by Ober, Josiah
35. Bresson 2000, 2007. Horden and Purcell 2000 characterize ancient Mediterranean trade as defined by intricate networks of interdependent regional exhange, based on a multiplicity of microenvironments.
36. Bloom, Canning, and Fink 2008.
37. Population growth leads to substantially lower living standards in most of the cities of Europe in 1500–1800: Allen 2001; Deaton 2013: 94–95. Disease: Scheidel 2007. Squalid conditions in the advanced economies of England and Holland: Kron forthcoming. Greek life expectancy: Morris 2004: 714–720, Kron 2005; von Reden 2007: 388–390. Improved life expectancy for those surviving childhood: Morris 2004: 715, Fig. 2.
38. On the question of life expectancy at birth, see Morris 2004: 7; Scheidel 2009a. Kron 2005, 2012 argues for longer life expectancies at birth for Roman and especially Greek populations than have other investigators. Deaton 2013: 82–83, drawing on exceptionally fine-grained data, shows that disease, rather than nutrition, was the primary limit to life expectancy in England in 1550–1850.
39. See, by way of contrast, Mayne and Murray 2002, on the archaeology of modern slums; Scobie 1986 on Roman slums.
40. Scheidel and Friesen 2009: 72–73 discuss the literature on egalitarian distribution of income and economic growth. Milanovic, Lindert, and Williamson (2011) examine inequality in 28 premodern societies (not classical Greece) on the basis of the the “extraction frontier,” thus measuring inequality by how close elites in a given society approached the feasible maximum of resource extraction, beyond which the non-elite would perish.
41. Morris 2004: 722–723.
42. Kron 2014, forthcoming. For Olynthos houses, Kron 2014: 129, table 2 estimates the Gini coefficient of inequality at 0.14, considerably lower (i.e., more equal) than later Hellenistic and Roman era Greek cities.
43. Cost of houses: Morris 2004: 723.
44. Kron 2011, 2014. Calculation is based on Athens citizen male population of ca. 31,000. Of these, 9,000 met the 2,000 drachma census in 322 BCE. Cost of a typical large Greek house: ca. 1,000–2,000 dr; house prices in inscriptions range as low as 200 dr.
45. Kron 2011, 2014. It is important to keep in mind that the overall Gini wealth index for Athenian society as a whole, including slaves and metics, would surely be substantially higher—I cannot say how much higher because I know no way to calculate wealth of metics or slaves. Wealth inequality, as measured by the Gini coefficient, is typically much higher than income inequality, which is considered below.
46. Foxhall 1992, 2002; Osborne 1992.
47. Morris 1998: 235–236. Quote, ibid. 236.
48. Scheidel and Friesen 2009: 84–85. By way of early modern comparisons, Milanovic, Lindert, and Williamson 2011 (Table 2) report the income Gini for Tuscany in 1427 = 0.46; Holland in 1561 = 0.56; England and Wales in 1688 = 0.45; France in 1788 = 0.56.
49. Scheidel 2010, figures p. 452; Scheidel and Friesen 2009. Of course, most people were not paid in wheat, and in many ancient societies they may not have eaten wheat as a staple. The method is similar to the converting of modern incomes from different countries and over time into, e.g., “1990 dollars per annum.”
50. Scheidel 2010: 454. Scheidel and Friesen 2009: 83 posit a bare subsistence minimum level of 335 kg of wheat equivalent per annum (i.e., 429 L wheat wage) per person in a tax-free environment, which translates to 1,716 L (4 × 429) for a family of four. This comes to 4.7 L/day/family. These figures are consistent with the calculations of Markle 1985, who argued that 3 obols/day (a juror’s pay in Athens = a wheat wage of ca. 4.5 L) was adequate to sustain a nuclear family at a subsistence level. Allen 2009 uses a somewhat different calculation (assuming a family as three rather than four “adult-consumption equivalents” and 250 rather than 365 wage-days per year), but he arrives at similar results for what he describes as a “bare bones” existence (2009: 340). In sum, an adult male wage earner at the “floor of the core” 3.5 L/day level might provide about two-thirds to three-quarters of his family’s minimum subsistence, meaning that women’s and children’s contributions to family income would be essential; see further, Scheidel 2010: 433–435, 454. Foxhall and Forbes 1982 offer a detailed examination of the role of grain in ancient diets—a key factor in any attempt to calculate actual subsistence minima.
51. Scheidel and Friesen 2009.
52. Scheidel and Friesen 2009: 71. Cf. Scheidel 2014, concluding, p. 191: “What little evidence we do have strongly suggests that Roman rule failed to deliver substantial benefits to workers in developed parts of the Mediterranean…. Roman economic development would not have differed greatly from that of most other pre-modern economies.” Allen 2009: 42–43 reaches similar results, based on the figures in Diocletian’s price edict of 301 CE. It is clear, however, that at least some regions of the Roman Empire in some eras of Roman rule performed well when compared to other ancient economies. The Roman economy is currently a very active research area; Scheidel’s conclusions are more pessimistic than those of some other scholars: see Bowman and Wilson 2009, 2011, 2013; Scheidel 2012a, 2012b; Temin 2013.
53. Scheidel and Friesen 2009: 74, 90.
54. Athenian figures cited and discussed by Scheidel 2010: 441–442, 455–456 are taken from epigraphic and literary sources collected and discussed in Loomis 1998: 111–113 and Markle 1985.
55. Scheidel 2010: 441–442, 453–458; Scheidel 2014: 188.
56. In Delos in the third century BCE, the wheat wage was 8 L/day—thus a multiplier of 2.3—just below the “middling” floor: Scheidel 2010: 442–443.
57. Half-day meetings: Hansen 1999: 136–137. Pay for assembly service: Hansen 1999: 150. Pay for other kinds of public service: Gabrielsen 1981; Hansen 1999: 314; Pritchard 2014.
58. Data summed up in Scheidel 2010; Allen 2009.
59. Optimistic and pessimistic estimates: Scheidel and Friesen 2009: 84–87. I focus on the late fourth century population because Hansen’s (2006b, 2008) “shotgun method” demographic model focuses on that period. Moreover, Athens did not have an empire in the late fourth century and was thus, in this way at least, more similar to other large poleis. For detailed discussion of fourth century Athenian demography, see Hansen 1986a, 1988, 2006c.
60. Definition: Davies 1971: xx–xxiv.
61. The limitation on rents and taxes exacted from farmers in Athens is emphasized by Wood 1988.
62. On Greek labor markets and interpolis movement of workers, see Davies 2007. My thanks to Barry Weingast for discussion of the labor market problem.
63. See, further, Whitehead 1977 (on Athenian metics), 1984 (on foreign residents in other Greek poleis, arguing that most poleis did have long-term foreign populations and that the conditions of their residency were formalized); McKechnie 1989.
64. Correlation: Bloom, Canning, and Fink 2008.
65. Roman growth rate: Saller 2005.
66. The estimates in Figure 4.3 are quite different from those suggested for Greece from 1—1900 CE by Angus Maddison (http://www.ggdc.net/maddison/Maddison.htm: accessed 2014.05.04). Maddison estimates long term historical population, total GDP, and GDP per capita for many countries, including Greece. His estimates for the population of Greece in 1000, 1500, 1700, 1800, 1900 seem to me much too high for “core Greece,” based on the archaeological and tax record information cited in Bintliff 2012 and on actual census data from the nineteenth century (it is unclear what geographic area Maddison’s Greece data is meant to cover). Maddison’s Greek GDP per capita estimates for ca. 1500 to the mid-twentieth century also appear too high, and the basis for his early GDP estimates is mysterious. The standard CNTS data (Banks and Wilson 2013) for per Greek capita GDP begins only in 1940, when annual per capita GDP was $90 (in 1940 dollars), which converts to $840 in 1990 dollars. Using Geary-Khamis 1990 dollars (which factors in power of purchasing power parity: PPP), Maddison estimates annual per capita GDP for Greece in 1940 at $2223. Allowing for a PPP rate of 1:1.5 (based on comparing CNTS and Maddison figures for 1990) brings the CNTS 1940 figure close to $1300. But Maddiso
n’s 1940 figure of $2223 still seems too high. Maddison’s population and GDP per capita figures for later decades of the twentieth century converge with the CNTS data, suggesting that his overestimates may be more extreme in the earlier, pre-twentieth century data.
CHAPTER 5 Explaining Hellas’ Wealth
1. The important role of action-guiding rules (institutions and political culture) in regulating market exchanges, and thereby addressing the failure of markets to deal adequately with “negative externalities” (e.g., toxic waste produced as a side effect of manufacturing being dumped into public waters) is a central feature of the “new institutional economics” (ch. 1). Valuing rules and regulation distinguishes institutional economics from free market fundamentalism, which generally rejects the idea of market failure, holding that negative externalities will be adequately dealt with by the market itself (i.e., people who can afford to will move away from the toxic zone; cheap enough real estate will attract less affluent others). Osborne 2009a goes further in embedding economics in social relations by positing that changes in ideas about political entitlement, citizenship, and elite standing can explain ancient Greek and Roman economic growth. This notion has obvious similarities to the arguments developed here, and Osborne rightly underlines the importance of the substitution of social capital for material wealth (e.g., through euergetism: voluntary public payments and services by the wealthy, who thereby gain in public esteem). But the mechanisms and microfoundations of the “changing ideas drive growth” theory remain underspecified. Bresson 2014 argues that assessing ancient economies in the terms of “capitalism” is justified.
2. Harris 2002. Cf. chapter 1 on the issue of the degree of horizontal vs. vertical specialization in the Greek world. I know of no comparable list for other ancient societies. Unfortunately, even if such lists were available, given the nature of the literary sources of the evidence for many of Harris’ specializations, and the lack of similar sources for most other ancient cultures, comparison would be difficult.
3. Cartledge 1998: 13.
4. List and Spiekermann 2013 demonstrate that a methodological focus on individuals as choice-making agents (in the form of “supervenience individualism”) is compatible with some forms of causal-explanatory “holism” in respect to considering institutions as collective actors. My two hypotheses assume that their compatibility thesis is correct.
5. Rawls 1999 [1971].
6. Emergence and the relationship between microlevel and high-level phenomena: Petitot 2010.
7. The reasons that a severe, long-duration collapse of civilization at the dawn of the Iron Age, in the context of distinctive geography, resource endowments, and climate, produced conditions especially conducive to the emergence of many small states (rather than a unitary empire) and of a norm of citizen-centered government (rather than monarchy) are discussed in chapter 6.
8. Here, I use “nature” in the thin sense employed by game theorists: exogenous facts about the world that set the framework in which human agents make strategic choices.
9. Meta-study: Finné et al. 2011.
10. Causal role of climate change in the EIA collapse: Kaniewski et al. 2010; Drake 2012. Cline (2014: 143–147) provides an excellent review of recent literature, concluding (2014: 170) that climate change was one of a number of factors that conjoined into a “perfect storm” to sink the Bronze Age.
11. Roman Warm: McCormick et al. 2012.
12. See further, Morris 2013b: 71 and literature cited; Bresson 2014: 51–52 suggests that climatic conditions cooler and moister than recorded for modernity could have favored agriculture in the Aegean by 500 BCE. My thanks to Ian Morris for bibliography and helpful discussion of these issues. He points out that we still need to explain why populations from Spain to China increased substantially in the first millennium BCE. Both climate studies and population estimates for the relevant periods are multiplying, and it is to be hoped that new data will enable us to better specify the extent, timing, and causes of population growth beyond the Greek world.
13. The modern Greek state’s coastline is 13,676 km. Among European countries, Greece’s coast/area ratio (104.68); is second only to Denmark (172.4); contrast Italy (26.0), Spain (9.95), and the United States (2.17): source of data: World Factbook, cited in http://en.wikipedia.org/wiki/List_of_countries_by_length_of_coastline. About 80% of Greece is mountainous; see discussion of polis elevations in chapter 2: http://www.nationsencyclopedia.com/Europe/Greece-TOPOGRAPHY.html. Mediterranean climate: ch. 2. Diversity and networks: Horden and Purcell 2000. Geophysics and the art of not being governed: Scott 2009.
14. Advantages in respect to location and climate: Xenophon Poroi 1.6–8 (specifically Athens). Aristotle Politics 7.1327b29 notes Hellas’ central location between Asia and Europe; Pseudo-Plato. Epinomis 987d regards the Greek climate, balanced between summer and winter, as optimal for the development of virtue.
15. Independence is, of course relative; per table 2.3; divergences in prosperity within regions of Hellas, in some cases linked to independence or its lack, are discussed in Chapters 6–11.
16. On the problem of endogeneity in explanation, see King, Keohane, and Verba 1994: 185–194. Another test would be to ask if other, non-Greek, societies that shared the advantageous location were also standouts in 800–300 BCE (or other periods); Cyprus, Thrace, and Sardinia are possible test cases.
17. Slavery and unfree labor: Scheidel 2005a, 2008a; Bang 2009.
18. Athenian empire and rents: Morris 2009; and this book, ch. 8.
19. Thrace and Athenian rents: Moreno 2008; and this book ch. 9.
20. Scheidel 2008a: 123–125.
21. My neologistic phrase “rule egalitarianism” (conceptually similar to what North, Wallis, and Weingast 2009 call “impersonality”) is modeled on the term “rule consequentialism,” commonly used by ethicists. The rule consequentialist focuses on social rules (as opposed to individual acts) that maximize aggregate welfare (or alternatively, aggregate preference satisfaction).
22. Eighth century egalitarianism: Morris 1987; and this book, ch. 6.
23. Greek egalitarianism: Morris 1996; Raaflaub 1996; Cartledge 1996. Runciman 1990 emphasizes the historically remarkable level of Greek egalitarianism. Foxhall 2002: 218 by contrast, regards “substantial inequalities in landholding” as a “paradox” that “I have never been able to resolve in my own mind.” The paradox arises, of course, if one supposes that egalitarianism requires either equal outcomes or equal opportunities (measured by equal access to all valuable resources). But, per above, rule egalitarianism assumes neither.
24. There is, of course, a good deal of specialization in centralized hierarchies—the complex monarchical civilizations of, e.g., Mesopotamia, Egypt, and Persia would have not been possible without specialists. Rational people may specialize because they are ordered to do so or because they are born into a social group traditionally assigned some specialized task. Such systems may work well, but they lack the marketlike features that, so I argue, drove classical Greek era innovation and creative destruction.
25. Histories 5.78. This passage is discussed in more detail in chapter 7.
26. Strategic calculation based on formal rationality in Greek thought: Ober 2009. Cf. North, Wallis, and Weingast 2009, who emphasize the behavioral implications of individuals being treated impersonally in institutional contexts. Note that I assume here not only formal equality of standing but also some degree of freedom of choice in respect to occupation. Obviously in practice the extent of free choice varied considerably, but it is the overall effect of differences in opportunities and incentives that produces the result of relatively greater investment in human capital. See further Ober 2012.
27. Xenophon, Memorabilia 2.7.7–10, esp. 2.7.10. Whether the women themselves would have concurred with Socrates’ assessment, we cannot say.
28. See, for example, Xenophon, Memorabilia 2.7.3–6.
29. This sort of investment in political, rather than specifically economic, skill
s may be a driver of increased use of slaves and other forms of unfree labor: Scheidel 2008a: 115–123. On the other hand, Xenophon (Memorabilia 3.4) points out that certain skills required for success in private business affairs are also valuable for managing public affairs. Public goods are in general nonrival (i.e., not a fixed quantity, so that that their use is not subject to zero-sum competition) and nonexclusionary (i.e., all relevant people have free access).
30. Hansen 1999: 314, and ch. 9 in this book.
31. Bankruptcy laws that limit personal losses and rules of incorporation that protect individual investors are familiar modern examples. Patron–client relationships and voluntary associations (e.g., burial societies) provided alternative, civil society, routes to similar ends in some ancient societies.
32. Gallant 1991 assumes high levels of risk aversion on the part of ancient Greek subsistence farmers and suggests possible family risk-buffering strategies, based in part on evidence from subsistence farming in early modern Greece. However, if the arguments presented here are correct, the classical Greek economy was not predicated on the risk aversion of families of subsistence farmers. Public insurance and risk: Burke 2005; Möller 2007: 375–383; Ober 2008: 254–258. Mackil 2004 shows how a somewhat similar risk insurance mechanism operated in some interpolis relations.
33. Numeracy: Netz 2002; banking and credit instruments: Cohen 1992; rhetoric etc. and social networks: Ober 2008 ch. 4.
34. Transaction cost economics applied to antiquity: Frier and Kehoe 2007; Ober 2008: 115–116, 214–220, 234–239; Kehoe, Ratzan, and Yiftach-Firanko forthcoming.
35. North, Wallis, and Weingast 2009, with discussion above, ch. 1.
