Stefan Dercon has a deal for you. Well, not for you, personally, but for the elites of the countries of the world that have not yet set themselves on the path out of poverty. The deal is a gamble: risk the comfortable equilibrium where the elite maintain power through clientelism and patronage, in exchange for a chance at widespread prosperity. What explains, according to Dercon, why some countries remain in desperate poverty is not whether a country has undertaken a particular slate of reforms (either liberalizations or interventions), but whether elites are willing to make the gamble on development, see it through, and are quick enough on their feet to make adjustments.

Dercon begins his book by outlining four broad models of the failure to develop (a lack of endowments, general market failures, increasing returns-style market failures, and bad governance), then discusses the views of many development economists in the context of those broad models: Jeffrey Sachs, Abhijit Banerjee and Esther Duflo, William Easterly, Angus Deaton, Dambisa Moyo, Joseph Stiglitz, Ha-Joon Chang, Paul Collier, and Daron Acemoglu and James Robinson.

Dercon sees the value in incorporating all these perspectives on how development works, although he casts shade at those most skeptical of state involvement, characterizing Easterly, Deaton, and Moyo as purveyors of “doom and gloom” (p. 24), and saving his sharpest jabs for Moyo. He sarcastically summarizes her as “Let’s just trust the market, especially global capital markets, the former Goldman Sachs advisor goes on to suggest,” then noting that Bill Gates has portrayed Moyo’s book as “promoting evil” and that her position “relies on some pretty dodgy evidence” (p. 25).

Subsequently, after describing his general hypothesis in the next chapter, and a perfunctory chapter on “yes, sustained economic development throughout the world is, in fact, happening,” Dercon then enters the second part of the book, where he identifies seven groups of countries which illustrate his position. These descriptive histories are very useful both in terms of seeing Dercon’s view from a number of vantage points, and as simply informative of the stories they tell. He riffs on the description of successful East Asian countries as “Asian Tigers” by titling chapters with a “menagerie of animal metaphors” (p. 88).

Dercon begins with the “dragon” (China), “tiger cub” (Indonesia), and “peacock” (India), as they represent such a large proportion of the world’s population. He uses them to make the well-worn point that the three differ greatly in terms of their histories and the nature of their states, despite their common development success, and then claims that what unites them in their success is their elite commitment to development, rather than whether markets or states are what is used to move development forward. The chapter is quickly followed up by a discussion of Sierra Leone and Malawi, two countries that are not quite disasters, but have an elite that is content in its equilibrium of economic stagnation and poverty.

The following five chapters each similarly hit a single theme: Kenya, Uganda, and Ghana remain impoverished but seem set for a takeoff; Nigeria and Democratic Republic of the Congo suffer from the resource curse; South Sudan, Afghanistan, Nepal, Lebanon, and Somaliland have dealt with, or currently deal with, conflict; Bangladesh succeeds due to elite commitment even though it started out in a very weak position; and Ethiopia and Rwanda are positioned to become the “African Lions.”

The last part of the book concerns how to effect change, with a focus on development aid, which is the mechanism Dercon is most experienced in. His position is roughly an amalgam of the views of the development economists listed above: aid is often wasted or even harmful, but that it is a policy mechanism that can be improved upon, and is ultimately a beneficial force for the world–especially if it can be used to change the nature of the elite bargain. While both historical experience and the wisdom of political economy imply the potential for aid to crush the freedom (and prosperity) of economies (Bauer, 2000; Coyne, 2013; Easterly, 2013), there may not actually be that much daylight between the position of Dercon and the findings from recent high-quality empirical studies on the effect of aid on economic freedom (Dutta & Williamson, 2016; Bologna Pavlik et al., 2022), which find somewhere between mixed, null, and maybe slightly positive results. Dercon’s views here are worth digesting.

I had a few points of concern as I read the book. Most substantively, the elite bargains described by Dercon bear a strikingly similarity to the political bargains described by Douglass North and his coauthors in Violence and Social Orders (North et al., 2009; from here on will be referred to as NWW). NWW describe societies as either “limited access orders” or “open access orders,” where the former build a state for the benefit of the elite, and in the latter, the structures of the state are used to benefit all. NWW describes at length how institutions function in these two societies and how the transition takes place from limited access orders to the open access orders. Dercon briefly acknowledges NWW but he chooses to contrast his approach with theirs, describing NWW as “considering a different question–the conditions for long-term growth over decades, if not centuries, and not, as in this book, the conditions for take-off growth and progress in fighting the most extreme levels of poverty” (p. 45).

But that is to overlook NWW’s lengthy discussion of the development of limited access orders. Under limited access orders, there is a “natural state” for the elite, which comes in three varieties, fragile natural states, basic natural states, and mature natural states (2009, p. 41–49). Fragile natural states are those rife with conflict and teeter on non-existence; graduating from a fragile natural state to a basic natural state is an implicit theme peppered in the background of Dercon in his discussion of several countries, especially Sierra Leone, Democratic Republic of the Congo, and of course the five countries in the chapter on conflict.

Mature natural states differ from basic natural states in terms of their capabilities to perform various functions; their institutions are able to do more than simply controlling violence and setting the rules for rent extraction. Per NWW, institutions must be built out on behalf of the elite (e.g., the development of the rule of law for the elite, or the creation of depersonalized, lasting bureaucratic structures). Dercon’s chief concerns appear to be about whether the elite are content to maintain power using the structures of a basic natural state, sustained by patronage and clientelism, or if they are willing to risk disrupting their current political equilibrium by further building out the institutional structures of the state. If they succeed, they will have achieved a mature natural state. Per NWW, the mature natural state may or may not become an open access order should the elite see it in their interest to grant the services and protections of the state they built for the benefit of all.

Where Dercon operates is therefore describing a particular stage of institutional development in NWW. NWW was not asking a different question, but a broader question. Dercon, when read in terms of NWW, is providing value by either filling in important details over how institutional development occurs, or by offering a different account of how countries at this stage of development begin to break free from poverty.

Dercon may or may not accept this reading. He views the work of Acemoglu and Robinson (2012) as a “sobering” read, suggesting that the implication of the “institutions matter” hypothesis is that if your history did not produce the preconditions for good institutions, then you are out of luck (p. 28). But although the development of institutions is often framed this way, the empirics of institutional change show that institutions can change rapidly and meaningfully at times (Lawson & Lawson, 2020; Murphy forthcoming) – and most (though not all) of what Dercon is describing is getting the elite committed to building institutions.

The second point I wish to raise is whether it is really reasonable to follow Dercon’s view that the approaches that the successful countries took were sufficiently diverse to dismiss any particular recipe as a means of escaping poverty. In Table 1, I put data to the question. I took the nine countries that Dercon describes as successes, or soon to be successes (China, Indonesia, India, Kenya, Uganda, Ghana, Bangladesh, Ethiopia, and Rwanda) and reported their scores in the Economic Freedom of the World index (Gwartney et al., 2022) in the first year they are reportedFootnote 1 in the data set, and then also in 2019 (to set aside COVID-19-specific changes). However, to emphasize how the world has liberalized overall, I take the mean and standard deviation from the 2019 data, and report the EFW data in terms of percentile in its initial year as if it were a country in 2019. For example, Rwanda’s score in 1970, its first year, is given a percentile from the distribution of countries in 2019. The percentiles are multiplied by one hundred to promote legibility.

Table 1 Economic freedom scores over time of development successes identified by Dercon (2022)
Full size table

In all cases, except for perhaps one, we observe significant liberalization. Note that the data which reads “0.01,” does not mean 1st lowest percentile—it means that, if there were 10,000 countries scored in 2019, a country receiving as score of “0.01” would be expected to be in dead last. This result is in part due to the normal distribution being imposed on data with a thick left tail, but it is owing more to the fact that so many countries liberalized over the period and shifted the mean. Four of the nine countries, Indonesia, Kenya, Rwanda, and Uganda, went from having (percentile) scores that ranged from 0.01 to 1.50 to an above average score (i.e., over 50.0), and another two (Ghana and India) improve to over the 40th percentile. That plus convergence yields growth, in all likelihood a lot of growth.

The country that has the strongest claim to not having its institutions fundamentally change is Ethiopia. Ethiopia lacks sufficient data to be scored before 2005, and in 2005 it rates as merely “very bad” (a quasi-percentile score of 4.98). It is possible (or even likely) that if we had data going further back, Ethiopia would be scored similarly to the others during the earlier periods, i.e., extremely low scores. But as of 2019, it holds the lowest economic freedom score among the nine. This fits with the narratives told in the text of Gambling on Development, as Ethiopia is the country that Dercon perhaps most emphasizes as exhibiting the characteristics of state-led development, rather than liberalization (pp. 233–252). So, does Ethiopia constitute the counterexample to liberalization? Well, maybe not quite. As Dercon was writing the book, Ethiopia was in the midst of a civil war and the possibility of a famine as its governance structures broke down, with Dercon expressing the belief that Ethiopia’s experience with rapid economic growth had probably come to an end (p. 250).

There is a diversity of the approaches of the nine (or eight) successful countries that Dercon lists, but they all, aside from Ethiopia, liberalized significantly. If Dercon is correct more broadly, perhaps it’s necessary for the elite commit to gambling on development, but in nearly all cases, that is going to mean an awful lot of liberalization, even if it is not the ideological focus of the elite. It may well be the case that elite commitment is an important driver of liberalization. The beliefs of elites are a topic studied on occasion in the literature on the causes of economic freedom (Dreher et al., 2009; Li et al., 2020; Murphy & O’Reilly, 2020), but it is hardly a pillar of the broader literature (Lawson et al., 2020). Maybe it should be a pillar, though this is not quite the direction literature has gone in recently (see, e.g., Levy & Peart, 2017; Holcombe, 2018).

A perhaps more “positivist” concern is that it’s a bit hard imagine how we could ever assess the hypothesis of Gambling on Development as we would assess more conventional explanations of development, such as the importance of institutions or education. Even in the case of NWW, the characteristics of open access orders were described in enough detail that we can now measure them (Murphy forthcoming), but it’s hard to imagine what an empirical counterpart to “we think the elites are taking development seriously this time” could be, short of the World Bank or some other group publishing some kind of rating. (The World Bank’s Country Policy and Institutional Assessment may be the closest we have to this right now, but it really isn’t supposed to be thought of like that.) To state this somewhat facetiously, the best way I can think of whether Dercon is actually right is to add a variable to a country-level data set that is set to one for Bangladesh, China, Ethiopia, Ghana, India, Indonesia, Kenya, Rwanda, and Uganda, and zero for everyone else, and then check back in ten (of fifty) years on the regression results.

Do I recommend Gambling on Development? Yes, the positives outweigh the limitations described above, if the “limitations” should even be called that. If used in the classroom, its overview of different development thinkers is rather complete and digestible in a manner I am not sure I have encountered elsewhere, and its narratives on the challenges faced by developing countries is informative, comparing favorably to say, the cumbersome Acemoglu and Robinson (2019). From the standpoint of political economy, it does rather well among scholarship outside the public choice tradition in avoiding romanticized politics, although, as I have said earlier regarding it and NWW, it situates itself poorly within modern political economy. While I believe Dercon underplays the case for liberalization as the primary vehicle for taking the gamble on development, his descriptions of how the process of development plays out practically plausibly fills in the details left unenumerated by NWW, and as such, Gambling on Development is a potentially important contribution to the scholarly literature.