by Joram Mayshar, Omer Moav, and Luigi Pascali
Why are some nations so much poorer than others? Two striking facts are revealed by the spatial distribution of economic activities since the Neolithic Revolution some ten thousand years ago. First, research has highlighted an exceptional persistence in the levels of economic, technological, and political development around the world until 1500 AD. Second, although the European colonization created a “reversal of fortune” among European colonies, once accounting for migration between countries, ancient economic disparities still explain a large portion of current economic disparities. Understanding the roots of global inequality therefore requires going back to the far past to answer two fundamental questions: What are the mechanisms that led to the development of complex states and sophisticated civilizations following the Neolithic transition? Why did it happen in some regions but not in others?
A key factor explaining the huge economic gaps between regions is the emergence of states that have the capacity to tax and provide the basic public goods of defence and law and order in some regions but not in others. Ever since Adam Smith and Karl Marx, scholars have attributed the emergence of hierarchy and states to one facet of agriculture: the surplus that was created from increased productivity. This surplus, it is argued, was a prerequisite for the rise of social hierarchy: an elite and a class of specialized bureaucrats that did not engage in food production for their own subsistence. This eventually led to the emergence of states and the great civilizations of antiquity.
In a recent research, we argue, in contrast, that surplus was neither a necessary nor a sufficient precondition for the emergence of hierarchy and was anyway unlikely to emerge following an increase in productivity. To understand why, consider a community of farmers with annual output above subsistence, who cultivate cassava (also known as manioc or yuca), a perennial root that can be harvested year-round but is highly perishable upon harvest. Since the crop isn’t stored and rots shortly after harvest, it is difficult to appropriate (i.e., to tax or to expropriate). The available surplus in this case in unlikely to facilitate the emergence of hierarchy. Consider now another farming community growing a cereal grain, such as wheat, rice, or maize, with no surplus, where each family’s annual produce equals its subsistence needs. Since the grain must be harvested within a short period and then stored, a visiting tax collector could readily confiscate part of the stored produce. Such ongoing confiscation may be expected to reduce population size; but this scenario demonstrates that surplus isn’t a necessary precondition for taxation by the elite. Finally, according to the Malthusian theory, increased productivity resulting from the transition to farming leads to increased population that eliminates any food surplus.
We propose that the crucial element for understanding why the transition to agriculture only enabled the emergence of significant hierarchies and states in some regions of the world and not in others is the farming of cereal grains that are highly appropriable, unlike less appropriable starchy crops such as roots and tubers. Thus, the non-emergence of complex hierarchy among hunter-gatherers was not because they live at subsistence, but because hunter-gatherers typically live hand-to-mouth, with little that can be expropriated to feed a would-be elite.
Utilizing multiple data sets spanning several millennia, we show how geography, through its effect on the type of crop cultivated, can explain differences in hierarchy. We further show that, in contrast to the conventional productivity-and-surplus theory, land productivity per se has no direct effect on hierarchy. We present these results by employing various datasets with information on social hierarchy: a cross section of societies, a panel of countries, and carbon-date archaeological evidence.
For our cross-section analysis, we use Murdock’s Ethnographic Atlas, which contains information on cultural, institutional, and economic features of 1,267 pre-colonial societies from around the world. Our main outcome variable is “Jurisdictional Hierarchy beyond the Local Community.” The Ethnographic Atlas also provides information on the major crop type grown by societies that practice agriculture. We measure the productivity of different crops using data provided by the Food and Agriculture Organization (FAO) on yields predictions. Consistent with our theory, the data show that when cereals are sufficiently more productive than roots and tubers, societies grow cereals and tend to have a more complex hierarchal organization. Moreover, we find that land productivity does not affect hierarchy once we control for the productivity advantage of cereals. Thus, we challenge the standard argument that it is an increase in productivity of land that led to the emergence of complex hierarchy and states. Moreover, we find that societies that practice agriculture are more hierarchical only where they cultivate cereals. This means that societies that cultivate roots or tubers have a similar level of hierarchy to that of non-farming, pastoral or foraging, societies.
Although the analysis accounts for a wide range of confounding factors, we cannot rule out completely that omitted variables may bias our cross-section analysis. To overcome this concern, we employ a panel dataset compiled by Borcan, Olsson and Putterman, based on present-day boundaries of 159 countries, with institutional information every five decades over the last millennium. With these data we exploit the “Columbian exchange” of crops between the New and the Old World as a natural experiment. The crops that became available following the transfer changed both the productivity of land and the productivity advantage of cereals over roots and tubers in the majority of the countries in our sample. Consistent with our theory, the panel regressions confirm that an increase in the productivity advantage of cereals over roots and tubers has a positive impact on hierarchical complexity, and an increase in land productivity does not.
The third empirical exercise gets us closer to the Neolithic transition and is based on cross-sectional data from various sources on the location of ancient cities and archaeological sites such as pyramids, ancient temples, palaces, and mines, which presumably indicate social hierarchy. Since modern agricultural yield predictions are presumably less accurate for crops several millennia ago, we construct two new datasets to measure the availability of different crops at the onset of the Neolithic transition. The first dataset covers the location of centres of first domestication of each relevant agricultural crop; the second captures the distribution of wild relatives (WRs) of domesticated crops (wild plants that are genetically related to cultivated crops). The number of WRs of a certain domesticated crop in a region proxies for the potential for domestication of that crop in that region. Using these data, we document two findings.
First, distance from the centres of independent domestication has a negative impact on the development of early civilization only if the centre domesticated cereals. The top figure below illustrates this result. It reports the cities founded before 500 BC: as can be seen these are clearly located around areas of cereal domestication. Second, the availability of WRs of domesticated cereal grains and lack of availability of WRs of domesticated roots or tubers explain a significant portion of the variation in the different indications of hierarchy in all the archaeological data sets, for data spanning various periods of antiquity (see the bottom figure). Although results are robust to many confounders, a limit of this data set is that it is cross-sectional.
We partially overcome this limitation in our fourth empirical exercise, in which we use data from the Archaeological Atlas of the World. Although this source was published more than 40 years ago, it has the advantage of providing radiocarbon estimates dating various archaeological sites, enabling us to count the number of pre-Neolithic and post-Neolithic sites in each area. The difference-in-difference estimates support the appropriability theory. Specifically, we find that the Neolithic transition only led to more traces of indications for complex hierarchical societies in areas where agriculture was more likely to start with cereals, based on our two proxies explained above (WRs of domesticated crops and proximity to areas of domestication). We find no evidence for the conventional productivity theory using these data.
In summary, our empirical analysis provides repeated evidence that the cultivation of cereals had a significant causal effect on the development of complex hierarchies and states, consistent with the appropriability theory. It also illustrates that the correlation between land productivity and hierarchy disappears after controlling for the cultivation of cereals, consistent with our critique of the conventional productivity-and-surplus theory. Moreover, the finding that it is unlikely that complex hierarchies would emerge when productive roots and tubers are available supports both the appropriability theory and the critique of the productivity theory. It is consistent with the prediction that farmers would, ceteris paribus, prefer to cultivate less appropriable crops that provide protection from bandits and tax collectors. Complex hierarchies, the data suggest, emerged when farmers were constrained to cultivating cereals. The availability of highly productive roots and tuber is a curse of plenty. In these regions, functioning states did not emerge and low income persisted.