African data col-
lection, such as Johnson, Szeresewski, Hopkins, Manning, Eltis, y
Austen, began to assemble their data sets during the 1960s and 1970s.3
Patrick Manning, Big Data in History (Basingstoke, 2013).
2
3 Marion Johnson, “The Cowrie Currencies of West Africa Part I,” Journal of African History,
XI (1970), 17–49; ídem, “Cloth as Money: The Cloth Strip Currencies of Africa,” Textile History,
XI (1980), 193–202; Robert Szereszewski, “The Process of Growth in Ghana, 1891–1911,”
yo
D
oh
w
norte
oh
a
d
mi
d
F
r
oh
metro
h
t
t
pag
:
/
/
d
i
r
mi
C
t
.
metro
i
t
.
mi
d
tu
/
j
i
/
norte
h
a
r
t
i
C
mi
–
pag
d
yo
F
/
/
/
/
4
7
2
1
9
3
1
7
0
6
5
9
7
/
j
i
norte
h
_
a
_
0
0
9
7
7
pag
d
.
F
b
y
gramo
tu
mi
s
t
t
oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3
A F RI C A N EC O N O M IC H IS T O R Y
|
195
By the 1980s, sin embargo, three forces had combined to push
African economic history to the intellectual periphery. Primero, African
economies were in decline. The oil shocks of the 1970s and the
consequent debt crises and IMF (International Monetary Fund)
structural-adjustment programs had diminished the resources avail-
able to African universities and deflated interest in explaining Africa’s
past fortunes. Segundo, as postmodernism became fashionable, histo-
rians concentrated more on cultural and social history than on eco-
nomic development. Tercero, the cliometrics movement of the 1970s,
which emphasized the mathematical and statistical aspects of eco-
nomics, heightened the methodological barrier between economics
e historia. As a result of these trends, African economic history,
both inside and outside Africa, went into a decline from which it
did not begin to emerge until the 2000s. Led by prominent econo-
mists, the renaissance of African economic history coincided with the
rise of Africa’s economic prospects. A methodological and techno-
logical revolution within the economics profession certainly had
something to do with the change. The new millennium also marked
the dawn of the data revolution in African history.4
Acemoglu, Johnson, and Robinson are widely credited with
the resurgence of African economic history through their provoc-
ative claim that disparities in economic development were largely
due to the institutional legacies of European settlers. Their main
contribución, sin embargo, was not so much their large new data
set as it was the causal inference that they drew from it using a
Journal of Development Studies, I (1965), 123–141; Anthony G. Hopkins, “Economic Imperialism
in West Africa: Lagos, 1880–1921,” Economic History Review, XXI (1968), 580–606; Manning,
“Slaves, Palm Oil, and Political Power on the West African Coast,” African Historical Studies,
II (1969), 279–288; David Eltis, “The Export of Slaves from Africa, 1821–1843,” Journal of Eco-
nomic History, XXVII (1977), 409–433; Ralph A. Austen, “Slavery among Coastal Middlemen:
The Duala of Cameroon,” in Suzanne Miers and Igor Kopytoff (editores.), Slavery in Africa: Histórico
and Anthropological Perspectives (Madison, 1977), 305–333.
4 Erik Green and Pius Nyambara, “The Internationalization of Economic History: Perspec-
tives from the African Frontier,” Economic History of Developing Regions, XXX (2015), 68–78
(doi:10.1080/20780389.2015.1025744)—argue that African economic history declined only in
Western institutions; Gareth Austin, “African Economic History in Africa,” ibid., 79–94
(doi:10.1080/20780389.2015.1033686), suggests that it was also in decline in Africa. See also
Morten Jerven, “African Growth Recurring: An Economic History Perspective on African
Growth Episodes, 1690–2010,” ibid., XXV (2010), 127–154; Austin and Stephen Broadberry,
"Introducción: The Renaissance of African Economic History,” Economic History Review,
LXVII (2014), 893–906; Fourie and Leigh Gardner, “The Internationalization of Economic
Historia: A Puzzle,” Economic History of Developing Regions, XXIX (2014), 1–14.
yo
D
oh
w
norte
oh
a
d
mi
d
F
r
oh
metro
h
t
t
pag
:
/
/
d
i
r
mi
C
t
.
metro
i
t
.
mi
d
tu
/
j
i
/
norte
h
a
r
t
i
C
mi
–
pag
d
yo
F
/
/
/
/
4
7
2
1
9
3
1
7
0
6
5
9
7
/
j
i
norte
h
_
a
_
0
0
9
7
7
pag
d
.
F
b
y
gramo
tu
mi
s
t
t
oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3
196
| J O H A N F O U R IE
novel econometric technique. Their claim about the persistence of
institutions attracted the interest of a younger generation of econ-
omists, eager to tackle the issues of African poverty and under-
desarrollo. Por ejemplo, Nunn based his extensive study of the
economic consequences of the slave trade on Curtin’s Big Data pro-
ject in African history from the 1960s, which culminated, following
much debate and controversy, in Eltis and Richardson’s Transatlantic
Slave Trade Database. Although the various versions of the slave data-
base that appeared during the past few decades have enriched a vast
scholarship about this issue, Nunn’s demonstration of a causal link
between the number of slaves exported and the level of African GDP
suggested a new direction for African economic history. A pesar de
the article later came under severe criticism, the consensus is that it
rekindled economists’ interest in African economic history.
Nunn’s later work (with Wantchekon) identified mistrust in
African societies as a legacy of the slave trade and (with Puga) presentado
how the slave trade pushed Africans into locations that still hamper
their ability to trade. These contributions laid the foundation for a
new trend in African economic history—the attempt to discover
causal mechanisms in the past to explain conditions in the present.5
Given that few effects are more persistent than geographical
unos, environmental information is crucial for linking historical
events to current circumstances. Nunn and Puga posited rugged ter-
rain as an explanatory variable. Alsan used temperature and humidity
5 Daron Acemoglu, Simon Johnson, and James A. robinson, “The Colonial Origins of
Comparative Development: An Empirical Investigation,” American Economic Review, XCI
(2001), 1369–1401; Nathan Nunn, “The Long-Term Effects of Africa’s Slave Trades,” Quarterly
Journal of Economics, CXXIII (2008), 139–176; Philip D. Curtin, The Atlantic Slave Trade: A
Census (Madison, 1972); José E.. Inikori, “Measuring the Atlantic Slave Trade: An Assessment
of Curtin and Anstey,” Journal of African History, XVII (1976), 197–223; Paul E. Lovejoy,
“The Impact of the Atlantic Slave Trade on Africa: A Review of the Literature," Diario de
African History, XXX (1989), 365–394; Eltis and David Richardson (editores.), Extending the
Frontiers: Essays on the New Transatlantic Slave Trade Database (nuevo refugio, 2008); Manning,
Slavery and African Life: Occidental, Oriental, and African Slave Trades (Nueva York, 1990); Lovejoy,
The African Diaspora: Revisionist Interpretations of Ethnicity, Culture and Religion under Slavery
(Bostón, 1997); Ewout Frankema and Marlous van Waijenburg, “Structural Impediments to
African Growth? New Evidence from Real Wages in British Africa, 1880–1965,” Journal of
Economic History, LXXII (2012), 895–926. The reason for the criticism is more explicit in the
working paper version, available at http://eh.net/eha/wp-content/uploads/2013/11/Frankema
Waijenburg_0.pdf. Nunn and Leonard Wantchekon, “The Slave Trade and the Origins of
Mistrust in Africa,” American Economic Review, CI (2011), 3221–3252; Nunn and Diego
Puga, “Ruggedness: The Blessing of Bad Geography in Africa,” Review of Economics and Statistics,
XCIV (2012), 20–36.
yo
D
oh
w
norte
oh
a
d
mi
d
F
r
oh
metro
h
t
t
pag
:
/
/
d
i
r
mi
C
t
.
metro
i
t
.
mi
d
tu
/
j
i
/
norte
h
a
r
t
i
C
mi
–
pag
d
yo
F
/
/
/
/
4
7
2
1
9
3
1
7
0
6
5
9
7
/
j
i
norte
h
_
a
_
0
0
9
7
7
pag
d
.
F
b
y
gramo
tu
mi
s
t
t
oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3
197
|
A F RI C A N EC O N O M IC H IS T O R Y
to assess the influence of the tsetse fly on development in Africa.
Bhattacharyya advanced the prevalence of malaria as the most im-
portant factor in explaining African poverty, at least from a statistical
perspectiva. Fenske and Kala’s annual panel of African temperatures
and slave exports indicated how past environmental shocks affected
the slave trade. Papaioannou’s research into court cases and the re-
cords of prisoners and homicides revealed how deviations from the
long-term rainfall pattern increased conflict in colonial Nigeria.
Papaioannou and de Haas extended this analysis to include other
parts of Africa with similar results. Climatic and other environmental
information geocoded at the micro-level enabled scholars to test
causal persistence with a high degree of statistical accuracy.6
The new variables gleaned from the data revolution not only
measure past events; they also serve effectively as contemporary
outcome variables. In the absence of data about regional African
economías, Michalopoulos and Papaioannou employed light den-
sity at night, obtained from satellite imaging, to test, on the one
mano, the effect of precolonial ethnic institutions and, en el otro,
the effect of the borders created during the European powers’
scramble for Africa on current economic performance. Notwith-
standing the strong criticism that Michalopoulos and Papaiannou’s
data sources and estimation techniques have attracted, light density
has become a popular outcome variable.7
6 Marcella Alsan, “The Effect of the TseTse Fly on African Development,” American Eco-
nomic Review, CV (2015), 382–410; Sambit Bhattacharyya, “Root Causes of African Under-
desarrollo,” Journal of African Economies, XVIII (2009), 745–780; James Fenske and Namrata
Kala, “Climate and the Slave Trade,” Journal of Development Economics, CXII (2015), 19–32;
Kostadis J. Papaioannou, “Climate Shocks and Conflict: Evidence from Colonial Nigeria,"
African Economic History Network Working Paper Series, 17 (2014); Papaioannou and
Michiel de Haas, “Climate Shocks, Cash Crops and Resilience: Evidence from Colonial
Tropical Africa,” paper presented at the World Economic History Congress, August 3–8,
2015, Kioto.
Stelios Michalopoulos and Papaioannou, “Pre Colonial Ethnic Institutions and Con-
7
temporary African Development,” Econométrica, LXXXI (2013), 113–152; ídem, “Further
Evidence on the Link between Pre-colonial Political Centralization and Comparative Eco-
nomic Development in Africa,” Economics Letters, 126 (2015), 57–62; ídem, “National
Institutions and Subnational Development in Africa,” Quarterly Journal of Economics, CXXIX
(2014) 151–213; Timothy Besley and Marta Reynal-Querol, “The Legacy of Historical Conflict:
Evidence from Africa,” American Political Science Review, CVIII (2014), 319–336; Nonso Obikili,
“An Examination of Subnational Growth in Nigeria: 1999–2012,” South African Journal of
Ciencias económicas, LXXXXIII (2015), 335–357; Denis Cogneau and Yannick Dupraz, “Questionable
Inference on the Power of Pre-Colonial Institutions in Africa,” PSE Working Papers,
No. 2014–25 (2014).
yo
D
oh
w
norte
oh
a
d
mi
d
F
r
oh
metro
h
t
t
pag
:
/
/
d
i
r
mi
C
t
.
metro
i
t
.
mi
d
tu
/
j
i
/
norte
h
a
r
t
i
C
mi
–
pag
d
yo
F
/
/
/
/
4
7
2
1
9
3
1
7
0
6
5
9
7
/
j
i
norte
h
_
a
_
0
0
9
7
7
pag
d
.
F
b
y
gramo
tu
mi
s
t
t
oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3
198
| J O H A N F O U R IE
Contemporary survey data, now with geocoded observations,
also frequently become outcome variables. A case in point is the
“Afrobarometer survey” by Nunn and Wantchekon to measure
levels of trust. Abel’s work regarding forced removals during apart-
heid also relies on the Afrobarometer survey. Michalopoulos,
Putterman, and Weil’s use of Demographic and Health Survey
(DHS) data in combination with information about the respon-
dents’ ancestral ethnicity discovered that descendants of agricultur-
alists are wealthier and better-educated than the descendants of
pastoralists.8
Not everyone agrees that such broad-brush studies, cual
link a historical episode to contemporary data, are always edifying.
The main complaint about such an approach is that it “compresses
historia,” oversimplifying a more elaborate causality. In a recent book,
Jerven criticizes what he calls the irresponsible manner in which
economists have deployed econometrics to explain African devel-
opment. In reviewing this book, de Waal calls for African economic
history to be “liberated from the tyranny of econometricians.”9
African economic historians have proposed an alternative
methodology that involves unearthing and digitizing colonial-era
archival records. For the last decade, scholars have scanned colonial
blue books (a compendium of records about the civil establish-
mento, revenue and expenditure, and other statistical particulars
of the colonies issued by the Colonial Office), tax censuses, voter
rolls, marriage registers, etc., to study the population size, wages,
incomes, education, fiscal systems, and transport networks of
8 The Afrobarometer survey series measures public attitudes about economic, political, y
social matters in several sub-Saharan African countries. More information is available at
http://www.afrobarometer.org/. Martin Abel, “Long-Run Effects of Forced Removal under
Apartheid on Social Capital,” paper presented at African Economic History meetings, Londres
School of Economics and Political Science, October 25–26, 2014; Michalopoulos, luis
Putterman, and David N. Weil, “The Influence of Ancestral Lifeways on Individual Eco-
nomic Outcomes in Sub-Saharan Africa,” paper discussed at NBER Reporter Number 4:
Program and Working Group Meetings (2014).
9 austin, “The ‘Reversal of Fortune’ Thesis and the Compression of History: Perspectives
from African and Comparative Economic History,” Journal of International Development,
XX (2008), 996–1027; Jerven, África: Why Economists Get it Wrong (Londres, 2015); ídem
and Deborah Johnston, “Statistical Tragedy in Africa? Evaluating the Data Base for African
Economic Development,” Journal of Development Studies, LI (2015), 111–115; Alex de Waal,
“Liberating African Economic History from the Tyranny of Econometric,” review of Jerven,
África: Why Economists Get it Wrong, available at: http://africanarguments.org/2015/06/24/
liberating-african-economic-history-from-the-tyranny-of-econometrics-by-alex-de-waal/.
yo
D
oh
w
norte
oh
a
d
mi
d
F
r
oh
metro
h
t
t
pag
:
/
/
d
i
r
mi
C
t
.
metro
i
t
.
mi
d
tu
/
j
i
/
norte
h
a
r
t
i
C
mi
–
pag
d
yo
F
/
/
/
/
4
7
2
1
9
3
1
7
0
6
5
9
7
/
j
i
norte
h
_
a
_
0
0
9
7
7
pag
d
.
F
b
y
gramo
tu
mi
s
t
t
oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3
A F RI C A N EC O N O M IC H IS T O R Y
|
199
African societies. Rather than a lengthy survey of every contribution
within this context, a single type of archival document—the military
attestation form, que incluye, among other things, soldiers’
heights—can represent the nature of the data revolution in African
economic history. These individual-level observations produce a
valuable proxy for living standards in the absence of other evidence.10
HOW HEIGHT DATA CAN HELP TO UNLOCK THE AFRICAN PAST The
use of human heights or stature as a proxy for living standards is
more than three decades old. Social scientists agree that height
10 Manning, “Historical Datasets on Africa and the African Atlantic,” Journal of Comparative
Ciencias económicas, XL (2012), 604–607; Frankema and Jerven, “Writing History Backwards or Side-
maneras: Towards a Consensus on African Population, 1850–2010,” Economic History Review,
LXVII (2014), 907–931; Fourie and Erik Green, “The Missing People: Accounting for the
Productivity of Indigenous Populations in Cape Colonial History,” Journal of African History,
LVI (2015), 195–215; Frankema and van Waijenburg, “Structural Impediments”; Rönnbäck
Klas, “Living Standards on the Pre-Colonial Gold Coast: A Quantitative Estimate of African
Laborers’ Welfare Ratios,” European Review of Economic History, XVIII (2014), 185–202;
Sophia Du Plessis and Stan du Plessis, “Happy in the Service of the Company: The Purchasing
Power of VOC Salaries at the Cape in the 18th Century,” Economic History of Developing
Regions, XXVII (2012), 125–149; Jerven, “A West African Experiment: Constructing a GDP
Series for Colonial Ghana, 1891–1950,” Economic History Review, LXVII (2014), 964–992;
Leandro Prados de la Escosura, “Output per Head in Pre-Independence Africa: Quantitative
Conjectures,” Economic History of Developing Regions, XXVII (2012), 1–36; Fourie and Jan
Luiten van Zanden, “GDP in the Dutch Cape Colony: The National Accounts of a Slave-
Based Society,” South African Journal of Economics, LXXXI (2013), 467–490; Wantchekon,
Marko Klašnja, and Natalija Novta, “Education and Human Capital Externalities: Evidencia
from Colonial Benin,” Quarterly Journal of Economics, CXXX (2015), 703–757; Fourie and
Dieter von Fintel, “Settler Skills and Colonial Development: The Huguenot Wine Makers in
Eighteenth Century Dutch South Africa,” Economic History Review, LXVII (2014), 932–963;
Nonso, Obikili, “Social Capital and Human Capital in the Colonies: A Study of Cocoa Farmers
in Western Nigeria,” Economic History of Developing Regions, XXX (2015), 1–22 (doi: 10.1080/
20780389.2015.1012712); Jörg Baten, and Fourie, “Numeracy of Africans, Asians, and Europeans
during the Early Modern Period: New Evidence from Cape Colony Court Registers,” Economic
History Review, LXVIII (2015), 632–656; jardinero, Taxing Colonial Africa: The Political Economy of
British Imperialism (Nueva York, 2012); Fourie, Ada Jansen, and Krige Siebrits, “Public Finances and
Private Company Rule: The Dutch Cape Colony (1652–1795),” New Contree, 68 (December
2013), 1–22; Frankema and van Waijenburg, “Metropolitan Blueprints of Colonial Taxation?
Lessons from Fiscal Capacity Building in British and French Africa, C. 1880–1940,” Journal of
African History, LV (2014), 371–400; Frankema, “Colonial Taxation and Government Spending
in British Africa, 1880–1940: Maximizing Revenue or Minimizing Effort?” Explorations in Eco-
nomic History, XLVIII (2011), 136–149; Papaioannou and Angus Edwin Dalrymple-Smith,
“Political Instability and Discontinuity in Nigeria: The Pre-Colonial Past and Public Goods
Provision under Colonial and Post-Colonial Political Orders,” Economics of Peace and Security
Diario, X (2015), 40–53; Remi Jedwab and Alexander Moradi, “The Permanent Effects of
Transportation Revolutions in Poor Countries: Evidence from Africa,” Review of Economics
and Statistics (próximo).
yo
D
oh
w
norte
oh
a
d
mi
d
F
r
oh
metro
h
t
t
pag
:
/
/
d
i
r
mi
C
t
.
metro
i
t
.
mi
d
tu
/
j
i
/
norte
h
a
r
t
i
C
mi
–
pag
d
yo
F
/
/
/
/
4
7
2
1
9
3
1
7
0
6
5
9
7
/
j
i
norte
h
_
a
_
0
0
9
7
7
pag
d
.
F
b
y
gramo
tu
mi
s
t
t
oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3
200
| J O H A N F O U R IE
accurately reflects an individual’s environmental conditions early in
vida, including access to nutrition and exposure to disease, y eso
changes in average height are reflective of changes in a society’s living
standards. Although heights might be expected to correlate posi-
tively with incomes, economic historians analyzing heights in
Western Europe and North America discovered a surprisingly neg-
ative correlation between height and income during the early phase
of industrialization in England and elsewhere. This apparent
anomaly became known as the “Early Industrial Growth Puzzle”
in Europe and the “Antebellum Puzzle” in the United States; para
most of the last two decades of the twentieth-century, scholars on
both sides of the Atlantic have attempted to explain it.11
The height of nineteenth-century Africans was the subject of
two articles by Eltis in 1982 y 1990, but the turn to heights to
document the evolution in the living standards of African peoples
during an era of unreliable data had to await Moradi and Baten’s
pioneering investigation of height differences between historical
birth cohorts constructed from DHS data. But DHS data were not
available until the colonial era ended in the 1950s. Another source
was needed to investigate the colonial and the precolonial eras.
Moradi found a sample of 1,046 Ghanaian recruits from World
War I and World War II and 730 Kenyan recruits from World
Segunda Guerra, adding information from surveys, to measure the effect
of colonial policies on African living standards. In a follow-up
work about the same question, his sample more than doubled.12
11 Robert W. Fogel et al., “Secular Changes in American and British Stature and Nutri-
ción,” Revista de Historia Interdisciplinaria, XIV (1983), 445–481; ídem, Stanley L. Engerman, y
James Trussell, “Exploring the Uses of Data on Height: The Analysis of Long-Term Trends in
Nutrition, Labor Welfare, and Labor Productivity,” Social Science History, VI (1982), 401–421;
Richard H. Steckel, “Height and Per Capita Income,” Historical Methods: A Journal of Quan-
titative and Interdisciplinary History, XVI (1983), 1–7; Steckel, “Stature and the Standard of
Living,” Journal of Economic Literature, XXXII (1995), 1903–1940; George Alter, “Height, Frailty,
and the Standard of Living: Modeling the Effects of Diet and Disease on Declining Mortality
and Increasing Height,” Population Studies, LVIII (2004), 265–279; Angus Deaton, “Height,
Salud, and Inequality: The Distribution of Adult Heights in India,” American Economic Review,
XCVIII (2008), 468; John Komlos, “Shrinking in a Growing Economy? The Mystery of Physical
Stature during the Industrial Revolution,” Journal of Economic History, LVIII (1998), 779–802;
ídem, “Anomalies in Economic History: Toward a Resolution of the ‘Antebellum Puzzle,’” ibid.,
LVI (1996), 202–214; idem and Bjorn Alecke, “The Economics of Antebellum Slave Heights
Reconsidered,” Revista de Historia Interdisciplinaria, XXVI (1996), 437–457.
12 Eltis, “Nutritional Trends in Africa and the Americas: Heights of Africans, 1819–1839,”
Revista de Historia Interdisciplinaria, XII (1982), 453–475; ídem, “Welfare Trends among the Yoruba
yo
D
oh
w
norte
oh
a
d
mi
d
F
r
oh
metro
h
t
t
pag
:
/
/
d
i
r
mi
C
t
.
metro
i
t
.
mi
d
tu
/
j
i
/
norte
h
a
r
t
i
C
mi
–
pag
d
yo
F
/
/
/
/
4
7
2
1
9
3
1
7
0
6
5
9
7
/
j
i
norte
h
_
a
_
0
0
9
7
7
pag
d
.
F
b
y
gramo
tu
mi
s
t
t
oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3
A F RI C A N EC O N O M IC H IS T O R Y
|
201
Moradi’s groundbreaking work stimulated interest in military
records elsewhere in Africa, often with startling results. austin,
Baten, and van Leeuwen found that in nineteenth-century West
África, Ghanaian and Burkinabe recruits were notably shorter than
northwestern Europeans but not shorter than southern Europeans.
In twentieth-century West Africa, Cogneau and Rouanet found
that the rate of increase in the heights of those born in Côte d’Ivoire
and Ghana during the late colonial period, 1925 a 1960, was almost
as fast as the rate observed in France and Great Britain during the
period from 1875 a 1975. Inwood and Masakure found that colored
South Africans were, on average, six cm shorter than white South
Africans at the start of the twentieth century, a significantly smaller
difference than today’s eight cm. The findings from these studies
now take their place within a larger literature about the living stan-
dards of indigenous populations across the world.13
in the Early Nineteenth Century: The Anthropometric Evidence,” Journal of Economic History,
l (1990), 521–540; Alexander Moradi and Baten, “Inequality in Sub-Saharan Africa: New Data and
New Insights from Anthropometric Estimates,” World Development, XXXII (2005), 1233–1265.
Other pioneering work included David E. Sahn and David C. Stifel, “Urban-Rural Inequality in
Living Standards in Africa,” Journal of African Economies, XII (2003), 564–597. Moradi, “Confront-
ing Colonial Legacies: Lessons from Human Development in Ghana and Kenya, 1880–2000,”
Journal of International Development, XX (2008), 1107–1121, ídem, “Towards an Objective Account
of Nutrition and Health in Colonial Kenya: A Study of Stature in African Army Recruits and
Civilians, 1880–1980,” Journal of Economic History, LXIX (2009), 719–754.
13 austin, Baten, and Bas Van Leeuwen, “The Biological Standard of Living in Early
Nineteenth Century West Africa: New Anthropometric Evidence for Northern Ghana and
Burkina Faso,” Economic History Review, LXV (2012), 1280–1302; Denis Cogneau and Léa
Rouanet. “Living Conditions in Côte d’Ivoire and Ghana, 1925–1985: What Do Survey Data
on Height Stature Tell Us?” Economic History of Developing Regions, XXVI (2011), 55–82; Kris
Inwood and Oliver Masakure, “Poverty and Physical Well-Being among the Coloured Popu-
lation in South Africa,” Economic History of Developing Regions, XXVIII (2013), 56–82; José M..
Prince and Steckel, “Nutritional Success on the Great Plains: Nineteenth-Century Equestrian
Nomads,” Revista de Historia Interdisciplinaria, XXXII (2003), 353–384; Howard Bodenhorn,
“The Mulatto Advantage: The Biological Consequences of Complexion in Rural Antebellum
Virginia,” ibid. (2002), 21–46; Aravinda Meera Guntupalli and Baten, “The Development
and Inequality of Heights in North, Oeste, and East India 1915–1944,” Explorations in Economic
Historia, XLIII (2006), 578–608; Inwood, Les Oxley, and Evan Roberts, “Physical Growth and
Ethnic Inequality in New Zealand Prisons, 1840–1975,” History of the Family, XX (2015), 249–
269 (doi:10.1080/1081602X.2015.1006653); Baten, Ines Pelger, and Linda Twrdek, "El
Anthropometric History of Argentina, Brazil and Peru during the 19th and early 20th Century,"
Ciencias económicas & Human Biology, VII (2009), 319–333; Baten, Mojgan Stegl, and Pierre van der
Eng, “The Biological Standard of Living and Body Height in Colonial and Post-Colonial
Indonesia, 1770–2000,” Journal of Bioeconomics, XV (2013), 103–122; Baten and Matthias Blum,
“Growing Tall but Unequal: New Findings and New Background Evidence on Anthropometric
Welfare in 156 Countries, 1810–1989,” Economic History of Developing Regions, XXVII (2012),
S66–S85.
yo
D
oh
w
norte
oh
a
d
mi
d
F
r
oh
metro
h
t
t
pag
:
/
/
d
i
r
mi
C
t
.
metro
i
t
.
mi
d
tu
/
j
i
/
norte
h
a
r
t
i
C
mi
–
pag
d
yo
F
/
/
/
/
4
7
2
1
9
3
1
7
0
6
5
9
7
/
j
i
norte
h
_
a
_
0
0
9
7
7
pag
d
.
F
b
y
gramo
tu
mi
s
t
t
oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3
202
| J O H A N F O U R IE
Information in attestations is important not only for measur-
ing living standards over time but also for tracing the effects of dif-
ferent colonial policies. Cogneau and Moradi constructed a much
larger sample, 11,940 recruits from Ghana and Togo, to evaluate
how the partition of German Togoland after World War I influ-
enced educational outcomes. The part of Togo that fell under
French mandate had lower levels of literacy after the partition than
did the parts that were ruled by the British. According to Cogneau
and Moradi, the French authorities were hostile to missionary
escuelas. Jedwab and Moradi’s analysis of colonial railways vis à vis
various economic outcomes, including heights, found that railways
increased the heights of those born closest to them and generally
improved income, boosting the production of cash crops like cocoa.14
Mariotti’s investigation of the heights of black mineworkers
in South Africa confirmed the positive effect of income improve-
ment on height. A 1974 plane crash that killed seventy-three
mineworkers provoked the Malawian government to ban the
migration of mineworkers to South Africa, forcing South Africa
to hire workers from the Transkei “homeland” in South Africa.
Mariotti showed that the resulting improvement in household
incomes for the newly recruited Transkeian mineworkers increased
the heights of children born during or immediately after 1974, y
only in those districts from which mineworkers came. Mariotti and
Dinkelman later turned their attention to the effects of the plane
crash and the sudden prevention of labor migration on Malawian
households.15
Attestation records provide a snapshot of living standards in the
absence of the other individual-level records that colonial author-
ities were often less-inclined to collect for indigenous populations
than for Europeans. The demographic information that is available
typically comes with a colonial-era bias that is difficult to exclude.
Although height information collected for military purposes is
14 Cogneau and Moradi, “Borders that Divide: Education and Religion in Ghana and
Togo since Colonial Times,” Journal of Economic History, LXXIV (2014), 694–729; Jedwab and
Moradi, “Permanent Effects.”
15 Martine Mariotti, “Fathers’ Employment and Sons’ Stature: The Long-Run Effects of a
Positive Regional Employment Shock in South Africa’s Mining Industry,” Economic Develop-
ment and Cultural Change, LXIII (2015), 485–514; Taryn Dinkelman and Mariotti, “Does
Labor Migration Affect Human Capital in the Long Run? Evidence from Malawi,” unpub.
paper (Dartmouth College, 2014).
yo
D
oh
w
norte
oh
a
d
mi
d
F
r
oh
metro
h
t
t
pag
:
/
/
d
i
r
mi
C
t
.
metro
i
t
.
mi
d
tu
/
j
i
/
norte
h
a
r
t
i
C
mi
–
pag
d
yo
F
/
/
/
/
4
7
2
1
9
3
1
7
0
6
5
9
7
/
j
i
norte
h
_
a
_
0
0
9
7
7
pag
d
.
F
b
y
gramo
tu
mi
s
t
t
oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3
A F RI C A N EC O N O M IC H IS T O R Y
|
203
unlikely to demonstrate such systematic bias, attestation forms can
have a selection bias of their own. Not everyone qualified for mil-
itary service. The minimum-height requirement meant that attes-
tation heights were at times truncated to the left. Truncation is
amenable to a statistical check (by using truncated regression
modelos, Por ejemplo), but another unobservable selection could
always be lurking. En efecto, Bodenhorn, Guinnane, and Mroz have
recently addressed this very possibility. They claim that the declin-
ing heights observed during the Industrial Revolution—the “Early
Industrial Growth Puzzle”—reflects not a decline in living stan-
dards but a change in the military’s selection process. When returns
in the private sector were small (at the start of the Industrial Revo-
lution), many of the stronger (or taller) men probably chose the
military as a career. As the economy grew, sin embargo, more of them
would have returned to private-sector employment, with no con-
comitant increase in the military population, leaving the weaker
(or shorter) men to join the army. Además, Inwood, Mariotti,
and Fourie used data about South African recruits in World War I
to show that this type of unobservable selection may also happen
because of changes in military technology.16
Selection-bias concerns, sin embargo, cannot prevent heights
from being a useful proxy for individual-level living standards in
the absence of other measures. A comparison of the results derived
from military attestation forms together with those derived from
survey data, Por ejemplo, suggests that selection into the military
may not have been strong enough to undermine our confidence
in earlier findings. The studies that have been done on heights
exemplify the contribution of the data revolution in African history
to new insights about how colonial-era policies affected the welfare
of subjects who remained outside the remit of recordkeeping.
Besides heights, attestation forms offer a wealth of data that
can contribute to fields outside economic history, in particular social
historia. The attestation forms of the South African Constabulary list
several characteristics of each recruit, including the ability to speak an
See for example, Bodenhorn, Timothy W. Guinnane, and Thomas A. Mroz, “Sample-
16
Selection Biases and the ‘Industrialization Puzzle,’” NBER Working Paper 21249 (2015);
ídem, “Caveat Lector: Sample Selection in Historical Heights and the Interpretation of Early
Industrializing Economies,” NBER Working Paper 19955 (2014). Fourie, Inwood, y
Mariotti, “Can Historical Changes in Military Technology Explain the Industrial Growth
Puzzle?” unpub. paper (London School of Economics, 2014).
yo
D
oh
w
norte
oh
a
d
mi
d
F
r
oh
metro
h
t
t
pag
:
/
/
d
i
r
mi
C
t
.
metro
i
t
.
mi
d
tu
/
j
i
/
norte
h
a
r
t
i
C
mi
–
pag
d
yo
F
/
/
/
/
4
7
2
1
9
3
1
7
0
6
5
9
7
/
j
i
norte
h
_
a
_
0
0
9
7
7
pag
d
.
F
b
y
gramo
tu
mi
s
t
t
oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3
204
| J O H A N F O U R IE
African language, the color of hair and eyes, vaccination history, el
ability to ride a horse or swim, and the presence of any tattoos. A
simple quantitative analysis can bring to light unsuspected details.
More than one-in-five Boer recruits from the two Boer Republics
had at least one tattoo, and the number of Boer recruits who were
redheads was nearly twice that of the Scottish or Irish recruits.17
DATA TRANSCRIPTION, AUGMENTATION, AND COLLABORATION The
use of large archival data sets is only one aspect of the data revo-
lution. New tools for data transcription, augmentation, and collabo-
ration allow for faster and less-expensive digitization, more rigorous
investigación, and better replicability and comparability of results.
The data revolution comprises not only new material but also a
new method—the use of computational power and statistical tech-
niques to expand the scale and scope of research questions.
Data transcription—the manual copying of archival sources,
often into a digital format such as Excel—is laborious and expen-
sive. Como ejemplo, Cifra 1 shows the attestation of Diederick
Alfred Joseph Yates, a Briton who enlisted in the South African
Constabulary in 1902. Although the form is standardized and the
text clearly legible to anyone familiar with late nineteenth-century
handwriting, it does not allow for automated transcription. Optical
character recognition (OCR) software is not yet able to capture the
information contained in such an attestation with a high degree of
fiabilidad. Historians must therefore spend countless hours tran-
scribing these attestations into analyzable text or train a research
team to do it. Both options are costly. They can also cut corners,
as Inwood and Masakure did when investigating South African
soldiers who served in World War I and World War II. A pesar de
several hundred thousand attestations are available in the Defense
Force archives, time and funding constraints forced Inwood and
Masakure to limit their transcriptions to only 10,000 attestations
for each of the wars. Such sampling introduces additional biases
for which researchers must take account when analyzing the results.
It also does not fully conform to the methodology now popular in
the era of Big Data, which advocates the use of all available data.
Fourie, Inwood, and Mariotti, “‘Poor South Africa! Will No Nice English People Ever
17
Come Out Here?’—The South African Constabulary of the Second South African War,"
unpub. paper (Stellenbosch University, 2015).
yo
D
oh
w
norte
oh
a
d
mi
d
F
r
oh
metro
h
t
t
pag
:
/
/
d
i
r
mi
C
t
.
metro
i
t
.
mi
d
tu
/
j
i
/
norte
h
a
r
t
i
C
mi
–
pag
d
yo
F
/
/
/
/
4
7
2
1
9
3
1
7
0
6
5
9
7
/
j
i
norte
h
_
a
_
0
0
9
7
7
pag
d
.
F
b
y
gramo
tu
mi
s
t
t
oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3
A F RI C A N EC O N O M IC H IS T O R Y
|
205
Fig.1
An Attestation Form from the South African Constabulary
yo
D
oh
w
norte
oh
a
d
mi
d
F
r
oh
metro
h
t
t
pag
:
/
/
d
i
r
mi
C
t
.
metro
i
t
.
mi
d
tu
/
j
i
/
norte
h
a
r
t
i
C
mi
–
pag
d
yo
F
/
/
/
/
4
7
2
1
9
3
1
7
0
6
5
9
7
There is reason to believe, sin embargo, that improved OCR soft-
ware will soon make the transcription of archival sources, como
attestation forms, less costly. This rapidly developing technology—
the most spectacular use of which is Google’s project to scan every
book title in existence (estimated in 2010 to be about 130 millón)—
already has wide applicability in history, for example in the digiti-
zation of old newspapers. But printed texts are relatively easy for
the OCR technology to digitize and transcribe. Eventually an OCR
algorithm will be able to read a natural image—like the attestation
in Figure 1—that contains written text. This technological break-
through in computational linguistics could have far-reaching con-
sequences for the field of economic history.18
/
j
i
norte
h
_
a
_
0
0
9
7
7
pag
d
.
F
b
y
gramo
tu
mi
s
t
t
oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3
18 Maya R. Gupta, Nathaniel P. jacobson, and Eric K. Garcia, “OCR Binarization and
Image Pre-Processing for Searching Historical Documents,” Pattern Recognition, XL (2007),
389–397; Sergey Milyaev et al., “Image Binarization for End-to-End Text Understanding
in Natural Images,” paper presented at the Twelfth International Conference on Document
Analysis and Recognition (ICDAR), August 25–28, 2013, Washington, D.C.
206
| J O H A N F O U R IE
Data augmentation, the second step in the new methodology,
signifies the addition of information from internal or external
sources to the original data. Although supplementing existing data
with new information has always been possible, researchers usually
had to do it manually. Computerization and the development of
matching algorithms have made this process much easier. En el
case of the attestation forms, scholars can now match recruits’
names and birth dates to, decir, genealogical records. Not only
can they learn more about their subjects; they can also solve pre-
viously intractable problems, such as calculating intergenerational
mobility using height as an indicator of living standards. Data aug-
mentation has countless new avenues to explore.19
The third step is collaborating across research networks. Datos
sharing has already resulted in several projects to compare historical
measures. At a macroeconomic level, series such as the Penn World
Tables and the Maddison project have provided scholars with cross-
country comparisons in history. Manning’s Big Data in History
project aims to “create a world-historical archive that will trace
the last four centuries of historical dynamics and change.” The
Global Collaboratory on the History of Labour Relations, based at
the International Institute of Social History in Amsterdam, tiene
assembled a team to collect quantitative benchmarks for a global
labor history—including population size and composition, trabajar-
force size and composition, occupations, and types of labor—setting
the years 1500, 1650, 1800, 1900, y 2000 as benchmarks (1950 es
included for Africa). Van Zanden has undertaken the Clio Infra pro-
ject, which aims to establish “a set of interconnected databases . . .
containing worldwide data on social, económico, and institutional
indicators for the past five centuries, with special attention to the
pasado 200 years.” This project has now evolved into the CLARIAH
(Common Lab Research Infrastructure for the Arts and Humanities)
proyecto, consisting of a consortium of more than forty partners who
store and share all types of data related to the arts and humanities in a
standardized and user-friendly format.20
Ver, Por ejemplo, James J. Feigenbaum, “Automated Census Record Linking,” unpub. paper
19
(Harvard University, 2015), available at http://scholar.harvard.edu/files/jfeigenbaum/files/
feigenbaum-censuslink.pdf.
20 Robert Feenstra, Robert Inklaar, and Marcel Timmer, “The Next Generation of the
Penn World Table,” NBER Working Paper 19255 (2013); Jutta Bolt and van Zanden,
“The Maddison Project: Collaborative Research on Historical National Accounts,” Economic
yo
D
oh
w
norte
oh
a
d
mi
d
F
r
oh
metro
h
t
t
pag
:
/
/
d
i
r
mi
C
t
.
metro
i
t
.
mi
d
tu
/
j
i
/
norte
h
a
r
t
i
C
mi
–
pag
d
yo
F
/
/
/
/
4
7
2
1
9
3
1
7
0
6
5
9
7
/
j
i
norte
h
_
a
_
0
0
9
7
7
pag
d
.
F
b
y
gramo
tu
mi
s
t
t
oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3
A F RI C A N EC O N O M IC H IS T O R Y
|
207
In African history, Frankema and his team have taken the lead
in digitizing and transcribing the colonial blue books, facilitating
research about colonial education, public finance, and population.
In an award-winning study, Frankema and van Waijenburg dem-
onstrated that African real wages were significantly above subsis-
tence levels and rising for most of the colonial period. En algunos
lugares, wages were much higher than comparable figures for Asia.
That study suggests that the idea of sub-Saharan Africa as always
having been poor is not supported by the evidence.21
Wages and prices can reveal much more than simply compar-
ative levels of development. Real wages also help to examine
changes in the degree of inequality between the rich and the poor
(or between the colonizer and the colonized), a topic that has
attracted much attention following the publication of Piketty’s
Capital in the Twenty-First Century. Because historical wage and
price series are often not readily available, several scholars are turn-
ing to such innovative sources as slave-ship provisions, Dutch East
India Company records, or agricultural journals.22
On a broader scale, Jerven has tracked macroeconomic trends
backward into the past. By reconstructing African GDP measures,
he has found considerable discrepancies between World Bank,
IMF, and Penn World Table estimates. In two highly acclaimed
books, Jerven emphasizes that Africa’s image as the “hopeless
History Review, LXVII (2014), 627–651; Manning, Big Data. For the Global Collaboratory, ver
https://collab.iisg.nl/web/labourrelations. Van Zanden et al. (editores.), How Was Life? Global
Well-Being since 1820 (París, 2014).
Frankema and Jerven, “Writing History Backwards or Sideways”; Frankema, "El
21
Origins of Formal Education in Sub-Saharan Africa: Was British Rule More Benign?” European
Review of Economic History, XVI (2012), 335–355; ídem, “Colonial Taxation and Government
Spending in British Africa, 1880–1940: Maximizing Revenue or Minimizing Effort?” Explora-
tions in Economic History, XLVIII (2011), 136–149. Idem and van Waijenburg, “Structural
Impediments,” won the 2013 Arthur H. Cole prize for the best paper published in the Journal
of Economic History.
22 Thomas Piketty (trans. Arthur Goldhammer), Capital in the Twenty-First Century
(Cambridge, Masa., 2014); ídem, and Emmanuel Saez, “Inequality in the Long Run," Ciencia,
CCCXLIV (2014), 838–843. For innovative sources, ver, Por ejemplo, Inikori’s attempt to extract
African market-price data from the slave-trade records to measure precolonial African economies
at https://www.wilsoncenter.org/person/joseph-inikori; Dalrymple-Smith and Frankema’s
collection of prices from slave-ship provisions at http://www.wageningenur.nl/upload_mm/7/
8/8/aa9b6927-3229-4323-b6f4-0a343dea34b4_Dalrymple-Smith%20Frankema%20-%20Paper.
pdf; Sophia du Plessis and Stan du Plessis, “Happy in the Service of the Company”; Willem
Boshoff and Fourie, “When did Globalization Begin in South Africa?” Stellenbosch Working
Paper Series WP10/2015.
yo
D
oh
w
norte
oh
a
d
mi
d
F
r
oh
metro
h
t
t
pag
:
/
/
d
i
r
mi
C
t
.
metro
i
t
.
mi
d
tu
/
j
i
/
norte
h
a
r
t
i
C
mi
–
pag
d
yo
F
/
/
/
/
4
7
2
1
9
3
1
7
0
6
5
9
7
/
j
i
norte
h
_
a
_
0
0
9
7
7
pag
d
.
F
b
y
gramo
tu
mi
s
t
t
oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3
208
| J O H A N F O U R IE
continent”—famously encapsulated by the May 13, 2000, cover of
The Economist—is largely a construction of the 1980s and 1990s.
Bolt and Gardner are compiling fiscal revenue and expenditure
data at the local level for the late colonial period. They pair these
new records with evidence about precolonial states to assess the
extent to which the fiscal capacity of local-government units re-
flects precolonial state centralization. Además, in an attempt to
understand the evolution of inequality during the colonial era (en
the absence of information about income at the individual level),
Bolt and Hillbom are compiling colonial-era records of occupations
and labor-market structures.23
These reinterpretations of the African past are the result of
digitizing and transcribing the vast amounts of data available in colo-
nial archives—projects that are ongoing. Meier zu Selhausen is dig-
itizing and transcribing vast quantities of demographic records kept
in missionary-station archives. These records are not without prob-
lemas. Selection into formal Christian mission stations may carry
hidden biases, and gaining the trust of bishops and others anxious that
the data should be put to proper use is often as time-consuming for
researchers as is the analysis. But the effort is worth the cost because
these records provide a glimpse of African demographic changes
and living standards unrecorded in the colonial blue books.24
The records available for the European settlers and their activi-
ties are more detailed. Verde, von Fintel, and Fourie are constructing
an annual panel data set of several thousand settler farmers for more
than 140 años. Once complete, this data set will offer a wide-ranging
and informative account of eighteenth- and nineteenth-century life
in colonial South Africa and allow economists to investigate the
Jerven, “African Growth Recurring”; ídem, “For Richer, For Poorer: GDP Revisions
23
and Africa’s Statistical Tragedy,” African Affairs, CXII (2012), 138–147; ídem, Poor Numbers:
How We Are Misled by African Development Statistics and What to Do about It (Ítaca, 2013); ídem,
África: Why Economists Get It Wrong; Bolt and Gardner, “De-compressing History? Pre-colonial
Institutions and Local Government Finance in British Colonial Africa,” paper presented at
the Economic History Association meeting, Septiembre 11, 2015, Nashville; Bolt and Ellen
Hillbom, “Potential for Diversification? The Role of the Formal Sector in Bechuanaland
Protectorate’s Economy, 1900–65,” Economic History of Developing Regions, XXX (2015), 1–30.
Felix Meier zu Selhausen, “Missionaries and Female Empowerment in Colonial Uganda:
24
New Evidence from Protestant Marriage Registers, 1880–1945,” Economic History of Developing
Regions, XXIX (2014), 74–112; idem and Jacob Weisdorf, “A Colonial Legacy of African Gender
Inequality? Evidence from Christian Kampala, 1895–2011,” Economic History Review, LXIX
(2016), 229–257.
yo
D
oh
w
norte
oh
a
d
mi
d
F
r
oh
metro
h
t
t
pag
:
/
/
d
i
r
mi
C
t
.
metro
i
t
.
mi
d
tu
/
j
i
/
norte
h
a
r
t
i
C
mi
–
pag
d
yo
F
/
/
/
/
4
7
2
1
9
3
1
7
0
6
5
9
7
/
j
i
norte
h
_
a
_
0
0
9
7
7
pag
d
.
F
b
y
gramo
tu
mi
s
t
t
oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3
A F RI C A N EC O N O M IC H IS T O R Y
|
209
economic consequences of such events as the abolition of slavery and
smallpox epidemics. Even though these statistical records were col-
lected for the European colonial administration, they contain detailed
demographic information about the Khoisan, a population group
almost completely neglected in other sources. The scale of this
project—several thousand individuals for each year—does not allow
for matching individuals manually across years. En cambio, an algorithm
confers unique individuals with a distinct ID that permits them to be
observed throughout their entire lifetime. By adding genealogical
records, we can also expand the algorithm to link families across
generaciones. Such an intergenerational panel would not have been
feasible without the advent of enhanced computing and powerful
statistical techniques.25
THE DIFFUSION OF THE DATA REVOLUTION Has the data revolution
produced a better understanding of the African past and the man-
ner in which historical factors still shape African destinies? Puede
be too early to pass judgment. Sin embargo, the new approach has
generated interest far beyond the narrowly defined disciplinary
borders of economic history. Econometricians, geographers, evo-
lutionary biologists, linguists, demographers, sociologists, y
computer scientists are beginning to exploit the rich quantitative
history of Africa using their own methods. Economic historians
should welcome this trend, not only because it generates an audi-
ence for their work but also because it provides new ways to test
conjectures and hypotheses. The risk, sin embargo, is that some eco-
nomic historians may be left behind, anchored to methods created
at a time when quantitative data were less available. Many of the
scholars who lack the statistical skills that the data revolution re-
quires are based in Africa, geographically isolated from the inter-
disciplinary opportunities to acquire them. Of the more than fifty
authors cited in the section above on the new economic history of
África, fewer than ten are from Africa.26
Fourie and Green, “Missing People”; Baten and Fourie, “Numeracy of Africans”; Jeanne
25
Cilliers and Fourie, “New Estimates of Settler Life Span and Other Demographic Trends in
South Africa, 1652–1948,” Economic History of Developing Regions, XXVII (2012), 61–86.
For evidence of Africans’ lack of the technical skills to keep up with advances in their own
26
economic history, see the list of contributors to the special issue “The Economics of Apartheid,"
published in Economic History of Developing Regions, XXIX (2014). For the introduction, ver
yo
D
oh
w
norte
oh
a
d
mi
d
F
r
oh
metro
h
t
t
pag
:
/
/
d
i
r
mi
C
t
.
metro
i
t
.
mi
d
tu
/
j
i
/
norte
h
a
r
t
i
C
mi
–
pag
d
yo
F
/
/
/
/
4
7
2
1
9
3
1
7
0
6
5
9
7
/
j
i
norte
h
_
a
_
0
0
9
7
7
pag
d
.
F
b
y
gramo
tu
mi
s
t
t
oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3
210
| J O H A N F O U R IE
Addressing this dearth of African scholars who are equipped
to participate in the field is necessary for reasons of both fairness
and intellectual diversity. As the new African economic history
has demonstrated, the negative effects of colonialism still linger
into the present; redressing these injustices by helping Africans
to engage in, and contribute to, their own history is the first step
to a more just intellectual discourse. But drawing African scholars
into the field is likely to create more than just equality. It will also
encourage the development of new questions, sources, approaches,
and answers that will broaden and deepen the scope of African
economic history.
A debate in a recent issue of Economic History of Developing
Regions summarizes the factors that contribute to the current intel-
lectual segregation. Green and Nyambara averred that economic
history research at African universities “is not only strong, but re-
mained vibrant even when African economic history was on the
decline at universities elsewhere. The lack of visible output in
major economic history journals is thus not a sign of weakness.
Instead it is an effect of the increased methodological specialization
of economic history in the Western world.” Green and Nyambara
thus promote the engagement of Western economic historians
with African scholars to prevent “regional isolation. En respuesta,
Austin suggested that “resource constraints,” along with “institu-
tional constraints and intellectual priorities,” are responsible for the
slow adoption of quantitative techniques by African scholars: "El
overwhelming priority that economics departments in Africa
rightly give to the study of current problems does not seem to be
combined with an awareness of the uses of history in fulfilling this
mission. Mientras tanto, the institutionalization of the humanities/
social sciences divide in many universities has made it less likely that
history graduates will be equipped to combine qualitative and
quantitative techniques, let alone focus on the latter.”27
Although online access may bring African students into closer
contact with new methodological tools, the existing disparities
Mariotti and Fourie, “The Economics of Apartheid: An Introduction,” Economic History of
Developing Regions, XXIX (2014), 113–125. Gardner and Fourie discuss this problem for other
developing regions in “The Internationalization of Economic History: A Puzzle,” Economic
History of Developing Regions, XXIX (2014), 1–14.
27 Green and Nyambara, “Internationalization of Economic History," 68; austin, “African
Economic History in Africa," 91.
yo
D
oh
w
norte
oh
a
d
mi
d
F
r
oh
metro
h
t
t
pag
:
/
/
d
i
r
mi
C
t
.
metro
i
t
.
mi
d
tu
/
j
i
/
norte
h
a
r
t
i
C
mi
–
pag
d
yo
F
/
/
/
/
4
7
2
1
9
3
1
7
0
6
5
9
7
/
j
i
norte
h
_
a
_
0
0
9
7
7
pag
d
.
F
b
y
gramo
tu
mi
s
t
t
oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3
A F RI C A N EC O N O M IC H IS T O R Y
|
211
may well become more entrenched. It is not entirely clear which
incentives will minimize the likelihood of a continued isolation of
African scholars, or who will offer them. One approach endorsed
by Green and Nyambara, as well as by Austin, is to encourage col-
laborative work through joint funding applications and research
programas. The danger of this strategy, sin embargo, is that the non-
African partners will tend to drive the research agendas. Otro,
more promising, approach is for European and U.S. universities to
recruit and subsidize Ph.D. students from Africa who could begin
the slow but sustainable process of training their African colleagues
and students in the new methods when they return.28
African economic history has already gained much from the data
revolution of the past two decades. Although the major historio-
graphical debates persist, we now know, Por ejemplo, that the
slave trade has made Africans wary of both their neighbors and
their political leaders, that real wages during the early twentieth
century were higher in many African countries than in Asian
countries, and that colonial railways boosted production and al-
tered the spatial distribution of cities. The continuing projects to
transcribe and digitize large numbers of colonial and postcolonial
records are likely to advance our knowledge about Africa’s eco-
nomic past.29
African economic history is, sin embargo, not just about the past.
A major issue for economic historians concerns how to explain
Africa’s apparent failure to sustain growth and development.
The renaissance of African economic history at the start of the
twenty-first century has coincided with a rise in African economies.
But for a variety of external and internal reasons, the momentum
seems to be slowing as pessimism about Africa’s prospects increases.
Can the tools of the data revolution bring a new understanding to
Africa’s past and promote policies for a better future? The answer is
likely to be negative if the African beneficiaries of this would-be
prosperous future cannot take part in the conversation and if
28 An ongoing project to write a freely downloadable textbook about African economic
history recognizes that technology can leapfrog many of the traditional constraints to accessing
higher education. See http://www.aehnetwork.org/textbook/.
29 Nunn and Wantchekon, “Slave Trade and the Origins of Mistrust”; Frankema and
van Waijenburg, “Structural Impediments”; Jedwab and Moradi, “Permanent Effects.”
yo
D
oh
w
norte
oh
a
d
mi
d
F
r
oh
metro
h
t
t
pag
:
/
/
d
i
r
mi
C
t
.
metro
i
t
.
mi
d
tu
/
j
i
/
norte
h
a
r
t
i
C
mi
–
pag
d
yo
F
/
/
/
/
4
7
2
1
9
3
1
7
0
6
5
9
7
/
j
i
norte
h
_
a
_
0
0
9
7
7
pag
d
.
F
b
y
gramo
tu
mi
s
t
t
oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3
212
| J O H A N F O U R IE
non-African scholars in Europe or the United States fail to establish
long-term partnerships with African universities and their faculties.
“The past is never dead. It isn’t even past,” William Faulkner
once said. This statement is nowhere more apt than in Africa,
where the deleterious effects of colonialism and slavery persist.
Equipping African scholars with the tools to take part in the data
revolution is not only necessary to redress the inequalities of the
past but also vital to build a thriving interdisciplinary academic dis-
curso.
yo
D
oh
w
norte
oh
a
d
mi
d
F
r
oh
metro
h
t
t
pag
:
/
/
d
i
r
mi
C
t
.
metro
i
t
.
mi
d
tu
/
j
i
/
norte
h
a
r
t
i
C
mi
–
pag
d
yo
F
/
/
/
/
4
7
2
1
9
3
1
7
0
6
5
9
7
/
j
i
norte
h
_
a
_
0
0
9
7
7
pag
d
.
F
b
y
gramo
tu
mi
s
t
t
oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3