Tuesday, May 31, 2011

More on the demographics of Canadian politics

As I blogged earlier this month, the radically changing appeal of different political parties to different ethnolinguistic and territorial communities played a major role in the massive surprises (Conservative majority, New Democratic Party official opposition, devastation of the Liberals and especially the Bloc Québécois) in this month's Canadian election.

Éric Grenier of the Canadian political blog Three Hundred Eight has recently analyzed in depth the demographics of the ridings of the three major political parties left in the realm of Canadian federal politics, and sent the resulting articles to Toronto's Globe and Mail. The changes have been significant. Take a look at hi notes on the Conservative Party.

With a gain of 23 seats over the party’s standing in the House of Commons when the election campaign began, the Conservatives now represent 19.2 million Canadians, or 2.8 million more than they did before the vote. Their caucus is still disproportionately based in the West, with 74 MPs coming from the four western provinces and northern Canada. This group forms the largest regional block in the Conservative caucus, but has shrunk to 44.6 per cent from 49 per cent of all Tory seats.

With 73 seats in Ontario, 44 per cent of the Conservative caucus is drawn from the province, up from 51 seats and 35.7 per cent before the election.

Almost 89 per cent of all Conservative MPs were elected west of Quebec, leaving only 3 per cent of the party’s caucus coming from the francophone province and 8.4 per cent from Atlantic Canada. While that is a small increase for the four Atlantic provinces, it is a drop from 7.7 per cent for Quebec. Only five of the 166 Conservative MPs in the House of Commons won their seats in Quebec.

[. . .]

The Conservative constituency remains overwhelming anglophone, with 72.4 per cent having English as their mother tongue, virtually unchanged from before the election. That is far more than the 57.2 of Canadians throughout the country whose first language is English.

With the seat losses in Quebec, only 5.8 per cent of Canadians in Tory ridings are native French speakers, down from 9.5 per cent before the election and a fraction of the 21.8 per cent of Canadians who are francophones.

Accordingly, the proportion of Conservative constituents who speak a language other than French or English has grown to 20.4 per cent from 16.3 per cent, almost reaching the national average. It is the same situation with the immigrant population in Tory ridings. They now represent 19.9 per cent of the Conservative constituency, almost identical to the 19.8 per cent of Canadians who are immigrants. Visible minorities, too, are now more represented in Conservative ridings, going from 10.6 per cent before the election to 15.3 per cent after the votes were counted. In 12 of the 166 Conservative ridings, visible minorities actually make up a majority of the population.


The New Democratic Party has also seen huge changes.

With a gain of 67 seats on election night, the NDP went from representing 3.8 million Canadians to 11.2 million, or roughly one-third of the country’s population.

In the process, the regional weight of the NDP’s caucus drawn from the West, Ontario, and Atlantic Canada was reduced dramatically – despite the party’s gains in each region. While the NDP increased its Western caucus from 14 to 16 seats, that now represents less than 16 per cent of the entire NDP caucus. Prior to the election, the West contributed almost two out of every five NDP seats.

With five seat gains in Ontario, the party’s caucus is still only 21.4 per cent from that province, while the two seat gains in Atlantic Canada come with a reduction in the region’s weight to only 6 per cent of all NDP seats.

Prior to the election, the one NDP seat in Quebec represented less than 3 per cent of its national caucus. Now, the 59 New Democrat MPs make up the majority (57.3 per cent) of the party.

[. . . I]t is in the linguistic profile of people the NDP now represents that we find the most change. In the 36 ridings held by the New Democrats before the election, English was the mother tongue of 66.1 per cent, well above the national proportion of 57.2 per cent. Only 7.9 per cent of the NDP’s constituents were native French speakers, while 24.3 per cent had a mother tongue other than English or French.


And finally, the ridings that the Liberal Party managed to keep--fragmented, concentrated in poorer regions of the country like Atlantic Canada, and having lost significant representation in French Canada and among immigrant and First Nations demographics--don't bode well for the party.

The regional distribution of Liberal seats has shifted eastwards. Whereas 48 per cent of the Liberal caucus (37 seats in all) came from Ontario when the government fell in March, the province now supplies only 32 per cent of Grit MPs. The proportion of seats coming from the West and Quebec has not changed significantly, but the representation of Atlantic Canada in the Liberal caucus has grown to over 35 per cent from 22 per cent. The 12 MPs from the four Atlantic provinces now form the largest regional block in the Liberal caucus.

[. . .]

The Liberals are still a party whose constituents are more educated and more diverse than the national average. But little more than 1 in 10 of the people they represent are native French speakers and more than one in three of their MPs come from Atlantic Canada. As the party re-tools with a view towards the next election that will take place more than four years from now, the demands of these different constituencies may clash with the message of a party striving to become a national force in Canadian politics once again.


The major problem with Grenier's analysis is that he analyses here not the people who actually voted for these parties, but rather the demographic composition of each of the different federal ridings. It's easy enough to imagine situations in pluralistic ridings where the swing of one group or another was responsible for a shift from one political party to another, other groups remaining firm in their support. Regardless these observations---only very partially excerpted here--deserve to be observed. It's difficult to avoid saying that the 2011 federal election in Canada was utterly transformative. Grenier helps explain how this transformation came about.

Saturday, May 28, 2011

The problems with sub-national immigration policies

Earlier this month, the Burgh Diaspora reflected on the history of immigration to the United States, the variant forms of immigration sought by different states, the tensions between different American states over differing attitudes towards immigration and immigrants, and made a proposal: why not let subnational entities control their own immigration policies?

I think we should embrace the anti-federalist mood swing. Allow states such as Pennsylvania to embrace talent immigration as each sees fit. Better yet, let cities decide. H-1B visas effectively tie foreign born labor to the employer through sponsorship. Municipalities could act in the same capacity, making the visa contingent on urban residence and site of work. Various schemes could be concocted to enhance geographic mobility. Green cards would issued after a few years, well before the end of the federal queue.


This does make a certain amount of sense. In the realm of actually existing subnational immigration policies, I'm most familiar with that of Québec, where--as I wrote in June of last year--concerns over the growth of English led to the adoption of a policy explicitly favouring Francophone and French-leaning immigrants over others (French and Senegalese and Congolese versus Britons and Indians and Guyanese, say). This shift has arguably made immigration more popular in Québec, removing the fears of language shift from French, and has the potential to provide Québec with the workers--including skilled workers--that it will need as the provincial population ages. The rest of Canada obviously doesn't share the priorities of Québec and wouldn't be as responsive to local concerns. So, inasmuch as the subnational jurisdiction of Québec's control over immigration policies go, it seems a relative success.

The problems with sub-national immigration policies? Immigrants don't necessarily fit the slots allotted to them. As I noted above, the gap between immigrant and native-born wages in Québec is even worse than in the rest of Canada, a product of many things including the non-recognition of skilled workers and difficulties with social integration. In many cases, it's not especially clear that local control over immigration would be an improvement.

The second problem is that of mobility. For municipalities to have control over immigration, as the above blog goes on to sugget, strikes me as a very bad policy move. Immigrants have to be mobile, geographically as well as socially, and a municipality doesn't necessarily offer sufficient scope. In Québec the overwhelming majority of immigrants may be concentrated in Montréal, but this isn't because they're forced to live there. Rather, immigrants are concentrated in Montréal because that's where immigrant communities have formed neighbourhoods, dense social networks, and the like. Restrictions on mobility are especially problematic if--as some propose--immigrants are assigned residences in hinterlands in an effort to try to boost stagnant or declining populations. The waste of potential in those cases is arguably as much a moral problem as an economic one.

Finally, there's the question of whether immigrants will stay in their localities once probationary periods are up. My native Prince Edward Island has an immigrant retention rate of 25%; for cited reasons of wages (the poorest province in Canada) and social integration, most immigrants do not stay. Many sub-national jurisdictions may not keep as many immigrants as planned--in Canada, the differences between the have and have-not provinces on this metric is notable.

In conclusion? Sub-national immigration may be a useful idea, but it's one that definitely has its serious issues. It may produce short-term gains, but those gains can quickly be dissipated with bad planning and bad underlying conditions. Beware.

Thursday, May 26, 2011

Some demography links

This afternoon--this afternoon in Toronto, at least--I thought I'd share with you five interesting demographics-related blog posts.

  • At Behind the Numbers, Mark Mather reports on the news that men in the United States are catching up to women in terms of life expectancy.

  • Dealing with my earlier post about biased sex ratios in the South Caucasus, fellow Livejournaler demographer has written--translated post here, original post here--on the subject, suggesting that the apparent deficit is a consequence of the underregistration of female births.

  • Geocurrents' Martin Lewis writes about the barrier along the Pakistani border built by the Iranians, dividing the historical region of Baluchistan. While built in an effort to control insurgents, the wall has a secondary use of limiting Pakistani migration into Iran.

  • The Global Sociology Blog links to an extended BBC series describing the effects of the male-biased sex ratio in India, describing things as varied as the grief of women forced to abort their daughters, the migration of women from more sex-balanced areas in southern India to the male-biased north, the shocking deterioration in the sex ratio in Kashmir, and Indian government efforts to encourage the birth of female children.

  • At Marginal Revolution, Tyler Cowen summarizes a new book by Youssef Courbage and Emmanuel Todd and the subtitle is A Convergence of Civilizations: The Transformation of Muslim Societies Around the World. This book makes the point, made here several times, that Muslim societies around the world are also going through the demographic transition, albeit each in their own distinctive ways.
  • Why Mongolia's set for massive urbanization

    I owe thanks to Sublime Oblivion's Anatoly Karlin for linking to Kit Gillet's article in the Guardian, "Vast Mongolian shantytown now home to quarter of country's population". It turns out that over the past two decades, the Mongolian capital of Ulan Bator, has grown immensely, developing shantytowns as the traditional nomadic herding lifestyle of the Mongols becomes non-viable and despite the influx of funds from Mongolia's new mines.

    Stretching north from the capital, Ulan Bator, an endless succession of dilapidated boundary markers criss-cross away into the distance.

    They demarcate a vast shantytown that sprawls for miles and is now estimated to be home to a quarter of the entire population of Mongolia.

    More than 700,000 people have crowded into the area in the past two decades. Many are ex-herders and their families whose livelihoods have been destroyed by bitter winters that can last more than half the year; many more are victims of desertification caused by global warming and overgrazing; the United Nations Development Programme estimates that up to 90% of the country is now fragile dryland.

    Yet with limited education, few transferable job skills and often no official documents, most end up simply waiting, getting angry with the government and reminiscing about nomadic lives past. Many take to alcohol.

    "More and more people arrive every year and there are so few jobs available," said Davaasambuu after queueing for 30 minutes to collect his family's daily drinking water from one of 500 water stations that dot the slum. "Nothing has changed in my neighbourhood since the last election [in May 2009]. There have been no new jobs or improvements. One little bridge has been added in the last four years, that's it," he said.

    [. . .]

    [W]hen temperatures plummet into the minuses for up to eight months, poorer residents are forced to spend upwards of 40% of their income on wood or coal for heating, which adds to their financial burden as well as to the heavy clouds of pollution that hangs over the city. Roads are simple, unpaved mud paths and streets have no signs, streetlights or even names but are merely the gaps naturally placed between two rows of tents or shacks set up by newly arrived migrants without any input from the government.

    "The quality of the infrastructure is a major problem," said Mesky Brhane, a senior urban specialist with the World Bank, who helped produce last year's report. "[People] are clearly frustrated by the lack of infrastructural improvements by the government.

    [. . .]

    Even in the more central ger areas, where many residents have lived for over a decade and built more permanent wooden or brick houses, running water and central heating are unavailable and the streets remain dark, mud roads with open sewage streams and rubbish piled high.

    Another big concern is the level of unemployment. While tens of thousands of rural migrants flood the city every year looking for work, setting up their tents at the point where last year's migrants stopped, unemployment remains a critical issue, especially in the ger districts where the unemployment rate can be as high as 62%, compared with 21% in the more developed areas of the capital.


    None of this should be a surprise. Dutch sociologist Paul Treanor's 2001 essay "Mongolia and Wyoming/Montana" predicted this very outcome. In the previous essay, Treanor compared the very poor Mongolia with the very rich American states of Wyoming and Montana. Mongolia and Wyoming/Montana, Treanor observes, are both territories which compare closely "in terms of climate, altitude, vegetation and population density". What are the differences?

    Density in both areas is very low by European standards. (In the European Union 8 inhabitants/km2 is the threshold for regional assistance to low-density areas). The Gobi desert is an empty area on any map of world population density. There are some almost empty areas in central Montana, and in semi-desert south-west Wyoming. Nevertheless, the rail infrastructure in the US is better than in Mongolia, and the local road network far better. The same will apply to energy infrastructure, and above all to the 'social infrastructure' such as schools and health care.

    Increasingly, however, both regions have the same economic basis: mining. No major industry ever developed in Wyoming and Montana anyway: and in Mongolia the non-extractive industrial sector has collapsed. So there has been a certain convergence of the economic base - but that base is better developed in the two US states anyway. Although reports on Mongolia refer to the 'massive' Soviet-built coal mines, Wyoming produces far more coal (over 300 million tons). In reality Mongolia's coal production (and per capita use) is tiny, in comparison with the US. Montana has a 120-year history as a copper-mining state. Both states are also well ahead of Mongolia, in the development of hydroelectric power.

    In Montana, agriculture is a larger sector than mining, - but part of the state has better climate and soils than any area of Mongolia. 18 % of Montana is arable land (concentrated in the north-east corner), compared with only 1% in Mongolia. Wyoming is, like Mongolia, about 75% grazing land.

    General agricultural productivity on Mongolian territory is very low. Compare Mongolia with agriculture in Poland (still considered a low-productivity agricultural sector in comparison with western Europe). In 1997, total cereal production per km2 was about 525 times higher in Poland. Meat production per km2 was 50 times higher in Poland. These figures are for total land area, and reflect primarily the difference in climate, geography, and ecology. In fact much of Mongolia is 'agricultural land', perhaps more than in Poland, but only in the sense that herds sometimes graze there. It took about 40% of the population to reach even that level of meat production. Cereal production was concentrated on the Soviet-built state farms. It has collapsed since 1989, from 416 kg/person to 81 kg/person, despite the present cereal shortage. That suggests that even these farms were only viable with subsidies, and outside technical assistance.

    The low agricultural productivity reflects the harsh climate of Mongolia. In fact the combination of cold and aridity is probably harsher than in Wyoming and Montana. In relation to the ecological limitations, the inhabitants had successfully adapted to these harsh conditions. The system of pastoral nomadism in Mongolia emerged over a period of thousands of years, along with others in the Eurasian steppes and deserts. It survived almost unchanged until about 1910. There was no similar range of pastoral nomadic cultures in North America.

    In other words, there was in Mongolia a unity of culture, history, economy and society based on pastoral nomadism. There was a pastoral-nomadic economy, in a pastoral-nomadic society, with a pastoral-nomadic type of culture, and a history characteristic for Eurasian steppe nomads. Mongolia is still inhabited by people who are culturally familiar with this unity: for many of them it is still daily reality. People outside Mongolia are also vaguely familiar with it: at least, they can associate Mongolia with yak herds, nomad tents, and Ghengis Khan.

    In contrast, the original inhabitants of Wyoming and Montana were militarily defeated, and marginalised for generations. (The Indian Reservations are known, even outside the United States, as examples of marginalisation). An entirely new society and economy was substituted for the existing version. The new population came primarily from rural Europe: for them, food production meant primarily the family farm. During the 19th century, the immigrants developed a cultural adaptation to the steppe/prairie zone: the cattle ranch. Although the cattle (and the horses) were imports from Eurasia, the system worked. But despite all the great cowboy mythology, the settlement of the American west was not primarily based on ranching. It certainly could not be based on ranching today: the ranch population is now a fraction of the state total.


    Treanor goes on to point out that Mongolia is far more removed from the major population centres of Eurasia than Wyoming and Montana are from the population centres of North America, that the relative cultural uniformity of the United States makes migration to and from Wyoming and Montana more possible than migration to and from Montana, that the tourism industry that thrives in Wyoming and Montana barely exists in Mongolia, and--crucially--that Mongolia is not part of a federal state that could subsidize its peripheral areas. At the same time, the nomadic herding lifestyle of the Mongols is becoming increasingly unsustainable, not least because of the quadrupling of the Mongolian population over the past hundred years. Arguably, most of Mongolia is too marginal to sustain local populations. The consequence of all this?

    [T]he entire pattern of settlement is both artificial and dependent on a base population of herders. It there is no nomadic herding, then there is little reason for anyone to live in zones of natural pasture. It the clusters of nomadic herding population disappear, then many [small] centres then lose their reason for existence. In turn, most aimak centres exist primarily as second-tier service centres, and if the base population migrates, their local economy would also collapse.

    The Gobi population is small enough, in absolute terms, to fit into a few mining and oil towns. In contrast, the forest-steppe zone will probably lose much of its population. Why this prediction? It is extremely unlikely that the nomadic pastoral lifestyle will survive for another generation: overall productivity is extremely low. If a high-productivity form of meat production replaced nomadic herding, the rural population might be partly stabilised. If not, then the rural population will have the choice of staying where they are, as the poorest people in Asia - or migrating. Given the predicted growth of the Chinese economy, and the demographic labour shortage in Russia and western Europe, emigration will probably be easier than at present. A special case is the Bayan-Ölgiy aimak: the population is mainly Kazakh. There has already been some migration to Kazakhstan: an oil boom there might attract much of the remaining Kazakh population.

    [. . .]

    The exceptional status of Ulaan Bataar is obvious. The industrial centres Darhan, Erdenet and (on a smaller scale Choibalsan), are the result of planned concentration of investment. They were created by decisions at national level. That is also true for the aimak centres, with primarily a service function. Industrialisation of the aimak centres seems improbable. They are remote and relatively small, with no existing industry, except processing meat and hides. Their 'gross product' is comparable to agricultural villages in western Europe. They will probably be much the same size in 2025 - about 15 000 to 25 000 inhabitants.

    That leaves Ulaan Bataar. The most reasonable prediction of the future population distribution is that the majority of Mongolians will live in one city. At present the best example of 'primate city' growth is Tirana in Albania. That is also a country with extreme rural poverty, and a collapsed industrial sector. Tirana has doubled (perhaps tripled) its population in a decade. However Albania also has a rich neighbour, Italy, and an extremely high rate of illegal emigration. And it has an urban tradition in the coastal regions, and an existing urban hierarchy with regional centres. Mongolia's medium-term future is extreme rural poverty, little emigration, and 100% concentration of development in Ulaan Bataar. That suggests massive movement to the capital.


    In all honesty, I don't see anything wrong with this. Leaving aside a lack of obvious sources of subsidies to rural Mongolia--Soviet subsidization of Mongolia was product of, among other things, Sino-Soviet tensions and an ideological commitment to support a long-standing Communist state and satellite--I agree with Treanor that expecting the bulk of Mongolians to remain employed in increasingly non-viable regions and lifestyles is unacceptable.

    Mongolia has to change if Mongolians are to enjoy acceptable standards of living. The job of the Mongolian government and its partners is to ensure that the transition is made as efficiently and quickly as possible. If not, mass emigration is always a possibility, whether to a China that's rapidly industrializing and is home to almost twice as many ethnic Mongols as in Mongolia, to a Russia that has long-standing connections with Mongolia and its own labour shortages, or to a South Korea that has cultivated close economic and demographic ties with Mongolia.

    Sunday, May 22, 2011

    Russian Demographics - Something Stirring in the East


    One of the reasons that I have always had a problem with Goldman Sachs' infamous notion of the BRIC economies was not the fact that it excluded other important economies such as e.g Chile or Indonesia, but rather that Brazil, India, Russia and China never belonged in the same group. The reason for this is largely because of demographics. Both Russia and China are consequently set to age much more rapidly than India and Brazil due to very rapid fertility transition in the 1990s. The demographic situation is especially dire in Russia which not only saw a dramatic and lingering decline in fertility in the 1990s but also saw a corresponding increase in mortality (aids and alcohol as big culprits).

    A recent piece by Carl Haub suggests however tha while doom and gloom used to be the prevailing tone on the state of Russian demographics recent trends suggest that this should change.

    Back in 2000, Russia achieved what Russians consider a dubious milestone, deaths (2,225,300) outnumbered births (1,266,800) by an astounding 958,500. The crude birth rate had sunk to 8.7 births per 1,000 population. Along with a crude death rate of 15.3, natural increase hit an all-time low of –6.6 per 1,000, or –0.7 percent rounded off. The total fertility rate (TFR) bottomed out at 1.195 children per woman. The crisis, as it was seen to be, was definitely noticed, but nothing really effective was done until 2007 when Vladimir Putin announced a baby bonus of the equivalent of $9,000 for second and further births. Putin has been an outspoken advocate for raising the birth rate and improving health conditions in order to avoid the consequences of sustained very low fertility. The program must have worked since births in 2007 jumped to 1,610,100 from 1,479,600 the previous year and have rising ever since. This is one of the very few “success stories” in the industrialized countries’ efforts to raise the birth rate.

    Together with the rest of Eastern Europe that was re-joined with the West after the fall of the Soviet Union, Russia experienced one of the most brutal fertility transitions ever seen. Indeed, history seems to have been extraordinarily cruel to many countries in Eastern Europe in handing them a second chance at the end of the 1980s just to take it away with the other hand as their demographic fundamentals collapsed. The birth dearth in the East even stretched into Eastern Germany where the total number of live births fell from 215700 to 88300 in the period 1988 to 1992.

    I have previously mused that perhaps those multinationals eager to expand eastwards would have to go all the way to Kamchatka to find qualified labour and perhaps even fail entirely and back in 2006, the only silver lining that the Economist's Berlin correspondent could find was how a residing population had led to a revival of wildlife with the lynx returning to Germany's Eastern borders.

    Perhaps though, it is time to put this discourse to rest?

    Russia in Transition



    From 1989 to 1999/2001 the total fertility rate in Russia fell from replacement levels to around 1.1/1.3 and notable effort [1] has been put into explaining why birth rates fell so much, so quickly.

    Grogan (2006) uses a household survey tracking data from 1994 to 2001 and finds that a large part of the decline in fertility among married couples can be attributed to the decline in household income in the same period. Grogan (2006) however also sheds light on other aspects of Russia's fertility during the Soviet era. In particular, the paper sets out to explain completed cohort fertility for women born between 1936 and 1961 and finds that women with higher education had considerably lower completed cohort fertility rates than their counterparts. This squares well with the notion of the quantity/quality tradeoff of fertility famously developed by Gary S Becker [2] and how parents substitute quantity for quality as their income levels rise (with education), but it comes with an important twist in the Russian case. Since female labour force participation was almost universal during the Soviet era and since women with less than higher education often earned the same (or more) than their better educated peers, Grogan (2006) seems to imply an inherent demand, by part of well educated women, for quality rather than quantity in their fertility decisions.

    The other driver of fertility decline in the form of the tempo effect is also present in Russia, but Grogan (2006) is skeptical as to its merits in explaining the sharp fall in fertility in the 1990s. It does appear to coincide with a change in attitude towards marriage and, specifically, births outside marriage, but from 1988 to 2000, mariage rates declined for the broad category of women (aged 15-44) as well as the share of total live births taking place outside marriage rose from about 14% to 26%.

    In essence, the tempo effect over the period in question is not linear and seems to neutralize itself over time.

    From 1989 to 1994, the share of births to mothers under 20 actually rose and then declined to just above 1989 levels in 2000. Not surprisingly, the share of total non marital live births among mothers aged less than 20 years rose sharply from 1989 to 2000. This suggests that the extent to which non-marital live births increased, it resulted in children being borne to young mothers. From a theoretical perspective, this is important in relation to how a change in the life course towards postponing marriage also leads to a postponement of childrearing. A norm of non-marriage child births may then serve to weigh against the tempo effect of fertility.

    This non-linearity of the tempo effect throughout what was essentially a sharp linear decline in fertility is interesting. The charts produced in Grogan (2006, p. 65 fig XI) clearly suggests that from 1989 to 1994 total live births for young mothers aged under 20 as well as those from 20-24 rose as share of overall birhts. This reverses somewhat in 1994 where live births for mothers aged 25-29 starts to increase as well as those aged 30-34. Yet Grogan (2006) notes that since there is no meaningful change in the fraction of total live births of "older" mothers in 2000 relative to 1989, the decline in fertility in Russia is not a postponement phenomenon.

    Brainerd (2006) builds on the points above by similarly latching on to the idea that the economic hardship bestowed on Russian citizens in the 1990s contributed to the decline in fertility. This suggests again a more permanent negative quantum effect at work rather than merely a postponement phenomenon. But the underlying causes of the fertility decline is cut very finely by Brainerd (2006). Notably, the paper argues for a pure negative income effect on birth rates and thus a reversal of the standard quantity-quality tradeoff as developed by Becker. The interesting thing here is that little evidence is found that general macroeconomic uncertainty (of the future) affect fertility even if women with more negative expectations of the future had a higher propensity of abortion.

    Quantitatively, Brainerd (2006) finds strong evidence for how marriage and a higher income per capita positively affects fertility using a fixed effect estimation with age specific fertility rates as dependent variable. Since marriage rates and income declined in the period 1989 to 1999, it leads to the conclusion that this caused the decline in fertility. I find this plausible, but would note that the estimation results suggests that underlying uncertainty of the future might still be affecting these results. For example, Brainerd (2006) shows how the effect of income on fertility is strongest for young mothers which indicates that permanent income may be a more useful proxy for linking fertility to income levels than the traditional method of using fluctuations in current income. It also sugggests that the income effect might be lower over time in the aggregate if we assume a general process of postponement, but this is dubious in Russia's case following Grogan (2006).

    And now lets go make some kids ... ?

    In general, the tendency of non-marital births is interesting to dwell on and Perelli-Harris (2008) [3] draws a sharp distinction between two reasons to explain it. The first relates to the notion of the second demographic transition [4] which postulates that the extent to which non-marital births occur in stabile cohabitations, as e.g. in the Scandinavian countries, it reflects a change in value towards marriage and thus a change in the life course. Contrary to this stands evidence, largely from the US, that non-marital births are associated with much less stable unions and, generally, poorer levels of society.

    Not surprisingly, Perelli-Harris et al (2008) do not ascribe either of these explanations to the rise of non-marital fertility in Russia, but rather; a mixture of both. One important aspect here is the extent to which, after a non-marital contraception, women with higher education tend to enter into marriage with a much higher probability than women with lower education. But everyone will be able to find sources to support their argument with for example this article by Sergei V. Zakharov and Elena I. Ivanova arguing for a more traditional second demographic transition process in Russia.

    One overarching conclusion which emerges on the fertility decline in Russia is that it was not driven primarily by birth postponement but seems to have been pushed by a more lingering quantum effect. The more specific driving forces of this quantum effect is much more difficult to get a hold on, but from the perspective of the macroeconomist it appears as if Russia entered a sinister spiral of increasing mortality and declining fertility just as the economy was meant to rebuild and then later take off on the much hailed wave of convergence. In particular, it appears as if the general adverse economic environment in Russia in the 1990s may have caused fertility rates to "undershoot".

    Pro-natalism in Russia, Action and Reaction?

    While we may certainly look upon Russia's demographic experience as a frightening example of the effect of negative population momentum, it would be unfair to say that the Russian leadership has been sitting idle. In 2006, Vladimir Putin announced a number of pro-natalist initiatives targeted at reversing the the decline of Russia's population. The plan included longer maternity leave, increased child benefits and most notably a full USD 9000 payment to women opting to have a second child Brainerd (2006).

    In May 2009, president Medvedev arranged for eigth families to be courted at the Kremlin where they were awarded the Order of Parental Glory; the Levyokin family chosen to represent the Moscow region had, at the time, given birth to no less than 6 children.

    Getting his Priorities Straight

    The question is whether it has worked?

    According to Carl Haub it has (see above), and if this is indeed one of the few success stories of how ageing economies can reverse their birth rate, it is worth paying more than scant attention to. The data here is subject to some uncertainty, but following Haub's lead the total fertility rate in Russia stood at 1.54 in 2010 which is up from a low point of 1.2 in 2000. In addition, Haub notes an important distinction between rural and urban fertility rates with the former standing at 1.9 in 2010 and the latter at 1.42. This last point is difficult to underestimate since it shines a rather pessimistic initial light on the strides to increase fertility in Russia. In particular, it casts russia in a more classic emerging market context witha a very abrupt quantity/quality trade-off at work whereby especially urban fertililty undershoots significantly below the replacement level.

    Still, the aggregate picture is improving.

    (click on charts for better viewing)

    In the jargon of the profession we must now be seriously asking whether Russia is about to join the very few nations that has managed to break free of the fertility trap defined here as how total fertility rates often don't recover (or has not recovered yet!) once they fall below 1.5. The only two other countries which have seen their fertility levels rebound from below 1.5 are Denmark and France.

    I would happily announce that this is the case, but the plot is just about to thicken.

    On the positive side and given evidence from the academic literature that the tempo effect is not a relevant phenomenon in a Russian context, it stands to reason that this rebound can be interpreted as a real change in sentiment towards having children.

    Score one for Russia's pro-natalist policies then?

    To some extent though Carl Haub pours water on this idea noting that the second derivative of the fertility increase is falling which leads him to ponder whether the rise of Russian births is losing steam. This argument is taken further by Kumo (2010) [5] who suggests that not only did Russia's pro-natal policies not work in the first place, but also that the rise in the number of births can be attributed entirely to fluctuations in the number of women in their reproductive age. More importantly however, Kumo (2010) emphasizes the difficulties of micro managing fertility and specifically the issue of just how difficult it is to get a lasting impact on fertility from cash transfers. In short, empirical evidence shows that pro-natalist policies rarely have a permanent effect. This is even more likely to be significant in a Russian context as the fund set up to dole out money to fertile mothers expires in 2016.

    Ageing in Russia, Adjusting for Mortality

    To assume that the Russian government's attempt to push up fertility rates will have a lasting permanent effect is probably as dubious as assuming that it will have no effect at all. In addition, if Russia is serious about securing a future balanced population pyramid, what is to say that there won't be more initiatives?

    Still, it appears that just as Russia seem to be making strides in the fertility department, the appalling situation for adult male mortality continues to taint the overall picture. Here, the optimists will call foul play and point out how the main story on Russian demographics has recently been a co-movement of improving mortality and fertility rates. This may be true, but overall conditions are still poor.

    According to data from the World Bank only a mere 47.4% of a male cohort can expect to celebrate their 65th birthday which contrasts with 78.5% for women. On average (from 1998 to 2009) only 44.5% of a given male cohort could expect to reach 65 years.

    Despite the visible improvement since the mid 2000s, the evidence from a birds eye view has not changed. Male life expectancy seems to be mean reverting around 61 to 62 (at birth) and mortality for adult males exhibits an increasing trend. An afinity to Vodka and other spirits as well as too many cigarettes appear to be lingering killers. Recent research (2009) from the medicinal sciences using mortality patterns from Tomsk, Barnaul and Biysk suggests alcohol was a cause of more than half of all Russian deaths at ages 15-54 years.

    As a result, the natural increase is still negative as the up-tick in births has still not managed to pip the mortality rate here even if it seems a more lasting change may be underway here.

    A Rare Sight

    Regardless of the permanency of recent years' improvement in fertility Russia cannot escape a rapid process of ageing. More than anything, this is why I so ardently argue against lumping Russia together with India and Brazil or more specifically; in Russia there is no positive demographic dividend in sight; rather what we have is a negative one.

    Of course, we cannot simply assume that the Russian population will fall from here on as one would assume (and hope) that Russia manages to reverse the trend in mortality. What we can see however is that in terms of the prime age group (35-54), Russia is likely to have peaked already in 2004 even if the effect of the double hump is interesting to consider (a result of assuming a perpetually declining population).

    In addition, the process of ageing means that there is almost no chance of Russia being able to contribute to global rebalancing by sustainably running an external deficit. This is one of the single most important macroeconomic characteristics which suggests why we should not label Russia as an "emerging" economy. Russia and the CEE will instead be fighting to escape the mantle that they may just have grown old befor they made it to become rich.

    Especially the younger part of the labour force will invariably be subject to a swift decline and the composition of the labour force is crucial to consumption smoothing on the aggregate level and thus capital flows.

    However, the most important aspect in the context of ageing in Russia is to adjust for the continuing high mortality rate among men. In a recent piece in Sciencemag Warren C. Sanderson and Sergei Scherbov argue that we should rethink ageing given that as the world population ages so does the threshold at which we can consider a person (or population) to be "old".

    (...) as life expectancies increase and people remain healthy longer, measures based solely on fixed chronological ages can be misleading. Recently, we published aging forecasts for all countries based on new measures that account for changes in longevity (58). Here, we add new forecasts based on disability status. Both types of forecasts exhibit a slower pace of aging compared with the conventional ones.

    This makes perfectly good sense and governments around the world are busy pushing up retirement ages to reflect this, but does this apply in a Russian context? What good would it do to push up the retirement age in Russia if less than half of a male cohort makes it to 65? The principle applied by messieurs Scherbov and Sanderson cuts both ways and in Russia's case we must incorporate an additional accelerant in our analysis of ageing to account for the continuing high rate of mortality and indeed an effect which will take some decades to pass through the pyramid.

    Something Stirring in the East?

    The recent improvement in Russia's demographic indicators begs the question of whether the glass is half full or half empty. On the former I would note two things. Firstly, Russia has indeed seen a noticeable improvement in both fertility and mortality and it seems to have coincided with the government's strategic aim to actually do something about the country's decaying demographics. Secondly, I will salute the effort in itself. We can all probably agree that Russia has veered a little too much towards the way of authoritarianism under Putin, but whatever the underlying ambitions to push forward a positive population agenda I think it is extraordinarily important.

    On the latter however, I am still worried that the trend in mortality have not been reversed and that, if anything, the situation is improving all too slowly. I am open to a more positive spin, but the data and an, admittedly scant, look at the evidence gives little comfort. As a result, ageing in Russia will be much more acute and its effect will have a much larger and negative impact than if life expectancy was a steadily increasing function of time. Indeed, given the continuing poor state of especially male health in Russia it is questionable whether the measures above of "peak growth" apply at all.

    The most important feature of Russia's demographic rebound is its potential permanency and especially we should watch whether Russia manages to stay above a fertility rate of 1.5. If this turns out to be case, we could harbour a hope that not only lynxes but also a rejuvenated Russian population may be stirring in the East.

    ---

    * All photos in this essay are taken from Creative Commons License accounts at Flickr. Data for the charts are from the World Bank Database and US Census Bureau (long term population forecasts).

    [1] - In the following I will make extensive use of Louise Grogan (2006) - An Economic Examination of the Post-Transition Fertility Decline in Russia and Elizabeth Brainerd (2006) - Fertility in Transition: Understanding the Fertility Decline in Russia of the 1990s.

    [2] - Becker, G. (1960) - An economic analysis of fertility, In Demographic and Economic Change in Developed Countries. NBER: New York

    [3] - Perelli-Harris, Brienna and Gerber, Theodore P. (2008) Non-marital fertility in Russia: second demographic transition or low human capital? In, Population Association of America 2008 Annual Meeting, New Orleans, US 17 - 19 Apr 2008. , 33pp.

    [4] - The second demographic transition has many sources but these ones by Dirk J. van de Kaa are a good starting place.

    [5] - Kazuhiro Kumo (2010) - Explaining fertility trends in Russia, VOX EU

    Saturday, May 21, 2011

    Why are sex ratios at birth so male-biased in the South Caucasus?

    Earlier this month, my attention was caught by a Behind the Numbers blog post by Carl Haub which described how sex ratios in the countries of the South Caucasus--Georgia, Armenia, and Azerbaijan--were among the most unbalanced in the world.

    Son preference and sex-selective abortion in China and India have frequently been in the news but who would ever have thought of it in the Caucasus countries? But it’s there, in Armenia, Azerbaijan, and Georgia, three former Soviet Republics and the trends have been well analyzed in a recent paper.* The normal global biological sex ratio at birth (SRB) is 105 male births per 100 female. Where son preference is strong and the sex of a fetus is determined, there has been a rise in the abortion of female fetuses in recent years.

    The SRB in those countries began rising in the mid-1990s and, in 2009, stood at 114 in Armenia and 118 in Azerbaijan. Birth reporting in Georgia has been somewhat erratic but the pattern is similar if not quite as skewed upwards. Abortion had long been used as a form of birth control in the Soviet Union and the practice has continued in the Caucasus. A low fertility rate is often thought to create a particularly likely impetus for sex-selective abortion when son preference is strong. In all three countries, the SRB for third births was well above the total SRB. In Armenia, the SRB for third births was about 140 in 2000-2004; in Azerbaijan, it was about 150; and, in Georgia, it was about 145. Of course, third births are much less frequent with low birth rates. In recent years, the SRB does appear to have leveled off, but their SRB’s are now exceeded only by China’s.


    Such a ratio can't help things at all. Georgia that--as I noted here in August 2008--has been seeing massive emigration since the end of the Soviet Union, just like Armenia to its south and even Azerbaijan to its east. Georgia and Armenia have cohort fertility rates substantially below replacement levels, while Azerbaijan is getting there. If future cohorts of women come to be so outnumbered by men, the South Caucasus' demographic catastrophes can only worsen. As my co-blogger Scott just mentioned, gender balance can well be a competitive advantage in the economic sense, as human capital is found and used more equitably without misogynistic attitudes. Again, this is profoundly self-destructive.

    But why is this going on? The Social Sciences in the Caucasus blog notes that the shift in sex ratios only occurred after the collapse of the Soviet Union, when sex ratios were more normal, while commenters suggested that this shift is to be expected in economically straitened times in societies where abortion is an acceptable method of birth control and male children are valued more than female children. Sabina Kiryashova's IWPR article "Abortion Used for Sex Selection in Azerbaijan" describes this mentality.

    Such abortions have become increasingly common in Azerbaijan in recent years, particularly since the use of ultrasound to determine the sex of the baby has became widespread.

    Up to four of the eight women who come into gynaecologist Gulustan Aslanova’s clinic each day for abortions do so because they are expecting an unwanted girl.

    Research conducted by the Women’s Crisis Centre in a main Baku maternity hospital found that abortions carried out because of the sex of the child are currently third behind those done because of the family’s financial situation and ones for contraceptive purposes.

    Azad Isazade, a psychologist at the centre, said that in Azerbaijan, as in many eastern countries, preference has always been given to boys. That is rooted in the pre-Islamic, pagan period when men were needed for farming and newborn girls were buried alive.

    That no longer happens, but some families see the lack of a male heir as a tragedy or even a curse.


    This is fine, but why in the South Caucasus? In the online essay collection Watering the Neighbour's Garden: The Growing Demographic Female Deficit in Asia, the paper by Meslé et al. "A Sharp Increase in Sex Ratio at Birth in the Caucasus. Why? How?" (73-88) makes the point that the shift is unprecedented, occurring in the absence of any state coercion, without any parallels either elsewhere in the former Soviet Union or elsewshere in the Greater Middle East.

    It is true that the countries of the Caucasus are very small compared to some of their neighbours and it is easy to imagine that by decreasing the scale to internal administrative districts the same phenomenon could be observed in some regions of Russia, Iran, and Turkey that are close to the Caucasus, particularly among the Azeri in Iran, in the Caucasian republics of southern Russia, and in the eastern regions of Turkey that border Armenia. It is for this reason that we have attempted to collect data by local districts in the countries of the Caucasus and also in the three closest large countries, as well as in Syria. [. . .]


    When this data is assembled on a detailed map of the region, two observations reinforce the contrast suggested by national indicators. On the one hand, within each of the three countries of the Caucasus, high sex ratios at birth are almost systematically observed in all the regions. The only notable exceptions concern two groups of Azeri rayons concentrated in the North on the border with Russia and in the South on the border with Iran. On the other hand, no such ratios are observed in the administrative districts nearest to the countries of the Caucasus, either in Russia, Iran, Turkey, or Syria. In Russia, the federated republics of the northern Caucasus such as Karachay-Cherkessia, Kabardino-Balkaria, Ingushetia, and Dagestan are all between 104 and 107. Only North Ossetia is slightly over 108. In Iran all the northern provinces and notably the four sharing a border with Azerbaijan (Gilan, Ardebil, eastern Azerbaijan, western Azerbaijan) are below 107. Finally, in Turkey there is a zone around the Van Lake, which is slightly over 107, but all the provinces along the border with Georgia and Armenia are between 104 and 107.


    The number of abortions a woman has rises if the first children born are female, but still. Why in the South Caucasus?

    Thursday, May 19, 2011

    US gender balance may be a significant competitive advantage

    Demographic gender imbalance in favor of males due to prenatal gender selection and misogynism in India and China will probably handicap economic and social progress in these countries relative to the United States. The US will have a long term economic advantage due to its greater utilization of its human capital and the future social disruption likely in countries with large gender imbalances.

    Despite rapid growth, India lets its girls die - World news - South and Central Asia - msnbc.com

    "Early results show India has 914 girls under age 6 for every 1,000 boys. A decade ago, many were horrified when the ratio was 927 to 1,000....In Morena, a sun-baked, largely rural district in the heart of India, the numbers are especially grim. This census showed that only 825 girls for every 1,000 boys in the district made it to their sixth birthdays, down from an already troubling 829 a decade ago."

    The article discusses the social reasons for this result. It is likely that there will be serious social disruption in the future due to this.

    NationMaster provides data on the gender ratio of the cohort under 15 years that shows China is worse than India:

    China: 1.17 male(s)/female 2011

    India: 1.13 male(s)/female 2011

    Tuesday, May 17, 2011

    A brief note on Lithuania's depopulation

    Thanks have to go to Edward Hugh, via his Facebook presence, for sharing the news that the Lithuanian population has fallen catastrophically, to a much greater extent than earlier predictions had suggested. I'm honestly quite surprised by the scope of the decline.

    The Lithuania prime minister bemoaned his shrinking population Monday, as the results of a census revealed it has fallen 10 percent in the past decade.

    "Unfortunately, we must admit that Lithuania is not only an emigrating nation but also a nation that is dying out," Prime Minister Andrius Kubilius said.

    Lithuania's statistics agency said the population is now 3.05 million, down from over 3.4 million in 2001. Twenty years ago, when the country split from the Soviet Union, the country had 3.7 million people.

    Results of the census - conducted from March to May - confirmed that rapid emigration and a falling birth rate continue to erode the country's demographics despite membership in the European Union and improvements to the quality of life over the past 10 years.

    Large communities of Lithuanians have sprouted in places such as Ireland and England during this time.

    "We have lost 700,000 (people) in 20 years - this is a lot for such a small nation," said Romas Lazutka, a sociology professor at Vilnius University.

    "The main driving force behind emigration is outdated economic and social policies. Those who do not have a job or cannot make ends meet, even while working, leave this country," he said.


    Agence France-Presse went into more detail, describing the economic consequences.

    The slump is a concern for the country's centre-right government, which is at the helm of a tough austerity drive brought in after Lithuania plunged into one of the deepest recessions in the European Union in 2009.

    The economy has gradually emerged from the doldrums after the sharp crisis.

    "It is very important that those people who left Lithuania in large numbers keep ties with Lithuania and see opportunities to come back," Prime Minister Andrius Kubilius told reporters on Monday. Analysts warn that the population decline may impede the economic recovery.

    "The population decrease has been very rapid due to negative demographic trends, but emigration is probably even more important," Rimantas Rudzkis, chief analyst at the bank DnB Nord in Lithuania, told AFP.

    "Such a decline will have negative consequences. Lithuania will be lacking workforce and that may deter investment," he added. The emigration trend started after EU entry in 2004.

    [. . .]

    Few have returned, despite the host countries' own economic slumps. Statistics showed that 83,500 people left Lithuania in 2010, mostly to Britain and Ireland.

    Officials have said that may include people who left earlier but made the move official to avoid the Lithuanian government's new drive to collect taxes from undeclared emigrants.


    Economic convergence between poor and rich countries--within the European Union, without--isn't inevitable. The sort of mass emigration that not only sharply reduces the size of the workforce but worsens problems of population aging can certainly harm a country's prospects.

    Wednesday, May 11, 2011

    "The Baby Bust of 2009 in the United States"

    Carl Haub at the Population Reference Bureau blog has pointed out that in 2009--likely because of the economic calamities of that year--has fallen significantly.

    The United States, long a target of envy by many European and East Asian industrialized countries for its “high” birth rate, has recently seen its own birth rate decline. The recession has been blamed and would seem to be a no-brainer as a cause, but one cannot assume cause and effect from births. Few demographic events cause as much interest as peaks and troughs in the birth rate. Nine months after events such as large power failures, snowy winters, and even the soccer World Cup, we get questions at PRB asking if there were a spike in births during these things. Couples spent more time at home, in the dark…you get the idea.

    The U.S. National Center for Health Statistics (NCHS) has released an interesting brief on who may have been most affected by the recent recession and where they might be. Births fell from an all-time high in 2007 of 4,316,233 to 4,131,019 in 2009, a decline of 4 percent. But, of course, rates tell the story better than just numbers. The numbers, however, can be very meaningful if you happen to be in the baby formula or diaper service businesses.

    In 2007, there were 69.5 births per 1,000 women ages 15 to 44 compared with 66.7 in 2009. This is referred to by NCHS as the fertility rate, but is also often called the general fertility rate (GFR). By age, the largest decline was among women ages 20 to 24. Childbearing fell from 106.3 births pr 1,000 women in that group in 2007 to 96.3 in 2009, a 9 percent decline. Does that reflect a lack of confidence in the outlook for jobs among women or couples “just starting out?” Fertility among the 25-to-29 age group, the next older, fell by 6 percent. That group, which has had the highest rate in recent years, dropped from 117.5 in 2007 to 110.5 in 2009. The decline was far less, only 2 percent among women in their 30s and even rose among women in their 40s, although latter group has much lower rates.


    Interestingly, fertility rates have fallen significantly more among most non-white groups than among whites: 3% for non-Hispanic whites between 2007 and 2009, 4% for non-Hispanic blacks, 9% for Hispanics, 3% for American Indians and Alaskan natives, and 4% for Asian and Pacific islanders. This, in turn, has had a significant effect on fertility in states where non-white groups are relatively more prominent than the American average. I might speculate that this occurred because rates of poverty and economic uncertainty are higher among groups other than non-Hispanic whites.

    It will be interesting to see if this effect is lasting. If the structural changes in the American economy--greater income inequality, higher rates of poverty, and so on--aren't reversed, then a long-term decline in fertility rates caused by the inability of parents to support the relatively large families common in the United States seems believable.

    Tuesday, May 10, 2011

    "Holiday in Fukushima: To the zone of exclusion"

    I wanted to point our readers to weblog Spike Japan's latest post, a report by the blogger of the visit he paid to Japan's Fukushima Prefecture, a district that not only suffered heavy damage from March's tsunami but is also having to cope with the damage and--perhaps worse--dangerous reputation from the Fukushima nuclear complex's meltdowns.

    Some lazy hacks have taken to calling everywhere along the Fukushima coast from Minami Soma in the north to Iwaki in the south “nuclear ghost towns”. They’re not, but Hirono is, and I hope never to see another one in my life. No rampaging steers running wild here, no cows lying dying in barns, no dogs turning feral as there are in the 20km zone of exclusion; this had been an orderly departure, leaving in its trace only silences and absences—of cars from garageless driveways, of washing from steel clothes poles, of people from the tidy sidewalks. Every roadside enterprise, from humble ramen stand to ubiquitous convenience store, was locked and deserted. What, I wondered, would an observer catapulted forward in time from two months ago (has it really only been two months?) make of this post-apocalyptic scene.

    As so often, it was the signs that were most poignant. One on a hillside proclaimed that Hirono was the town, by virtue of its southerly location, that announced the coming of spring to the north-eastern Tohoku region, while another called for support for the women’s soccer club Mareeze (yes, it’s a portmanteau of “marine” and “breeze”) of Fukushima Daiichi operator TEPCO—like the string of four pure-play nuclear seaside towns to the north, Hirono is a TEPCO company town, thanks to its mixed gas and coal thermal power plant.

    [. . .]

    On returning home, I discovered the depth of the Faustian compacts in which these Fukushima seashore towns had engaged with TEPCO. While the prefectural average per capita income in the year to end-March 2009, the latest year for which data are available (Japanese-only link to a mine of fascinating Fukushima factoids here), was around Y2.75mn ($34,000 at the current rate), it was Y5.65mn (over $70,000) in Hirono, by far the highest in Fukushima, and Y4.85mn (over $60,000) in Okuma, home to most of Fukushima Daiichi. In the sublimely implausible event that Hirono and its 4,500-odd inhabitants were to declare independence, it would rank somewhere above Switzerland and below Norway as one of the nominally half-dozen wealthiest nations on the planet. Remember that the next time you fork over for your electricity bills, Tokyoites.

    The nuclear shoreline is also impervious, it would seem, to the vicissitudes of recession. While the rest of the prefecture—and the rest of the world—were left reeling in wake of the global financial crisis, the Soma district (essentially the Fukushima coast minus Iwaki) was clocking up gross product (i.e., GDP at a local level) growth of 6.4%, a figure that would not bring dishonour to the average emerging economy.


    And all this has been wrecked.

    This essay, photographs included, provides an invaluable on-the-ground description of the Tōhoku region that had been suffering from economic malaise and out-migration even before the disaster.

    Readers might also be interested in the blog's latest post, "Tobacco economics", which takes a look at the ongoing collapse of Fukushima's tobacco farming. The Japanese tobacco monopoly seems to be in the process--again, aided by Fukushima's reputational damage--of shutting down an agriculture that produces tobacco leaves at ten times the price of southern Africa.

    A fresh look at US population trends

    The United States Census Bureau completed its constitutionally mandated population count last year and has been gradually releasing data and analysis of the results. The most important information, that of state population totals, required to provide the basis for re-apportionment of seats in the US Congress as required by the Constitution was released first.

    In March of this year the Census released Population Distribution and Change: 2000 to 2010, a report with more detailed data and some high-level analysis. The results of the count show significant changes taking place.

    The decade ending in 2010 saw the US population increase by 9.7 percent, the lowest rate of increase since the Depression decade of the 1930's. It seems quite possible that the current decade may see even slower growth. Another US government agency, the National Center for Health Statistics reported in Births, Marriages, Divorces, and Deaths: Provisional Data for 2009, that births in the US declined in 2008 and 2009 (the most recent data available). Well known analyst Calculated Risk noted here that "it is common for births to slow or decline during tough economic times in the U.S. - and that appears to be happening now." Combining this trend with the fact that a large segment of the "baby boom" population will reach 65 years of age by 2020 (with the associated higher mortality due to age-associated factors) provides grounds for a prediction of slower population growth.

    The Census divides the US into four regions for comparison purposes: Northeast, Midwest, South, and West. By 2010 the population of the West region (at 71.9 million) surpassed that of the Midwest region (at 66.9 million), of course for the first time. Political power will shift due to the transfer of House seats out of states in the Midwest; and in economic terms the Midwest has weakened significantly.

    Population growth was heavily concentrated; only six states accounted for more than half of the total. The six are Texas, California, Florida, Georgia, North Carolina, and Arizona. The highlighted states also were key components of the national housing boom and crash. Overall, the national population is concentrated in metropolitan areas with roughly 84% of residents living in designated metro areas. In addition, 1 of every ten residents lived in either the New York or Los Angeles metro areas.

    By contrast, several fringe regions continued to lose population or remained stagnant, including an area of the Appalachians, the Great Plains, the Mississippi Delta and northern border areas. This shift from outlying areas to core regions indicates that urbanization is continuing.

    Slow, concentrated growth with continued migration south and west seems to be the outlook until the next census is conducted.

    Sunday, May 08, 2011

    How demographics shaped the 2011 Canadian federal election

    My previous post here touched upon the remarkable consequences of the Canadian federal election held last Monday. The New Democratic Party made massive breakthroughs, especially in Québec, to become the official opposition, while the separatist Bloc Québécois was devastated (from 47 to 4 seats) and the Liberal Party was more than halved. Maps do a good job of illustrating this transformation. Copied from here, this map shows the distribution of seats following the 2008 federal election in Canada. Blue is for the Conservatives, red for the Liberals, orange for the New Democratic Party, and teal for the Bloc Québécois.

    Results of 2008 federal election, Canada

    Here is a map showing the results of the 2011 election.

    Results of 2011 federal election, Canada

    Population changes played a major role in this outcome. Take voter turnout, which rose 2.3% from 59.1% in 2008 to 61.4% this time. The low voter turnout is a major concern for Canadians, but so far no one has come up with a solution. Disenchantment with politics is pervasive.

    The NDP made massive gains, going from 36 to 108 seats (by comparison, the previous peak was 43 seats in 1988). Many established parties were displaced. My own downtown Toronto riding of Davenport, represented by Liberal parliamentarians since 1962, switched massively over to the New Democratic Party, the NDP candidate outpolling the Liberal incumbent by 2:1. Davenport, with a large Portuguese-Canadian community, has traditionally voted Liberal. Why the change?

    The Davenport riding is in transition. As Portuguese-Canadians become prosperous and move to the suburbs, the gentrification that has taken over The Annex neighbourhood is spreading west into Davenport. Since I've moved here, a townhouse complex has been built on the other side of the street, while an artists' community centre is by the laundry and any number of old storefronts about being converted into residential units.

    Store made house (3)

    It may not be that whereas Liberal Mario Silva made his name partly through his aid to immigrants, the NDP's Andrew Cash first came to attention as a punk rocker, then as a journalist for the alternative press. And yes, you probably won't be surprised to hear that I voted NDP myself. I was frustrated with the Liberal Party and wanted a change. Clearly, I was not alone.

    One very notable instance of demographics on the election--here, the tendency of Québec to vote as a bloc and to swing massively from one party to another was the massive success enjoyed by the NDP, which crushed the Bloc Québécois and won 58 of the province's 75 seats. By way of comparison, before the election the NDP had only one seat in Québec, in Montréal. Québécois seem to have tired of the Bloc, and voted for the only other party in massive numbers. The NDP didn't have to try: one candidate, a unilingual Anglophone who spent a good part of the campaign in Las Vegas on vacation and hasn't even visited her rising beat a 13 years' incumbent.

    What's very interesting about this is that the NDP is now the dominant party of French Canada, not just Québec. I wrote earlier here about the substantial assimilation facing most Francophone populations outside of Québec, but for the time being there are close to a million Francophones in adjacent Ontario and New Brunswick. These populations also voted substantially for the NDP. Taken from Wikimedia here, the below is an edited map showing electoral outcomes in central and eastern Canada, with maps of outcomes in the three Canadian cities with the largest Francophone populations (in descending order, Montréal, Québec City, and Ottawa-Gatineau) added.

    2011 election results, eastern Canada

    Even Francophone neighbourhoods in Ottawa voted for the NDP.

    What this means, given the weakness of the governing Conservatives in the province, is that the NDP is the only political party to have large numbers of members in both French Canada and English Canada. If the party can assimilate its new huge cohort, this straddling the language frontier could well make it an inevitable party of government.

    Finally, here in Toronto the Conservatives pierced what was once a Liberal stronghold and split the city with the NDP. The first map below shows the situation before the election; the second shows the situation after the election.

    20081014_GTA_results[1]

    201152-2011-GTA-ridings[1]

    This division corresponds substantially to the geographical divisions manifested in Toronto's recent mayoral election, where the left-wing George Smitherman held the downtown but the right-wing Rob Ford carried the day with the outer peripheries of the city: as described by the "Three Torontos" paradigm, poorer, more multiethnic, and with more problems about a left-leaning city government that didn't seem to do much for them. These areas traditionally voted Liberal federally--for instance, the Portuguese Canadians--because the Liberal Party is the one responsible for the lifting of discriminatory immigration legislation and the introduction of multicultural policies in the 1970s. As immigrant communities have become larger and more stable, and the Conservative Party has made efforts to appeal through social and economic conservatism, many of their members have switched allegiances to the Conservative Party. While a good marker of assimilation, this is not good for the Liberals.

    What does all this mean? Considering the Conservative Party's gains in Ontario, and the NDP's massive success in Québec and substantial improvement in the rest of Canada, the future of the Liberal Party as a potential party of government is in doubt. Without any particularly coherent geographic or communal base and no longer the party of central Canada's cities (and patches elsewhere), how can it succeed? As for Québec, the effective destruction of the Bloc Québécois and the defection of Québécois voters en masse to the federalist NDP may signal significant reverses for the local separatist movement.

    It is exciting times for Canadian politics. Demographics both established and changing can be thanked for a good chunk of it.

    Thursday, May 05, 2011

    After the election, more on Canada's census

    In the aftermath of Monday's federal election in Canada, which not only saw the creation of a Conservative majority in the federal parliament but the emergence of the social-democratic as the Official Opposition (thanks substantially to the near-total collapse of Québec's federal separatist Bloc Québécois party) and the reduction of the Liberal Party to third place, many questions about the future policies of the new secure federal government have been raised. Some of these have to do with Canada's census, which (as I wrote here in February) no longer includes the more detailed long-form questionnaire as mandatory, threatening the utility of the census as a planning tool and the comparability of data sets. One junior Conservative parliament talked about getting rid of the census altogether; thankfully, if not altogether reassuringly, the Prime Minister's office contradicted the MP.

    The Prime Minister’s Office said contrary to what the St. Catharines Tory MP told his local paper, the 10-question mandatory survey is not headed for the scrap heap.

    “The short form census will remain as is,” PMO spokesman Dimitri Soudas said.

    That may not be the end of the matter, however.

    Statistics Canada itself is investigating whether to ditch or augment the mandatory short-form census in favour of alternatives such as data mining, chief statistician Wayne Smith said in a February interview with The Globe and Mail.

    It was the Harper government who requested this review.

    The Conservatives, who last year scrapped the mandatory long-form census on the grounds it was wrong to coerce Canadians into answering intrusive questions, have asked Statistics Canada to rethink the way it collects population data.

    Both the 50-question long-form survey – which is now optional – and the short-form questionnaire that collects basic data are used by researchers, policymakers, economists and others to get a richly-detailed picture of Canada. These users complained bitterly about Ottawa’s decision to make the longer survey voluntary, warning it will erode the quality of data gathered.

    [. . .]

    After the PMO spoke out to quell confusion over the matter, the Tory MP later talked to the paper and said he’d been “unclear” in his comments.

    He said he was merely talking about already-announced plans to remove threat of jail time for refusing to fill out the short-form census.

    Mr. Dykstra said the short form and its basic questions would still remain mandatory.

    “What I should have said was we were going to reduce the penalties. We couldn't because we had an election, but we will be introducing that at some point in the new parliament,” Mr. Dykstra told the Standard.


    There's a minor movement afoot in Canada for census-takers to request that the long-form census be sent to them. Laudable, I guess, but the literal self-selectivity of this isn't going to help the census' accuracy.

    I like what Frances Woolley wrote on a Globe and Mail blog about the importance of the census, not just as a planning tool but as a historical document.

    In 1911, census data was collected by enumerators, who walked from house to house, knocking on each door. For each person in Canada, they collected just one line of information. Name, address, sex, relationship to household head. Date and place of birth. Racial or tribal origin, nationality and religion. Occupation, trade or means of living. Earnings, education, language and -- for reasons that no doubt seemed important at the time -- insurance coverage.

    The Canadian and provincial censuses up to 1911 are now available on-line, as are censuses from the U.K., the U.S. and other countries.

    A census form can bring your own personal history to life. No other kind of record will tell you how your family lived -- did the kids stay at home until they were thirty? Were your ancestors servants, or did they have servants? Were they rich or poor, educated or not?

    That’s the reason I happily filled out my census form and answered yes to the question about releasing personal information. It’s a gift to my great-grandchildren.

    But will my great grand-children care?

    Most of the questions on the 1911 census are now asked on the new, voluntary National Household Survey. It will be sent out to about 4.5 million households in about four weeks time. If you’re one of those lucky households, you have a chance to leave a footprint, a time capsule, for future generations to find.

    A typical Canadian, however, only has an opportunity to fill out the mandatory, short form census. It starts out by collecting the same information as the 1911 census – name, address, sex, relationship to “person 1” (households do not have heads anymore), age and date of birth.

    The government needs this information to function effectively. For example, federal–provincial transfers for health care, education, and so on are based on the number of people in each province. The census is the way governments know how many people there are. Or, to take another example, our democracy is based on representation by population. Census information is used to draw up electoral boundaries between ridings. Basic demographic information is fundamental to the functioning of democracy.


    How can people in a society know what's going on without knowing what the society actually includes?