What the Roman Empire's demographics were like and what they mean

Vaclav Smil is a writer and researcher who occupies an enviable niche with his studies the interactions between human beings, the societies they create, and the natural environments that they inhabit, aiming to determine things like the factors influencing the sustainability of human civilization. One of his more recent books, this year's Why America Is Not A New Rome, was written with the aim of taking apart the myth that the United States is a new Roman Empire, complete with decadence and empire and ultimate collapse. This analogy works, I grant you, except for the many, many ways in which 21st century America and the 2nd century Roman Empire are fundamentally different. The United States is much more technologically and socially innovative than Rome ever was; by all accounts, Americans behave more realistically with regards to the wider world, and are certainly less prone to inhumanity; and at every level, most unlike Rome, the United States is part of a highly complex and globally integrated economy.

One chapter of particular interest to Demography Matters readers is his fourth, the simply-titled "Life, Death, Wealth." For students of demographics, the biggest problem with studying the populations of the classical era lies in the lack of data. What we know about the populations of the Roman Empire (and there were numerous distinctive regional populations, thanks to deep-seated environmental, cultural, and technological differences between Rome's provinces, as opposed to a single Roman Empire population) is a combination of careful analysis of surviving documents, archeological studies, and analogies with modern societies. Rome, Smil points out, was a society trapped in the first stage of the demographic transition, with very high fertility rates largely counterbalanced by very high death rates. Walter Scheidel's papers on classical demography go into detail on specific areas of the classical world; Wikipedia's article on classical demography provides a useful overview; Wikipedia user G.W.'s detailed primer goes into much greater, and gorier, detail.

Inhabitants of the Roman Empire had a life expectancy at birth of about twenty-five years. Although the figure relies more on conjecture than ancient evidence, which is sparse and of dubious quality, it is a point of general consensus among historians of the period. It originates in cross-country comparison: given the known social and economic conditions of the Roman Empire, we should expect a life expectancy near the lower bound of known pre-modern populations. Roman demography bears comparison to available data for early twentieth-century India and rural China, where life expectancies at birth were also in the low twenties.

About 300 census returns filed in Egypt in the first three centuries CE survive. R. Bagnall and B. Frier have used them to build female and male age distributions, which show life expectancies at birth of between twenty-two and twenty-five years, results broadly consistent with model life tables. Other sources used for population reconstructions include cemetery skeletons, Roman tombstones in North Africa, and an annuities table known as "Ulpian's life table". The basis and interpretation of these sources is disputed: the skeletons cannot be firmly dated, the tombstones show non-representative sample populations, and the sources of "Ulpian's life table" are unknown. Nonetheless, because they converge with low Roman elite survival rates shown in the literary sources, and because their evidence is consistent with data from populations with comparably high mortality rates, like eighteenth-century France, and early twentieth-century China, India, and Egypt, they reinforce the basic assumption of Roman demography: that life expectancies at birth were in the low twenties.

As no population for which accurate observations survive has such a low life expectancy, model life tables must be used to understand this population's age demography. These models, based on historical data, describe 'typical' populations at different levels of mortality. For his demographic synopsis of the Roman Empire, Bruce Frier used the Model West framework, as it is "the most generalized and widely applicable". Because it is based on only one empirical input, the model life table can provide only a very approximate picture of Roman demography. On two important points, the table may seriously misrepresent the Roman situation: the structural relationship between juvenile and adult mortality, and the relative mortality rates across the sexes. In any case, Roman mortality should be expected to have varied greatly across times, places, and perhaps classes. A variation of ten years would not have been unusual. A life expectancy range of between twenty and thirty years is therefore plausible, though it may have been exceeded in either direction in marginal regions (e.g., malarious urban districts on one end; high-altitude, low-density settlements on the other).

The specifics of any ancient age distribution, moreover, would have seen heavy variation under the impact of local conditions. In pre-modern societies, the major cause of death was not the chronic, end-of-life conditions that characterize mortality in industrialized societies, nor primary malnutrition, but acute infectious disease, which has varied effects on age distributions in populations. Pulmonary tuberculosis, for example, characterized much of the Roman region in antiquity; its deaths tend to be concentrated in the early twenties, where model life tables show a mortality trough. Similarly, in pre-modern societies for which evidence is available, such as early modern England and early eighteenth-century China, infant mortality varies independently of adult mortality, to the extent that equal life expectancies at age twenty can be obtained in societies with infant mortality rates of 15% to 35% (life table models omit this; they depend on the assumption that age-specific mortality ratios co-vary in uniform, predictable ratios). No ancient evidence can gauge this effect (there is a strong tendency to overlook infant death in the sources), and the model life tables may overstate it, but comparative evidence suggests that it is very high: mortality was strongly concentrated in the first years of life.

Smil notes, quite rightly, that the classical demographic pattern has no parallels in our contemporary world.

Today an analogue of these Roman values exists only in terms of continuing high birth rates and total fertility rates in Western, Eastern, and Central Africa; in 2007 these regions had average birth rates of, respectively, 42, 41, and 46/1,000, and average total fertilities of their countries ranged between 5.5 and 6.4. But even in these sub-Saharan countries, where the demographic transition has yet to run its full course, death rates have been already reduced quite significantly, to between 5 and 17/1,000, still nearly twice the current global mean of 9/1,000 but only about 35%-40% of the high Roman value (123).

[. . .] There is no modern population--even among the worst-off countries of sub-Saharan Africa--whose growth, longevity, and age structure would even remotely resemble the ancient Roman pattern. Perhaps the best contemporary analogy would be to imagine a population that is even more destitute and desperate than those of Sierra Leone or [Burkina] Faso and then to contrast it with the long-lived, formerly fast growing and now rapidly aging U.S. population, whose mean life expectancy at
birth is more than three times as high as the Roman Empire's (126).

Taking a look at a list of the world's countries by life expectancies, the only countries that come close to the likely life expectancy of the average Roman are a long list of terribly poor countries, all but Afghanistan located in sub-Saharan Africa, most with very high levels of HIV infection in addition to any number of other illnesses. But even the worst-off country, Swaziland, comes at 39.6 years at least a decade ahead of the Roman average and on par with the luckiest Roman districts. Combine this with the very high disease load of the average Roman and sustained undernourishment--Smil cites evidence suggesting that, at least as measured by average height, the food supply improved after the Roman Empire's collapse in the west--and the picture of a congenitally unhealthy ppopulation is inescapable. Combine that with the abundant evidence for the exceptionally unequal distribution of wealth and power within the Roman Empire, and with the significantly lower level of economic output (estimated by Smil to be inferior to that of central Africa and, of course, lacking the imported technologies available to central Africans) and you have a Rome that stands few comparison with even the worst-off countries of our 21st century world.

Wikipedia's G.W. pointed out that--to say the least--mortality and illness on this scale hampers economic growth.

Mortality on this scale discourages investment in human capital, hindering productivity growth (adolescent mortality rates in Rome were two-thirds higher than in early modern Britain); it creates large numbers of dependent widows and orphans; and it hinders long-term economic planning. With the prevalence of debilitating diseases, the number of effective working years was even worse: health-adjusted life expectancy (HALE), the number of years lived in good health, varies from life expectancy by no more than eight percent in modern societies; in high-mortality societies such as Rome, it could be as much as one-sixth beneath total life expectancy. A HALE of less than twenty years would have left the empire with very depressed levels of economic productivity.

It's difficult to avoid concluding that death and suffering on this scale had an effect on the cultures of the time. Smil remarks that, even though Romans accepted the fundamental humanity of slaves, Roman slaveowners--like their counterparts in the Atlantic slave/sugar economy more than a millennium later, in pre-revolutionary Haiti say--were still quite willing to treat their human property quite inhumanely (137-138). This habituation to extreme suffering, this coarsening of human sensibilities, isn't the sort of mindset that would support the sorts of rational, non-zero-sum social and political bargains that created the institutions undergirding the 21st century world.

Demographic patterns not too far removed from the Roman Empire's are normal for human beings, almost 95% of whom lived on Earth before it broke from the established demographic patterns in the 20th century. Even in favoured long-prosperous Canada, they prevailed within a century of my birth.



Sadness
Originally uploaded by etherflyer

(This photo of a late 19th century Torontonian child's grave was taken by a friend of mine, and used here and in another blog post of mine; he's since taken others.)

We at Demography Matters are concerned with seeing where established trends will take us. Tonight, a day before the new year, I thought I'd take a look back to see where we escaped. It's worth gauging the distances between then and now, I think.

On Spike Japan

Spike Japan, maintained by Tokyo blogger Richard Hendy, is one of the more interesting blogs out there, certainly among the more original blogs taking a look at the intersection of demographics with economics. Documenting his travels to areas of Japan outside of metropoli like Tokyo in acute essays and well-chosen photos, places that are slowly (or quickly) falling apart owing to a combination of two decades of slow-to-no economic growth and ever accelerating depopulation, Hendy got some international attention via this article from The Guardian written by Chris Michael earlier. How did he start? He describes Spike Japan's genesis in his introductory essay "Down the benjo: The ruin/nation of Japan".

It may come as a shock to almost all of you living outside of Japan, and to some of you living in the center of its big cities, that as we approach the summer of 2009, swathes of the country are in ruins. It came as a shock to me, too, I have to confess, having lived for almost all of the last decade in the bubble of central Tokyo and only venturing outside occasionally to get to the airport, nearby beaches, and old friends in the mountains.

As the financial results came in for Japanese companies for the last quarter of 2008, in late January and early February, they were suggestive of a complete cessation of activity in certain sectors of the economy. My interest in what happened outside of the bubble I live in was only really piqued, though, by a report in the Financial Times that many of the Brazilian emigrants to Oizumi, the town in Japan with the highest percentage of such residents, were ready to pack for Sao Paolo. I went out to investigate and was fascinated and amazed by the combination of immigration, industry, bleakness, and normality. A series of e-mails I had anyway been writing to a friend gained traction from this visit, and developed into a series titled “Down the benjo”; “benjo” is a no longer particularly polite Japanese word for “bog” or “john”. I owe inspiration for the title from a reference early in the year by Willem Buiter, Professor of European Political Economy at the London School of Economics, on his maverecon blog at the Financial Times, to the Icelandic economy having been flushed “down the snyrting”. Much the same, it struck me when I read his post back in January, was happening to the Japanese economy.

The visit to Oizumi led to another, this time to Hitachi, the home of the multinational conglomerate of the same name, which prompted a longer e-mail essay, which in turn prompted a ramble down a recently abandoned railway line less than an hour from Tokyo, which resulted in a 15,000-word essay that took the best part of a month of free time to put together.

Hendy has visited rural areas and small cities, two stand-out travelogues being his extended exploration of the northernmost and most recently settled major island of Hokkaido, and his recent sojourn to the
Amakusa islands off the west coast of Kyushu. Whether in Hokkaido, Amakusa, or elsewhere, the regions of Japan that Hendy visits are all areas located away from its prosperous industrial and urban centres, substantially rural, blighted economically and scenically by Bubble-era constructions, lacking in innovative local enterprises, heavily indebted, and sharing in the general drain of the young to the cities. After the recent financial crisis, the prospects that these regions might receive the investment that might turn things around--making them destinations for retirees, say--are trivial. The dream of a return to the land is ridiculous.

According to the Rural Depopulation Research Association, "There are probably a lot of people who would like to move to the countryside if the conditions were right, (but) it's difficult to see how the number could increase with the present situation. The local communities need to maximize their areas' resources."

The "I-turn" movement (moving from the city to the country) and the "U-turn" movement (people from the country heading to the city, then back again) have been around since the 1980s. However, the things that discourage more young people from moving to the countryside are the same as ever.

"As things are," says Yuzawa. "Even if people want to go back the countryside, often there is nowhere for them to work and nowhere for them to live."

[. . .]

A survey by the Rural Depopulation Research Association in 2000 found that "company work" is the most popular choice for those that have already moved to the countryside. In other words, they avoid the shortage of work by commuting to the city. Relatively few work in the government construction industry, which plays a major part in the rural economy. Other U-turn and I-turn employment ranges from tourism to geriatric care to traditional crafts.

With the help of the help of the organization, local governments try and match jobs to candidates; but the right work isn't always available, and sometimes idealism isn't enough to persuade young people to give up city salaries.

But the Japanese country needs help from somewhere.

In a 2000 survey, more than a third of Japan's municipalities were classified as depopulated more than half Japan's land area. All had lost more than a quarter of their residents since the 1960s.

The thing is, what's happening to rural and marginal Japan is going to happen to urban and core Japan sooner or later. Trend economic growth, as Hendy observes, is inexorably trending downwards towards nothing, as the workforce continues to contract, the dependency ratio tips crazily, productivity stagnates and foreign competition grows and infrastructure ages.

It might just be, however, that despite recent evidence to the contrary, Japan has embarked on a vicious demographic spiral, in which a variety of complex feedback mechanisms set to work: aging results in declining international competitiveness, which results in greater economic hardship at home, which results in a suppressed birthrate; aging results in ballooning fiscal deficits, which in the absence of debt issuance must result in higher taxes or cuts to government spending, which cause economic pain, driving down the birthrate; aging, as the elderly dissave, results in a decline in the pool of domestic savings on which government borrowing is an implied claim, reducing room for fiscal maneuver and resulting in less ability to withstand exogenous shocks; aging further entrenches conservative attitudes to everything from pension reform to immigration, resulting in greater government outlays and smaller government receipts; aging leads the electorate to fear for the future of the pension system, resulting in more saving by the economically active, depressing consumption, which drives manufacturers offshore and raises unemployment, which is strongly correlated with the birthrate.

Japan might be an extreme case, not least because of its lack of immigration--South Korea has become much more of an immmigrant country in a shorter period of time--but it's certainly not the only global economic power out there with lowest-low fertility. There's Germany, say, and certainly the various descriptions of the former East Germany's rapid population aging and shrinkage doesn't sound out of kilter with what Hendy has been writing about and photographing.

Spike Japan is one of those blogs that works on two different levels, as a personal travelogue and as an extended meditation on the existential economic problems of post-growth societies. Visit it for both of these reasons.

On sputtering integration in Toronto

One of the less cheerful tags ar my personal blog is "three torontos". The tag comes from a phrase in the title of a 2007 report by the University of Toronto's David Hulchanski, who found that Toronto's neighbourhoods could be divided into three categories based on patterns of income growth: neighbourhoods which saw significant income growth over the 1970-2000 period; neighbourhods which more-or-less stagnated (growth or decline of less than 20%); and, neighbourhoods which saw significant income decline. These divisions map onto enduring social, geographic, and ethnic divisions in Toronto, onto any number of patterns like the distribution of cyclists, voting in the recent municipal election, and the boundaries of the once-autonomous communities in an amalgamated Toronto. A follow-up study, available here and covered in the Globe and Mail by Anna Mehler Papierny, suggests that on the balance of existing trends Toronto's going to be polarized into two areas, have and have-nots.

Toronto is becoming a city of stark economic extremes as its middle class is hollowed out and replaced by a bipolar city of the rich and poor – one whose lines are drawn neighbourhood by neighbourhood.

New numbers indicate a 35-year trend toward economic polarization is growing more pronounced: The country’s economic engine, which has long claimed to be one of the most diverse cities in the world, is increasingly comprised of downtown-centred high-income residents – most living near subway lines – and a concentration of low-income families in less dense, service- and transit-starved inner suburbs.

Three years ago, University of Toronto professor David Hulchanski published a paper on Toronto’s “Three Cities,” illustrating a growing socioeconomic disparity among the city’s census tracts. But the three-way divide Prof. Hulchanski and his fellow Cities Centre researchers described is swiftly being reduced to two, according to a new paper they will release Wednesday. Toronto, a predominantly middle-class metropolis just three decades ago, is increasingly dominated by two opposite populations – one with an average income of $88,400, and another of $26,900.

These two groups live in different neighbourhoods, work in different sectors, send their children to different schools and have divergent and unequal access to city services and public transit. Even the 905-area suburbs outside of Toronto are seeing a dramatic drop in the proportion of middle-income earners in their population, the report finds.

Those in the lowest-income areas are also more likely to be immigrants and visible minorities.

“It’s only going to become worse,” Prof. Hulchanski said. If the trend continues, the paper suggests, Toronto in 2025 will have a concentration of high-earners along the lakefront and the city’s subway lines surrounded by low-income areas – with almost nothing in between.

[. . .]

It also seems to contradict Toronto’s most prized mottos – “Diversity our strength” and “The city that works.” Neither of those rings true any more: Toronto’s diversity is becoming balkanized, turning it into a weakness where it could otherwise act to the city’s advantage. The creation of economically polarized pockets of high- and low-income residents means Toronto simply won’t “work” as a municipal entity.

“We used to brag about it,” Prof. Hulchanski said. “ ‘Toronto’s an efficient city – it works.’ We know now that’s not true.

“To have so much poverty in one geography and for it to be so deep and for the social distance to be so large … that isn’t healthy.”

In a five-year period alone, average incomes declined in 34 of the city’s census tracts (about 7 per cent of its total) – 23 of those areas became predominantly low-income. At the same time, 12 areas became high-income and nine earned “middle-income” status.

This has to do with the exclusion of immigrants from the labour force. As the Toronto City website boasts, Toronto's population is quite cosmopolitan, absolutely and relative to other Canadian cities. A variety of sources suggest that new Canadians just aren't fitting into the labour market, as evidenced by current unemployment rates.

While jobless rates dropped both nationally and locally – to 7.6 per cent Canada-wide, the lowest level in two years, and to 6.7 per cent from 9.2 per cent earlier this year in Toronto – unemployment is ramping up for people who have come to Canada in the past five years.

In Toronto, 19.7 per cent of recent immigrants are unemployed. That’s far higher than the 13 per cent who were jobless just a year ago, and nearly three times the jobless rate for Canadian-born residents.

It’s not unusual for immigrants to be hit harder by recession and to take longer to recover their job prospects. But Toronto relies more on immigrant labour now than it has in the past: As of 2011, virtually all of the city’s job-market growth depends on immigrants.

“Because of the fact that more than 50 per cent of our residents are foreign-born, there’s a sharper thrust and a higher stress for us to do really well,” said Ratna Omidvar, president of the Maytree Foundation.

What’s perennially missing in a city with a plethora of disconnected services and growing socioeconomic stratification, advocates argue, are the tools to connect immigrants to jobs. To this end, the federal government has pledged $2.3-million in funds to help Torontonian immigrants integrate – cash that has gone to programs started in May of this year.

And a new initiative through Scotiabank is teaming up with the Toronto Region Immigrant Employment Council to try to link immigrant professionals with their Canadian counterparts.

Immigrants who’ve arrived in Canada since the financial crisis face a growing Catch-22 of employment barriers, says TRIEC executive director Elizabeth McIsaac. They are at a disadvantage from the start, and the longer their lack of Canadian experience bars them from the job market, the harder it is to join and the longer their unemployment is a drag on the rest of the economy.

“If you landed in the middle of a recession and you didn’t get your first opportunity, your time out of the market exacerbates the challenges you had getting into it.… It begins to have a multiplying effect – a real scarring effect on immigrants.”'

The gap between immigrant and native-born worker incomes is taking an extra generation to close, a combination of competition from guest workers and the semi-legal labour market, a lack of Canadian-recognized credentials, and--quite possibly--the continued exclusion of new Canadians from the closed social networks of established employers and professional groups which let people join the labour market at a level befitting of their skills.

So. Ontario--including Toronto--may be doing better than in Québec in integrating immigrants into the mainstream labour market, and it's certainly doing a better job of avoiding creating a metic class than countries with less porous immigration regimes like Germany. Even with a national economy that has been performing quite strongly relative to most of its First World peers, this still isn't good enough to avoid creating a very problematic social and geographic pattern of relative deprivation linked to ethnic and national origins in Ontario's, and Canada's, largest city. This can lead in very negative directions. On ethical grounds alone, this is unacceptable.

Any suggestions as to how Toronto--and other cities--could pull out of this? Getting a sufficiently dynamic labour market, and associated economy, is key. Is there best practice to be productively shared?