Published by EH.Net (June 2024).

Leandro Prados de la Escosura. Human Development and the Path to Freedom: 1870 to the Present. New Approaches to Economic and Social History. Cambridge: Cambridge University Press, 2022. 250 pp. $29.99 (paperback), ISBN 978-1108708586.

Reviewed for EH.Net by Sakari Saaritsa, University of Helsinki.


Leandro Prados de la Escosura has done a service to the economic and social history community by pulling together more than a decade of his leading work on the application of the Human Development Index (HDI) to historical data globally. Prados, Professor of Economic History at Universidad Carlos III, Madrid, has been one of the most prolific authors on historical HDIs since the 2010s. Compiling – and altruistically sharing – a vast amount of global data, he has meticulously analyzed divergence and convergence patterns between the West and the Rest, as well as the contributions of the different components of the index to its gradual advancement in different parts of the world and over different periods.

Prominent reviews and lively debate about this book and the HDI in general have already been published by economic historians (see, for example, issue 1 / 2023 of Rivista di storia economica / Italian Review of Economic History, with a critical review by Jan Luiten van Zanden followed by an extended exchange with Prados and further comment by Nicola Amendola, Giacomo Gabutti, and Giovanni Vecchi). The aim here is to describe what is in the book, including its global empirical chapters, while first touching upon some pertinent theoretical points.

The book contains six chapters and is divided into two parts. The first part, “An Aggregate View,” introduces Prados’s preferred Augmented Human Development Index (AHDI) and its construction, and then discusses global trends, inequalities, and uneven evolution of human development over the last 150 years on its basis. The second part, entitled “The OECD and the Rest”, takes a regional approach, first looking at the gaps and catch-up dynamics between the enriched OECD club and the less fortunate rest of the world, and then at Latin America and Africa. The choice of the latter as the current ‘laggard’ is obvious (in the 1960s, it could have been Asia). Latin America is selected for closely matching the world average. In addition to a plethora of data in figures and exhaustive tables, the empirical chapters always contain concise high-quality overviews of related research literature and debates on each theme or region, from origins to recent contributions. This greatly augments the pedagogical usefulness of the book.

As a tool for analysis, the HDI, now approaching the sage age of 35, has simultaneously been intellectually stimulating, widely applied, and open to fundamental criticism over ultimate usefulness and transparency. Building on Amartya Sen’s capabilities approach, the HDI aimed to capture core dimensions of the human condition that are necessary for the positive freedom to independently shape one’s life: income (GDP), knowledge (education and literacy) and a long healthy life (life expectancy). While the key motivation for its launch in 1990 was de-centering economic growth as a metric of development and ranking countries based on achievements more dependent on public goods, it was not intended to be a paternalistic social welfare function. The prerequisites of a market economy and political freedom for it to be meaningful were explicitly spelled out in the first UNDP Human Development Report (HDR 1990, p. 1).

The HDI has been subject to various revisions, both by the UNDP between the HDRs and in the hands of scholars, including economic historians. The first victim of this stimulating R&D work has been comparability. The pioneering work of Nicholas Crafts (1997), for instance, was based on the HDI formula in effect from 1995 to 1999, already revised from 1990, using formulations by Tony Atkinson for transforming income. Crafts’s key observation was that currently rich countries scored historically lower than present-day developing countries, mainly due to low life expectancy and education. In their innovative work looking at regional development and the questione meridionale of the South in Italy using the HDI, Emmanuele Felice and Michelangelo Vasta adopted what is called the “hybrid” HDI or HHDI (Felice and Vasta 2015). A major UNDP overhaul of the HDI in 2010 changed the parameters of the equation, introducing the use of the geometric mean and requiring metrics rarely available in historical data such as schooling expectancy and the GNI (HDR 2010). Therefore, UNDP researchers formulated the HHDI for longitudinal studies, with old inputs preserved within the new equation. The research and data by Prados contain different formulations again. Few of these works are directly comparable. The computational choices involved – especially, as recently argued by Amendola et al. (2023), the ones over the marginal rate of substitution (MRS) of the dimensions – can generate completely different narratives of development, at least at the level of decades and countries. As the MRS ultimately depends on the ethical valuations of the analyst, this smuggles paternalism in through the back door.

Some key theoretical choices made by Prados thus merit scrutiny. Firstly, there is the whole issue of whether the HDI should be considered strictly an ordinal measure, good for ranking countries but with no sensible interpretation for distance, relative growth, inequality, or the like. This is because it reads like a utility function, only capable of ranking bundles of goods. This strict interpretation has for long been advocated by some, e.g., Giovanni Vecchi (Vecchi 2017, pp. 454-491), as well as Anand and Sen, who stated flatly in 1992: “But the human development index in the 1990 (and subsequent) reports was constructed expressly as a measure of relative performance across countries at a point in time. No special significance is attached to the absolute value of the index, the entire analysis being conducted in terms of the ranking of countries relative to one another.” (Anand and Sen 1992, pp. 8-9; HDR 1993, p. 110.) Had this stance been adopted, this book would probably not have been written.

Secondly, Prados makes the case for a version of the HDI where, following Nanak Kakwani, the health and education indices are transformed using logs of distance from maxima so that increases in the components from higher initial levels lead to larger increases in the HDI than increases from lower levels. The argument is based on an understanding of the HDI as a measure of the performance and achievement of societies (or perhaps polities and their rulers): increasing life expectancy or education from an already high level is more difficult and should therefore be seen as more impressive. The same has been suggested by Sen (1981, p. 292). Prados also points to the existence of theoretical or empirical maximum limits for education and life expectancy but not for income, which calls for levelling of the playing field through the convex modification of the former.

Since this key transformation runs through the results and the data of the book, it’s worth pondering critically. Is it fully commensurable with the idea of the HDI as measuring the expansion of the positive freedom of individuals, rather than, e.g., state capacity? It also has been asked whether learning to read should not be considered more significant to the expansion of peoples’ capabilities than getting a PhD (Vecchi 2017, p. 468; Noorbakhsh 1998, p. 519). When discussing life expectancy, Prados points out that using relative improvements in levels would grant larger HDI changes for poor countries for reducing infant mortality than for rich countries for advancing the longevity of their oldest cohorts, which would “arbitrarily give more weight to saving the lives of younger people than the lives of older people” (p. 16). However, such weighting might be seen entirely justified if more life years were thus spared, and would intuitively match the preferences of many, if not most, societies. It is also not entirely obvious even from a performance perspective that rolling out basic sanitation or primary schooling for the first time in an LDC is easier than adopting novel cancer treatments in some of the richest countries in the world.

Downgrading the impact of income is a canonical procedure, even if some of the arguments in favor are not entirely obvious. It is stated that while education and health should be seen as capabilities in themselves, income “simply represents an input that can be turned into a capability: being able to live a full, meaningful life”, and its returns are expected to diminish at higher levels, unlike for the other dimensions (p. 14). But as Friedrich Hayek once put it, “If we strive for money it is because it offers us the widest choice in enjoying the fruits of our efforts” (2006 (1944), pp. 92-93). Both in some of the richest and the poorer countries of the world, this choice nowadays also crucially includes being able to buy better education and health care for one’s children in an environment that has become increasingly marketized and unequal since the 1990s. In European welfare states, health and education are still to varying degrees effectively off the market.

Prados’s AHDI explicitly incorporates a metric for freedom, which is what would separate the HDI from a “basic needs” index where individual choice is not a consideration. The measure selected, the Liberal Democracy Index, aspires to account for a broad array of political and civil liberties, such as electoral democracy and an independent judiciary, but apparently not economic freedom as described in the first HDR: “People must be free to exercise their choices in properly functioning markets” (HDR 1990, p. 1). In full, the AHDI then consists of the geometric mean of the log transformed concave income index (log GDP per capita), the Kakwani transformed convex health (life expectancy at birth) and knowledge (years of schooling for population 15 or older) indices, and the Liberal Democracy Index, not subject to similar transformations. Equal weights are given to all, the default HDI setting known to conform to long-run Principal Component Analysis for the three original dimensions (HDR 1993, pp. 109-110), but not grounded in the actual choices of individuals (Ravallion 2011). In concrete historical situations, individuals might trade, for instance, risk of death against expected income when attempting to cross the Mediterranean into Europe on leaky boats to circumvent immigration controls; or freedom against promises of social mobility when supporting movements that promise to prioritize children of “workers and peasants” in higher education but would strip everyone of meaningful political rights.

The empirical findings of the book are presented via exhaustive, and occasionally exhausting, tables, graphs and descriptions of trends and ratios. In the aggregate, the AHDI shows notable progress, but with specific periods of acceleration and deceleration. As usual in HDI-based analyses, life expectancy and schooling are what make the biggest contributions to human development over time, not (concaved) income. Major strides in these dimensions were based on global processes of diffusion of knowledge and institutional patterns which were relatively independent of economic growth or political regimes. For life expectancy, what mattered in the first stage was an epidemiological transition underpinned by the practical adoption of public health policies informed by germ theory, but preceding biomedical remedies like vaccines or antibiotics (cf. Saaritsa, Simanainen, and Ristola 2023). This process, which accelerated from the 1920s also in many parts of the world under colonial regimes, required particularly local government capacity and has been seen closely connected to the spread of primary schooling (Riley 2008). Based on a literature review, Prados associates the spread of education closely with nation building in a broad sense and the global ascendancy of the centralized state, requiring mass socialization through schools regardless of the political flavor or degrees of democracy involved. Nation building in 19th century Europe, or post-independence state activism in autocratic former colonies from the 1960s, where majorities had often been completely excluded from public mass schooling, could underpin similar leaps in the knowledge index.

An important observation in the book is the lack of evidence of any consistent virtuous circles between the elements of the index, particularly at the decadal and regional level of analysis. Authoritarian regimes have carried forward global trends in health and schooling, and significant progress in the other dimensions of human development have taken place during notorious periods of economic stagnation, like in Europe in the 1930s and Latin America in the 1980s. Trends in political freedom followed known waves of authoritarianism and democratization, including the rise and spread of Fascist and Communist governments in early 20th century and postwar Europe, the emergence of often initially one-party postcolonial states in Africa, and the diffusion, connected to the Cold War and aided by foreign intervention, of right-wing dictatorships in Latin America.

In terms of global AHDI inequality, Prados demonstrates key differences in the evolution of inequalities (cross-country “inequality 1”, or population-weighted “inequality 2”) in relative and absolute terms and between the different components, and by looking at the incidence of growth in different parts of their distributions. Inequalities in health and education peaked and started to decline much earlier than income inequality, which kept growing until the late 20th century. These trends were driven by the aforementioned processes of diffusion of institutional innovations and state capacity from the 1920s. Most of the increases in life expectancy and schooling accrued to the middle and the bottom of the distribution, whereas those in economic growth and political freedom benefited the top as well as the middle – with the global “human development middle class” thus benefiting most consistently. Measured in absolute terms, however, inequalities have for the most part constantly grown, not decreased.

Looking at OECD and “the Rest”, Prados discusses dispersion within and between the groups, drivers of development in the main regions, and the characteristics of catching up. Inequality between these two groups drove global AHDI inequality until the 1980s, when within-group inequality took over. Between 1910s and 2010s, the ratio of the AHDI score of the “Rest” to the OECD rose to about one half from about a quarter. The non-income elements were more dominant in the “Rest”, with the exception of the immediate postwar decades, and the period from the 1920s to the 1980s was generally one of major progress in health and education. While life expectancy was typically a prominent driver of development, all regions had somewhat different stories, with e.g., schooling more important for AHDI progress in East Asia and liberties holding it back in North Africa. The second health transition increasing the longevity of the older cohorts in the rich countries but not yet in the “Rest” has driven a new divergence in life expectancy. Detailed charts of the relative contribution of each dimension in catching up by region are presented, providing plenty of insights beyond already familiar narratives. The regional chapters also drive home how little can be understood from the indicators alone without being able to complement the trends with knowledge on the historical contexts. Descriptions of metrics going up or down by decade or region, of which there are plenty on the pages of the book, would remain empty signifiers without headnotes on what was happening at the time.

The chapters on Latin America and Africa both contain rich, comprehensive country level analyses as well as overall regional diagnostics. In Latin America, persistent income inequality was combined with strides in human development also in bleak periods like the 1980s, with life expectancy in the lead while educational systems tended to start top heavy and elitist like in historical Europe, and a deterioration in freedoms in the 1960s and 1970s, more understandable in the context of the Cold War. Countries had their specific trajectories, and looking at cases like Cuba, Jamaica or Chile, GDP, health, schooling, and liberties often had particularly low correlation in Latin America, even as some of the dimensions forged ahead.

The chapter on Africa is valuable for once again codifying relevant literature as well as for providing an analytical perspective beyond unwarranted pessimism or optimism through the prism of the AHDI, together with subregional and national analysis to the extent allowed by the data. In general, schooling was key for the continent’s gradually advancing score, particularly from the 1920s to the 1990s, while performance in terms of life expectancy was patchier, with acceleration from the 1920s before decolonization until the 1980s, a slowdown in the 1950s, and deceleration and stagnation in the 1990s, followed by a rebound in this millennium. Liberties advanced immediately after independence but took a beating with the emergence of autocratic dispensations since the 1960s, to be reversed again after the Cold War was over. In terms of income, there has been little catch-up to report despite of absolute growth. The reference to “syndromes” (p. 217) drawing from Collier and O’Connell, meaning ‘salient episodes of purposive failure attributable to human agency within the society’ such as regulatory and redistributive failures and excesses as well as state breakdowns, is illuminating in the case of Africa. It would probably be so in many instances in historical Latin America or Europe as well. Violence and state collapse in Africa were, as in the case of Latin America, often a function of the Cold War turned hot by foreign proxies.

An interesting point from the perspective of possible dynamic effects from improved HDI raised by Prados on both Latin America and Africa is that looking at height data, it seems that reductions in infant mortality were not necessarily accompanied by improvements in cohort health in poor countries. This could potentially foster generations that survived infancy only to be scarred by weak nutrition and health, dampening productivity and future growth.

All in all, the book demonstrates how useful the HDI, particularly when decomposed, can be for at least presenting a broad and harmonized picture of the development of core dimensions of human well-being over time and in different regions. Together with the incisive summaries of state-of-the-art research on aspects of human development and rich regional and national analyses, it makes for a handy global comparative reference of essential characteristics of development beyond the GDP in the last 150 years for economic and social historians. Although the fundamental dilemmas of the HDI remain unresolved, this book will benefit scholars and students for years to come.


Amendola, Nicola, Giacomo Gabbuti, and Giovanni Vecchi. “On Some Problems of Using the Human Development Index in Economic History.” European Review of Economic History 27, no. 4 (November 2023): 477–505.

Anand, Sudhir, and Amartya Sen. “Human Development Index: Methodology and Measurement.” Human Development Report Office Occasional Papers. UNDP Human Development Report Office, 1992.

Crafts, N. F. R. “The Human Development Index and Changes in Standards of Living: Some Historical Comparisons.” European Review of Economic History 1, no. 3 (1997): 299–322.

Felice, Emanuele, and Michelangelo Vasta. “Passive Modernization? The New Human Development Index and Its Components in Italy’s Regions (1871-2007).” European Review of Economic History 19, no. 1 (2015): 44–66.

Hayek, Friedrich A. von. The Road to Serfdom. Reprint. Routledge Classics. London: Routledge, 2006.

Noorbakhsh, Farhad. “A Modified Human Development Index.” World Development 26, no. 3 (March 1998): 517–28.

Ravallion, Martin. “On Multidimensional Indices of Poverty.” The Journal of Economic Inequality 9, no. 2 (June 2011): 235–48.

Riley, James C. Low Income, Social Growth, and Good Health: A History of Twelve Countries. California/Milbank Books on Health and the Public. Berkeley: University of California Press and Milbank Memorial Fund, 2008.

Saaritsa, Sakari, Eero Simanainen, and Markus Ristola. “Nurses, Doctors, and Mortality: The Effectiveness of Early Health Professionals in Rural Finland, 1880–1938.” European Review of Economic History 28, no. 1 (February 2024): 91–119.

Sen, Amartya. “Public Action and the Quality of Life in Developing Countries*.” Oxford Bulletin of Economics and Statistics 43, no. 4 (1981): 287–319.

United Nations, ed. Human Development Report 1990. New York: Oxford University Press, 1990.

United Nations Development Programme (UNDP), ed. Human Development Report 1993. New York: Oxford University Press, 1993.

United Nations Development Programme (UNDP), ed. Human Development Report 2010: 20th Anniversary Edition: The Real Wealth of Nations: Pathways to Human Development. Houndmills: Palgrave Macmillan, 2010.

Vecchi, Giovanni. Measuring Wellbeing: A History of Italian Living Standards. New York: Oxford University Press, 2017.


Sakari Saaritsa is Professor of Social History, Faculty of Social Sciences, University of Helsinki. His research focuses on the history of human development, specifically health, education, gender, and historical demography.

Copyright (c) 2024 by EH.Net. All rights reserved. This work may be copied for non-profit educational uses if proper credit is given to the author and the list. For other permission, please contact the EH.Net Administrator ( Published by EH.Net (June 2024). All EH.Net reviews are archived at