Friday, April 29, 2011

What last year's natural increase in Estonia means

Thanks to Facebook's Urmas for letting me know the surprising news that Estonia has become--so far as I know--the first former Soviet republic in Europe (Azerbaijan and the rest of the Caucasus, perhaps, excluded--thanks for the correction, Anatoly) to experience an excess of births over deaths since the fall of the Soviet Union. Yes, there seems to have been some improvement--I posted recently about Russia, after all--but it still stands out remarkably. Compare neighbouring Latvia's ongoing collapse.

Births, deaths and natural increase, 1990–2010 (Estonia)

According to the revised data of Statistics Estonia, in 2010 35 people more were born than died. The population of Estonia was 1,340,194 on 1 January 2011.

15,825 people were born and 15,790 people died in 2010. The number of births exceeded the number of deaths last 21 years ago in 1990.

In 2010, 62 children more were born than a year earlier but the number of births was still smaller by about 200 than the last decade’s record in 2008 when more than 16,000 children were born.

On the contrary the number of deaths has been rapidly decreasing during the last three years and in 2010 291 people less died than a year earlier. Thus the positive natural increase was mainly achieved due to the decrease in the number of deaths.

617,757 males and 722,437 females lived in Estonia at the beginning of 2011. Population growth continued due to the natural increase in Harju and Tartu counties.

Harju County's is Estonia's most populous, its 524 thousand people amounting to 39.2% of the Estonian population, while Tartu County is Estonia's third most populous, home to just shy of 150 thousand people and 11.2% of the total population.

It's noteworthy that Idu-Viru County, Estonia's third by population and located in the northeast of the country around the city of Narva on the Russian frontier, does not figure in this; so far as I know, Ida-Viru is continuing to experience continued natural decrease.

One of the most important facts about Estonian demographics is the ethnic cleavage between ethnic Estonians (now roughly 70% of the total, up from 60% in 1990) and Russophones, largely descended from Soviet-era migrants attracted to industrial jobs and a higher standard of living than they could find in other Soviet republics like Russia, Ukraine, and Belarus. See this 2007 post from Itching for Eestimaa for a decent overview of the situation. The two populations behave differently in many ways, including in demographic patterns. The three most populous counties fit into three different categories, Harju being roughly 60% ethnically Estonian by population, Tartu ~85%, but Ida-Viru 20%. (Somewhat ethnically mixed up to the Soviet era, post-war displacement of ethnic Estonians and mass immigration to the northeast's industrial centres created the only one of Estonia's fifteen counties with a non-Estonian majority population.)

I've a draft of a much longer post hidden in the archives. Briefly, I would like to say that this all fits with my noting of with ethnic Estonians now inclining to a relatively Nordic pattern (relatively high fertility, significant postponement of births, very significant extramarital fertility) and Russophones behaving in the opposite manner. See Puur et al.'s brief abstract "Fertility patterns among foreign-origin population: the evidence from Estonia" for an outline.

The analysis reveals a remarkably strong contribution of foreign-origin population to the total number of birth. In the 1970s and 1980s it accounted for more than a third of births registered in the country, leaving a long-term imprint on the ethnic and linguistic composition of the population. The cessation of massive inflow after the turn of the 1990s has somewhat reduced the proportion of births to women of immigrant background. In the recent years it has accounted for less than 30% of the total.

The comparison of completed cohort fertility rates allows to distinguish between two different patterns among women of immigrant origin born during the 20th century. In older cohorts, born in the first quarter of the century, the foreign origin population shows noticeably higher fertility, reflecting the later onset of fertility transition in the regions from which the immigrants originate. The progression of fertility transition in the latter resulted in the continuous decline and the convergence of levels with the native population in the birth cohorts of the late 1920s. However, the state of convergence proved temporary and in the generations born in the 1930s and later, the levels diverged again with foreign origin women having a systematically lower fertility compared to their native counterparts.

The examination of parity progression ratios and the ultimate parity distribution reveals that the lower completed fertility stems mainly from the less frequent progression to a second, and in particular, to a third birth among the foreign origin population. Compared to the native population, the corresponding measures have been twice or even more than twice lower, demonstrating the largest difference across parity distribution. On the other hand, the proportion of women with one-child has been markedly higher among immigrants. At the same time it is interesting to note that childbearing has typically occurred at an earlier age among the foreign-origin population.

Meanwhile, as Lars Agnarson notes in his paper "Estonia’s health geography: West versus east – an ethnic approach", partly because of Russophone concentration in Soviet-era industry--exactly the sorts of industries which got hit by the post-Communist transition--Russophones evidence significantly higher mortality than their ethnic Estonian co-residents, mortality rates apparently rising in proportion to the homogeneity of Russophone communities. Between 2000 and 2009 the populations of ethnic Estonians and ethnic Russians each decreased by roughly nine thousand, but the ethnic Russian population is less than 38% of the size of the ethnic Estonian.

What does this imply? For Estonia as a whole, last year's natural increase might indicate that Estonia is moving from a post-Communist demographic system (low fertility, high mortality) to a Nordic one characterized by relatively high subreplacement fertility, low mortality, and a post-modern approach to family relationships. The question of emigration is noteworthy, and over the 2000-2008 period Estonia does seem to have seen the emigration of 13 thousand people. The volume of emigration has been significantly less than in Latvia or Lithuania, however, and much of it has been directed towards a Finland that is both geographically and culturally quite close to Estonia, Helsinki and Tallinn being separated by no more than a ferry ride. Much of what Estonian emigration has developed may be more temporary in nature. Estonian-Finnish migration is even bidirectional: Estonia can offer job opportunities for Finnish workers, too, and the lower cost of living in Estonia compared to Finland may be a long-run advantage.

It's also worth noting that a relatively less dire demographic situation than that of neighbouring Latvia may well provide Estonia with if not advantages, fewer disadvantages--including economic ones--over its southern neighbour, and others, too.

Within Estonia, if the past two years' sustained difference between a relatively high fertility/low mortality ethnic Estonian population and a relatively low fertility/high mortality Russophone population remains, then the continued shrinkage and aging of the Russophone population is inevitable. This will have significant effects on everything from the spatial distribution of the Estonian population (what will happen in the northeast) and the futures of economic sectors depending heavily on Russophone labour to the balance of political power in Estonia and Estonian relations with its neighbours.

Expect more later, please; consider this a taster.

Thursday, April 28, 2011

On Doug Saunders' Donner Prize win for Arrival City

I wanted to share with our readers that friend of the blog Doug Saunders' book Arrival City, a book exploring the roles of cities in global migrations that I blogged about back in September, won Canada's prestigious Donner Prize literary award.

Globe and Mail European bureau chief Doug Saunders has won the $35,000 Donner Prize for Arrival City: The Final Migration and Our Next World, which the jury for the annual award for public-policy writing declared is “a work that analyzes a global trend that we ignore at our peril.”

Arrival City, Saunders’s first book, highlights global migration as a defining force of 21st-century life and predicts that the volatile communities it creates will become the sites of either great creativity or violence. “This book enriches our understanding of how to interpret the dynamics of migrating people and the ecosystems that best support them,” the Donner jury said. “It deserves a broad readership and is a valuable public-policy contribution.”

Saunders’s book has drawn favourable reviews in Canada and abroad. “Serious, mightily researched, lofty and humane, Arrival City is packed with salient detail and could hardly be more timely,” a New York Times reviewer wrote, going on to compare the book to Jane Jacobs’s influential The Death and Life of American Cities.

In a brief interview, Saunders called winning the Donner “a huge and very flattering recognition for a book that doesn’t fit the usual mould of public-policy books. I’m very impressed that the judges were even willing to consider it.”

Established in 1998, the Donner Prize is awarded in recognition of excellence in Canadian public-policy thinking, writing and research. Previous winners include economist Eric Helleiner, pollster Michael Adams and political scientist Tom Flanagan.

Congratulations! And yes, I strongly recommend you pick up the book, too.

Four links for you

I'd like to apologize for not posting in a while: offline life intervened. I have accumulated four links that I think might be of interest to you all.

  • Over at Far Outliers, Joel quotes from Ryszard Kapuscinski's 2002 The Shadow of the Sun, describing how Tanzanian Sango-speakers (from central Tanzania) found ethnolinguistic kin in coastal Dar es Salaam.

  • Following a ball of yarn, they will finally arrive at the house of a countryman. The neighborhood is called Kariakoo, and its layout is more or less planned—straight, perpendicularly aligned sandy streets. The construction is monotonous and schematic. The so-called swahili houses predominate, a type of Soviet-style housing—a single one-storied building with eight to twelve rooms, one family in each. The kitchen is communal, as are the toilet and the washing machine. Each dwelling is unbelievably cramped, because families here have many children, each home being in effect a kindergarten. The whole family sleeps together on the clay floor covered with thin raffia matting.

    Arriving within earshot of such a house, Edu and his kinsmen stop and call out: “Hodi!” It means, in effect: “May I come in?” In these neighborhoods the doors are always open, if they exist at all, but one cannot just walk in without asking, so this “Hodi!” can be heard from quite a distance. If someone is inside, he answers, “Karibu!” This means: “Please come in. Greetings.” And Edu walks in.

    Now begins the interminable litany of greetings. It is simultaneously a period of reconnaissance: both sides are trying to establish their precise degree of kinship. Concentrated and serious, they enter the primevally thick and tangled forest of genealogical trees that is each clan and tribal community. It is impossible for an outsider to make heads or tails of it, but for Edu and his companions, this is a critical moment of the meeting. A close cousin can be a great help, whereas a distant one—significantly less so. But even in this second instance, they will not go away empty-handed. Without a doubt, they will find a corner under the roof here. There will always be a little room for them on the floor—an important consideration, since despite the warm climate it is difficult to sleep outside, in the yard, where one is tormented by mosquitoes, by spiders, earwigs, and various other tropical insects.

  • Global Voices, meanwhile, features the story of Yao Jiaxin in China, who seems to have murdered a rural migrant, a woman named Zhang Miao because she--as ruralites would, Zhang said in his defense--would try to blackmail him after he hit her with his car. Interesting fault lines, here, and abundant quotations from an enraged Chinese blogosphere.

  • Zhang Miao, how silly you were?

    When Yao Jaxin checked on your injury, why did you stare at him. You probably thought that he was going to take you to the hospital.

    You were wrong. He was checking if you were rural peasant or city dweller.

    Zhang Miao, how silly that you let him know that you were from the rural area? In Yao Jaxin's eyes, a rural peasant's life is worthless. Death would cost less than injury. But he forgot to crush you with his car. Instead he stabbed deep into your body with his knife, eight times.

    Zhang Miao, how silly you were?

    Even though Yao Jaxin checked on your injury with a knife,

    Even though it took him awhile to figure out your rural background,

    The time for a premeditated murder was too short. They called this murder out of rage, rather than intentional murder.

    We are all too ignorant to hear of such a term before.

  • At the Population Reference Bureau's blog, Carl Haub notes the continued rise in Russian fertility rates. Along with declining deaths, this has significantly reduced the rate of natural decrease. The big problems is that the cohorts of women born in the 1990s--a time of a markedly low birth rate--are coming to childbearing years and the situation will become more difficult, and that immigration is becoming problematic.

  • The social scientists have offered their opinion that the only way to avoid Russia’s population dropping below 140 million is to improve health conditions and lengthen life. They have recognized the relatively small number of women who will soon enter childbearing age and also warned that the supply of Russian-speaking migrants from other former USSR Republics is dwindling, particularly since many prefer to seek higher-paying work in the European Union. Migrants from former Central Asian republics of do not have work qualifications. The scientists have recommended that greater efforts be made to assimilate those migrants and prepare them for more worthwhile participation in the labor force.

  • Finally, at the unsurprisingly Toronto-centered blog Torontoist Kevin Plummer writes about how Italian migrants in post-Second World War Toronto sacrificed in order to own their own, constantly improved, houses.

  • By the mid-1950s, the annual influx had grown to about 20,000. Over that period of rapid growth, Italian men comprised 8% of the workforce in Metro Toronto. They worked unskilled or semi-skilled jobs as construction workers or general labourers, digging sewers. With long hours in abysmal working conditions, it was gruelling work at the kinds of jobs that, one Italian man told Iacovetta, "they give only to the immigrants."

    Some newcomers benefited from a network of kinfolk and paesani, and could board with earlier immigrants from their family or village. Some employers, like small construction companies, placed employees in company-owned shared accommodation. Other men lodged with other male newcomers in boarding houses run by Italian families or enterprising newcomers. In any of these shared accommodations, quarters might be tight, with several men to a room. The idea of large groups of foreigners living together also drew the attention of well-meaning outsiders who suggested that the Italians were being exploited as much in housing arrangements as in their work conditions. It occurred to few experts that, as Robert F. Harney in an essay in the anthology, The Canadian City (Carleton University Press, 1984), Italian "sojourners might prefer such a boarding system was lost in a haze of moral outrage."

    Almost always, boarders were from the same region of Italy as their hosts—Calabrese boarded with Calabrese, Abruzzese with Abruzzese—so these housing arrangements also provided the comfort of cultural camaraderie. "It was better being with your own," one interviewee recalled to Iacovetta. "You had somebody to talk with at night. We talked about Italy, and about the jobs—what else?" The boarding house cuisine also reflected familiarity of home and the common ethnicity of the residents. (As a cost-saving measure, pasta-heavy diets were only occasionally supplemented with meat.)

    "Familial priorities loomed large in the lives of Italian men," Iacovetta writes. And the frugal housing situation allowed the male workers—who took pride in being their family's bread-winner—to stretch their meagre wages. Maximizing savings, they could send money to their families in Italy or save for a down payment on a house. "It's common for sixteen people to live in one well-kept eight-room house," Robert Thomas Allen noted in Maclean's (March 21, 1964). "When the newcomer does land a job, he makes every nickel count. Italians are so frugal that they kid themselves about it."

    Go, read.

    Tuesday, April 19, 2011

    The great real estate bust of the 21st century

    Negative population growth in much of the developed world through 2050 means that residential real estate prices will have a long term negative trend.

    For example, the Population Research Bureau predicts (pdf) that Japan’s population, now around 127 million, will shrink to roughly 95 million by 2050. This is due to the aged population with below replacement fertility. Assuming a household size of 3 persons, this would amount to a reduction in housing unit requirements of roughly 10 million units, or about 250,000 units per year between now and 2050.

    Other countries with significant population loss to 2050 forecast by the PRB include:

    Germany – decline from 82 million residents to 72 million residents

    Russia – decline from 142 million residents to 127 million residents

    Ukraine – decline from 46 million residents to 35 million residents

    Poland – decline from 38 million residents to 32 million residents

    Many other countries in East Asia and Europe are forecast to have roughly zero population growth during this time frame. This suggests that in these areas real estate will be at best a non-factor in GDP growth.

    Sunday, April 17, 2011

    Debunking the Demographics Irrelevance Proposition

    In a seminal paper [1] from 1958 Franco Modigliani and Merton H Miller showed why investors should not care about whether firms were financed with debt or equity. This led to the idea of the the debt irrelevance proposition and although the DIP is a theoretical benchmark rather than a real world rule the 1958 paper by Modigliani and Miller remains a key contribution to the finance literature. We should not however extend the same role to the recent attempt by researchers [2] to re-invent the DIP in a new guise replacing "debt" by "demographics". Allow me to explain why.

    Demographics, Just Forget About It ...

    My point of departure is Edward Chancellor's recent GMO letter in which he tackles what he considers to be the non-issue of Japan's dire demographics. He emphasizes two things; firstly, that economists are notoriously poor at predicting demographic variables and secondly he notes that whatever relevance demographics might have for macroeconomic analysis at large (of which Mr Chancellor appears skeptical) it is irrelevant for the investor;

    Besides, long-term demographic forecasts aren’t particularly relevant for equity investors. It’s true that changes in the population have a sizable impact on GDP growth. But stock market returns are not positively correlated with economic growth. Returns from equities are a function of valuation and future returns on capital – a subject to which I will return later – rather than changes in GDP. Nor is there a positive correlation between population growth and stock market returns. In short, investors should not get too hung up on inherently unreliable long-term demographic projections for Japan.

    It is important to underline, in fairness to Chancellor, that the points are made with specific focus on Japan but the the argument seems to have a more general hue. This is even more obvious in relation to one of Chancellor's main references in the form of Morgan Stanley analyst Alexander Kinmont's note entitled The Irrelevance of “Demographics”? Kinmont puts up the following four points which I will use as my points of reference;

    1. It is not clear that demographic estimates are accurate over long time frames. In fact, while spurious specificity is one of the attractions of demographics as a talking point, the fact that neither death rates nor birth rates have proven predictable should caution one against accepting any assertion about demographics.

    2. It is not clear that demographics are the critical variable in determining the level of economic growth. That role falls to the growth rate of TFP.

    3. It is not clear that equity returns are related to absolute levels of growth. Equity returns are an issue of valuation. Nominal returns are greatly affected by inflation too.

    4. It is not clear that demographic change, even if it is allowed as a negative for economic growth, is necessarily negative for stocks, as certain forms of demographic change may be associated with a rising equity market multiple. Demographic change could in fact represent a benign environment for stocks.

    On the first point Kinmont makes points to the irony in that the worry about Japanese demographics seems to be peaking just as Japanese fertility is on the mend. This is a cheap shot though and not one which stands up to scrutiny. First of all on the fertility trend itself I get the same chart as Kinmont's below using data from the World Bank showing a rebound in Japan from a low point of 1.29 in 2003 and 2004 to 1.37 in 2009. However, Indexmundi which takes its data from the CIA World Factbook has fertility much lower and actually declining in Japan. The latest data point from the CIA World Factbook reports an estimate of TFR in 2011 is 1.21. This is a pretty steep difference and I invite comments as to suggest the right number or at least the right trend.

    (click on picture for better viewing)

    All this is of course underlines Kinmont's point that we don't know the future and that economists have a proven track record for abysmal forecast performance. Still, we should get our concepts right at the offset. Long term projections in age structures are likely to be robust as they are a function of people already being born and while migration may change the course of ageing in any given country the fact that we are all ageing at one at the same time means that there are fewer migrants to go around. I would then claim that ageing does matter and that understanding how an economy such as Japan adapts to the ageing of its population remains one of the most vexing and important issues for social scientists and investors alike.

    So when Kinmont implies that low fertility in Japan is a non-issue I have to strongly oppose. Just take a look at the chart above Kinmont himself uses. Fertility has been below replacement levels in Japan since 1970 and on current growth rates (assuming a constant growth rate of fertility which in itself is dubious to the extreme) fertility levels would reach replacement levels some time in 2030-40. So, that would be 60 years with below replacement fertility. Even if fertility in Japan (and again in most of the OECD) took a discrete jump to replacement levels it would do very little to change the outlook for ageing in the immediate future.

    In claiming that demographics do not matter Chancellor are Kinmont are taking a very wide brush over the general recognition in the academic literature that our economic systems tend to hit a snag once fertility falls below a certain level (a TFR of 1.5). This is also called a fertility trap and what it means is that it becomes very difficult to escape negative population dynamics once they set in. I emphasize this since it highlights that we are not, as a friend of mine likes to point, simply shooting arrows into the void when we point to the importance of these issues. I recommend the following presentation by Wolfgang Lutz et al and the paper that goes with it or this old post at AFOE by Edward if you are still not convinced.

    In terms of the postulated increase in Japanese fertility since the mid 2000 it is a positive development, but as is evident from the data this rebound is extremely uncertain. In addition, we need to know whether this is just an echo of the tempo effect (and thus how large the rebound is likely to be) or whether it reflects a real change in attitudes on quantity. I am open to contributions here but the only thing we can for certain is that ageing, in Japan and the rest of the OECD, will continue its march onwards. Here I also feel that Kinmont puts up a straw man when he invokes the idea of Japan's population going to zero;

    The unrevealed assumption, then, behind the mathematics used to arrive at widely-used population estimates is that the Japanese population will drop to zero. One cannot help but suggest that the logic of demographic pessimism is circular.

    I want to re-emphasize that the issue here is not predicting fertility and death rates but recognizing the effect that the current and past trends have on ageing today and tomorrow. Try the recent work by Wolfgang Lutz, Warren C. Sanderson, and Sergei Scherbov if you want to see the cutting edge here and while uncertainty is still a key variable ageing remains a tangible reality. The main question issue I would like to get across is then that the demographic transition manifests itself in a transition of ageing and that this essentially becomes our main unit of analysis.

    Growth and Demographics, No Connection?

    Kinmont and Chancellor argue that demographics are likely to be less important for growth over time as total factor productivity (TFP) growth tends to be the main driver of growth.

    Japan could quite easily grow at a good rate, especially in per capita terms, for a high-income developed country even in the face of a falling population (or more precisely a falling working age population). All that is required is for TFP growth to accelerate back to the level of growth enjoyed by Japan prior to the bursting of the Bubble in 1989. TFP slowdown preceded the population peak. Variation in TFP performance not in labour input growth is likely to be larger than the negative effects of population change.

    This is an important point and more importantly, Kinmont offers an argument to explain the declining labour input in Japan’s economy which links in with the fact that Japan has been stuck in deflation and at the zero lower bound for the best part of two decades (my emphasis).

    Labour input has in fact fallen at an accelerating pace over the past 20 years. It is clear that the fall is principally a decline in man-hours. This cannot be simply a function of a decline in the working age population because that decline only began in 2000. Instead, its origins must lie in rising unemployment and under-employment. A persuasive new paper, The Paradox of Toil, by a researcher at the NY Fed [3] argues that a decline in labour input is a natural consequence of a deflationary economy with zero (or effectively zero) interest rates.

    In short, the declining labor input in Japan is a function of deflation and being stuck at the zero lower bound. In addition, this Fed Researcher Kinmont refers to is Gauti Eggertson who studied under Krugman at Princeton and did most of his initial work on the liquidity trap and the zero lower bound. So, I would be careful getting in his way without a strong look at the argument.

    I think however that we might be dealing with the problem of a missing link in the sense that demographics may be one of the primary sources of deflation and the liquidity trap in the first place. This is an argument that has been pushed in Japan’s case in the sense that it was a lack of pent up demand that held Japan back in the 1990s as well as deleveraging. Indeed, Japan may hold a cautionary tale on the effects of a balance sheet recession in an economy where fertility has been below the replacement level for an extended period. The Eurozone periphery (ex Ireland) who have even ceded monetary policy to Frankfurt are case studies to this theory I think.

    I would also emphasize that as labour input declines so does, obviously, consumption (aggregate demand) input which again feeds into the the paradox of thrift in the closed economy (or perhaps even a realisation crisis?). In an open economy it leads to export dependency as domestic investment actvity responds to foreign demand as well as the excess income you earn from a positive net foreign asset position (if you are so lucky as to have one) becomes a crucial source of growth.

    Another more fundamental point is that if the total factor productivity growth (TFP) is a residual what is actually hidden in this residual? Well, I had a wack at the whole argument a while ago from the perspective of the academic armchair.

    Technology and productivity are famously assumed exogenous in the Neo-Classical tradition while New Growth theory as it was developed in the 1980s and 1990s emphasised the need to specifically account for the evolution of technology. Today, I would venture the claim that there is a consensus that productivity and technology is a function of what we could call, broadly, institutional quality which encompass almost anything imaginable from basic property rights to the level of entrepreneurship. Indeed, a large part of research is still devoted to pinning down exactly which determinants that are most important here both across countries and through time. Now, I would argue that, in the context of standard growth theory, this is where the scope for the study of the effect of population dynamics is largest. Thus I don’t think it is unreasonable to expect the level and evolution of productivity growth and technological development to be a function of the current population structure but also its velocity which is a function of e.g. migration (new inputs?), future working age size etc. Also, this is also where human capital and the evolution of technology is joined at the hip through the idea of innovative capacity and readiness.

    Once we venture into the notion of endogenous growth theory and thus the attempt to directly explain the sources and components of total factor productivity growth there is growing evidence that age structure/demographics alongside a host of other variables are important. Try this one for a recent literature review, and for the general link between growth and demographics the list of contributions is long. You just need to read around a bit.

    I would argue then that growth and prosperity of the modern capitalist welfare state is highly conditional on some form of demographic balance and Japan has long since moved beyond into unbalanced territory. Basically, Japan is stuck in a liquidity trap as well as a fertility trap. The latter works along the lines of depressing consumption demand and making it very difficult to maintain key economic structures such as e.g pension systems. In addition, ageing affect the growth path of an economy and leads to export dependency, this last point however which I concede is not yet an established fact in the literature.

    What about stocks then?

    We seem to have two intertwined arguments here. Firstly, the extent to which demographics may have an influence on growth it is irrelevant for the investor since you can't buy GDP growth anyway. Secondly, the evidence of a correlation between demographics and equity prices is weak and indeed, if anything, should be bullish for Japan (this last point is made by Kinmont).

    Thus the FT summarized the latest findings of the London Business School team of Dimson, Marsh and Staunton, as published in the Credit Suisse Global Investment Returns Yearbook, 2010. The LBS academics examined all the available data (83 markets), and concluded that “99 per cent of the changes in equity returns could be attributed to factors other than changes in GDP”. (...) Growth is not all that it is cracked up to be. This analysis underscores previous academic findings showing that growth
    per se to be of only small importance to stocks.

    It would be unwise to disagree with the gist of this point. Even if I can make a connection between demographics, growth and investor performance it is very likely that buying into such a story at too high a valuation will lead to poor returns. Buying at the right value is the most important aspect of any investment decision.

    This however is not the same thing as saying that just as you make sure to "buy cheap" poor demographics, low growth etc are completely irrelevant. Rather, I think that the extent to which the modern investor needs to understand a decidedly more complex macro picture with lingering deflation, heightened risk of sovereign defaults and zero lower bounds the understanding of demographic dynamics is key. We are then again discussing the question of deflation and low interest rates in Japan;

    The origin of Japan’s problems is falling valuation when compared with the rest of the world. When we note in addition that it is excesses of inflation or the arrival of deflation (that is, monetary phenomena reflecting policy errors) which tend to reduce market average valuations, we feel it safe to conclude that demography will have next to nothing to do with the longer-term return profile of the Japanese market either in nominal or real terms.

    I feel this is a very dangerous claim to make because it assumes that the deflation dynamics of Japan and indeed the problems facing the Bank of Japan in reviving credit growth are unrelated to demographics. In addition there is the unintended consequence of BOJ having to monetize an ever greater amount of JGB issuance in the future which in itself becomes more paramount as Japan ages.

    On the second point regarding a direct relationship between demographics and stock prices (asset prices in general if you will) I think Kinmont does better especially because he does not fall into the asset meltdown hypothesis trap. In short the asset meltdown hypothesis states, in a US context, that as the baby boomers retire they will dissave and thus need to sell off their financial assets to a market which cannot support the flow, because the generation in the working age years is smaller, and that this will lead to an "asset meltdown". Generalized, this is then the classic (and naive) nexus between life cycle economics and financial markets which postulate that dissaving into old age is rapid and imminent.

    There are two problems here. Firstly, the empirical (and indeed theoretical literature) has found it very difficult to verify that dissaving occur among elderly cohorts to the extent postulated by the standard life cycle theory. Secondly, the relationship between asset prices and broad demographic aggregates appear weak. Results differ from country to country and most studies take place in a US and Anglo-Saxon setting which tend to bias the results further.

    Kinmont does however point to a study by Geanakoplos, Magill and Quinzili [4] which show how the ratio of the 45-54 age group to the 25-34 age group is closely related to P/E ratios. As this ratio is set to increase in Japan, Kinmont ventures the idea that, if anything, perhaps you would want to buy Japan on the basis of demographics.

    I have read the research by Geanakoplos et al and I find it intriguing, but my problem is that it does not control for the old age dependency ratio which suggests that the key ratio will be correlated with ageing in general. But I should be hesitant disregarding it on the basis of this hunch. I am preparing a large panel data set at the moment on demographics and stock prices with the aim to essentially rejuvenate a literature which seems too focused on the asset meltdown hypothesis noted above.

    On a more general level, demographics and investment has been a core theme in the post crisis flow into emerging markets which, by and large, share the characteristics of being in the middle or at the end of their demographic dividend. Again, this does not nullify the importance of valuation and certainly, the recent soft patch notwithstanding, many emerging markets are still looking expensive.

    Where goes the DIP then?

    If you build your story up around the notion that investors buy value and not GDP growth you can easily come to the conclusion that demographics are irrelevant for the investor at large. This however would be a mistake.

    I would be the first to wish for a return to a state of affairs in which investors needed only to look at valuation and firm fundamentals to make their decisions. Today however, you need to understand the macro backdrop and in order to do that you need a firm grip on how demographics affect macroeconomics. Pointing out that we are poor at predicting birth and death rates as well as pointing to weak evidence between growth and demographics do not cut it. We need not predict fertility and mortality but instead we need to understand the effects of ageing already present and there is plenty of evidence that demographics affect the growth rate and growth path of the economy.

    I am more sympathetic to the strict relationship between stocks and demographics which is fickle and not well understood. Clearly, there is not presently any convincing model or framework which suggests how and why you might be able to buy sound demographics on a beta level. My main bet is that demographics should, at least, be used to qualify the notion of the global market portfolio and especially that demographics be used to re-balance such a portfolio over time.

    In conclusion, Kinmont and Chancellor bring up some valid and good points in their attempt to brush away demographics as an important input variable to investment and macroeconomic analysis but you shouldn't be fooled. Just as was the case with the original DIP you accept this new version at your peril.


    [1] - Franco Modigliani and Merton H Miller (1958) – The Cost of Capital, Corporation Finance and the Theory of Investment, American Economic Review 48 (June 1958) pp. 261-297.

    [2] - GMO White Paper - After Tohoku: Do Investors Face Another Lost Decade from Japan?, Edward Chancellor and Morgan Stanley Japan Strategy - The Irrelevance of “Demographics”?, Alexander Kinmont. I realise that I have lately been referring to sources and pieces of research which by nature of their origin (banks, research firms etc) are behind subscription walls. I am sorry, but I will make sure to produce relevant quotes so that my readers can follow the issues and arguments. I cannot upload full PDF versions of the reports for obvious reasons and I hope my readers will understand.

    [3] - The Paradox of Toil, Gauti Eggertsson, Federal Reserve Bank of New York Staff Reports, no. 433, February 2010

    [4] - Demography and the Long-run Predictability of the Stock Market. John Geanakoplos, Michael Magill, and Martine Quinzili; August 2002, Revised: April 2004. Cowles Foundation Discussion Paper No. 1380

    Wednesday, April 13, 2011

    On censuses and ethnicity in the western Balkans

    I've blogged here in the past about censuses, especially when censuses run into problems with accuracy or become politicized. In the western Balkans, censuses are considerably more problematic than in Canada or China, simply because demographics--in particular, ethnic identity--matter so significantly when it comes to power and boundary claims.

    Croatia, Kosovo and Montenegro have started to count their populations as part of a year of censuses across the ethnically tense Western Balkans to keep up with EU countries doing the same.

    [. . .]

    Because of the painful history organisation of the count sparked debate and controversy throughout the region: from Macedonia, where ethnic Albanians fear that their importance will be reduced, across Montenegro, where there are complaints that Serbs are being “assimilated”, to Bosnia which did not even manage to adopt a census law.

    In Kosovo, where the last census was held in 1981, while it was still a province in the Yugoslav federation, Belgrade called on ethnic Serbs not to take part in the count organised by ethnic Albanian authorities in Pristina.

    Serbia, which refuses to recognise Kosovo’s 2008 declaration of independence, insists Pristina’s institutions are not authorised to conduct a census.

    [. . .]

    According to estimates, Kosovo has a population of 2.1mn of whom 90% are ethnic Albanians. Kosovo authorities will be helped during the census by the European statistics bureau Eurostat.

    Belgrade, which has the largest number of inhabitants in the region – almost 7.5mn – has postponed the start of its own census to October 1 due to lack of funds, and was planning to also hold it in Kosovo, which it still considers part of Serbia.

    In Montenegro, where 32% of the some 620,000 inhabitants in 2003 declared themselves as Serbs, a campaign was launched recently seemingly pushing people to identify as Montenegrins.

    The national television broadcast programmes insisting on the “Montenegrin identity” of the Orthodox population in the tiny mountainous country.

    In response pro-Serb political parties and the Serbian Orthodox Church have slammed such attempts at “assimilation”.

    In Albania where the census was postponed until November questions about ethnic or religious affiliation sparked debate.

    There are fears that results will show an important rise in the Greek community as many Albanians in recent years changed their identity and religion to obtain residency and working permits in neighbouring Greece, an EU member.

    [. . .]

    In Bosnia the census is so problematic that Sarajevo will likely not even organise one.

    The first post-war head count based on ethnicity is likely to reveal the full extent of so-called ethnic cleansing and upset the division of political power in many communities.

    At the moment power at local levels is often divided along ethnic lines, based on the number of Muslims, Serbs or Croats living in a community before the war. A new census would change all that.

    It's a bit of an irony that ethnic balances are still so controversial in the former Yugoslavia, since many of the successor states are far more homogeneous than before. In Croatia, for instance, the flight in 1995 of ethnic Serbs from the self-proclaimed Serb republics led to a jump in the proportion of ethnic Croats from 78.1% to 89.6% (and a decline in the total population by some 300 thousand), this and other refugee influxes in turn boosting the proportion of Serbs in Serbia, most notably in northern province of Vojvodina. In the two component republics of Bosnia and Herzegovina, of course, pre-war heterogeneity has given way to homogeneity though the country as a whole remains multiethnic and divided. The countries of the western Balkans after a decade of war have achieved the ethnic homogeneity of central Europe; they are near-model nation-states.

    Tuesday, April 12, 2011

    On the Portuguese predicament

    The title of Barry Hatton's Associated Press article "Europeans seek new lives in old colonies" is a bit misleading. The article doesn't deal with "Europeans" generally, but rather is concerned almost entirely with Spain and Portugal, with passing mention of Greece and Ireland.

    Portuguese are packing their bags for booming Angola and Mozambique in Africa, and for emerging economic powerhouse Brazil, where there is a shortage of engineers to prepare the country for the 2014 World Cup and 2016 summer Olympics in Rio de Janeiro. Spaniards are being drawn to their former colonies in Latin America.

    While analysts say the true scale of the new migration is still hard to determine because official statistics lag behind trends, anecdotal evidence and fragmentary data point to what's going on.

    Portugal's Emigration Observatory says the number of Portuguese registered at consulates in Brazil jumped from 678,822 in 2009 to 705,615 the following year. In Angola, the number went from almost 57,000 in 2008 to just over 74,500 in 2009. The number of Mozambican residence permits granted to Portuguese in 2010, meanwhile, was up almost 13 percent on the previous year, to nearly 12,000.

    Spanish electoral registers show around 30,000 Spaniards moved to Argentina between June 2009 and November 2010 — an 11 percent increase over that period. Some 6,400 went to Chile — a jump of 24 percent in the same timeframe — and 6,800 headed for Uruguay, an increase of 16 percent.

    Between Spain and Portugal, Hatton implies that the situation in Portugal is much more serious. Before the economic crash, Spain not only enjoyed sustained--if unviable--economic growth and became a very significant net receiver of immigrants. Portugal, in contrast, suffered a lost decade with little economic growth, a consequence of low productivity and underdeveloped human capital, competition from China and post-Communist Europe in low-end manufactures, and a strong Euro that weakened competitiveness.

    "The emerging markets are where it's happening, that's where the jobs are," says Jorge Borges, a 35-year-old Portuguese civil engineer.

    Disheartened by bleak career prospects in Portugal, whose crippling debt crisis pushed it this week to seek a bailout like Greece and Ireland, Borges crossed the border five years ago and tapped into Spain's building boom.

    Then the overleveraged Spanish economy also collapsed, and Borges recently lost his job. Now he wants to move on again, but Europe's wretched economies are not an option — and his online job hunt is targeting vacancies in Brazil and Angola, distant Portuguese-speaking countries.

    "The first chance I get, I'm going overseas," Borges said from Zaragoza, Spain, where he is awaiting the call to go abroad.

    Brazil in particular is a magnet. The Latin American giant is recruiting foreign civil engineers and architects to meet demand for major public works projects, including more than $200 billion — close to Portugal's annual GDP — in energy infrastructure. Brazil's economy grew 7.5 percent in 2010, the highest growth rate since 1986, and is expected to expand by more than 5 percent a year through 2014.

    Angola and Mozambique, too, are absorbing large numbers of Portuguese emigrants. Lusophone countries are hardly the only destinations for Portuguese: large Portuguese communities exist elsewhere, in Canada and the United States, in France and Switzerland and Luxembourg, and now in Germany, the Netherlands, and the United Kingdom. If Portuguese go abroad, they have options. Not so in their homeland.

    Portugal's low-voltage economy can't absorb the best-educated generation in its history.

    More than 60,000 graduates are idle in their prime. Many more are in low-pay, dead-end jobs.

    A recent song by pop group Deolinda set young people's grievances to music and went viral online as it struck a chord with a generation. The song, called "What a fool I am," lists their gripes, including being stuck at home with their parents despite investing years to polish their CVs.

    One group of twenty-something graduates turned the music into a battle cry. Through a Facebook page they organized national protest marches last month, and more than 100,000 turned out in a dozen Portuguese cities.

    "Not taking advantage of our generation ... is national suicide," says 25-year-old Alexandre Carvalho, one of the organizers.

    Carvalho and his co-organizers want to pursue careers in Portugal but, he says, "it's hard to stay. We'll probably end up going abroad."

    "Parva que Sou" is a moving song.

    The mass emigration is something I commented on back in 2009. The situation has deteriorated further, bad news for the future since a recent Eurostat press release notes that fertility has fallen sharply from 1.44 children per woman in 2003--definitely sub-replacement, but higher than elsewhere in southern Europe--to 1.32. Portugal is fast running out of its existing youth cohorts and isn't going to replace them in the long run.

    (As João Peixoto and Catarina Sabino, in their 2009 Real Institut Elcano paper "Immigration, Emigration and Policy Developments in Portugal", observe, the immigrants from the Lusophone world and eastern Europe who came in the 1990s have been leaving for some time. Replacement migration is not an option.)

    Saturday, April 09, 2011

    On immigrant under-representation in politics

    Vancouver Sun journalist Douglas Todd's blog The Search has an interesting focus on exploring questions of religion and identity in and around and beyond Vancouver and northwestern North America. One of his most recent Vancouver-themed posts highlighted a remarkable fact: descendants of Celtic immigrants, especially Scots, have monopolized the office of mayor.

    The last three mayors of Vancouver - Gregor Robertson, Sam Sullivan and Larry Campbell - all value their rich Scottish ancestries.

    Even prior to the election of these three kilt-loving men, the number of Vancouver mayors who have had Scottish surnames has proved impressive. Three of the past nine mayors of Vancouver, for instance, have been Campbells, which means "crooked mouth" in Gaelic. They include Gordon Campbell and Tom Campbell.

    Vancouver's first mayor, Malcolm McLean, was born in Tiree, Scotland. The city went on to elect a host of mayors with surnames long associated, at least in part, with Scotland, including Malcolm McBeath, Louis Taylor, Gerry McGeer, Tom Alsbury (born in Edinburgh), George Miller, Fred Hume, Charles Thompson and William Rathie.

    This West Coast metropolis, in addition, has had far more than its normal share of those with a combination of Scottish, or Welsh-Irish, surnames, such as Philip Owen, William Owen, Art Phillips, William McGuigan and Charles Jones. (See John Mackie's lively list of Vancouver mayors.)

    Indeed, with the addition of a few mayors with English-Norman surnames (such as Michael Harcourt and Charles Tisdal), the city of Vancouver has almost entirely lacked mayors of non-British ancestry in its 125 years.

    The one well-known exception is mayor David Oppenheimer (1888-1891), who had Jewish-German roots.

    Where have been the Italian, Scandinavian or French mayors -since up until the 1970s people of this ethnicity, along with Germans, together accounted for almost 15 per cent of the city's residents?

    And - the question that's most obvious in this age of high Asian immigration - why has there been no Vancouver mayor of Chinese extraction?

    One might even joke about a Celtic cabal. (And no, McDonald though I am, I can't say anything about this. Really. Can't.)

    This sort of underrepresentation isn't unique in Canada to Vancouver. Toronto's mayoral elections last year, held in a city where half the population is foreign-born, all six of the leading candidates were of European descent, three were of British background, and in the end mayor David Miller passed on his position to Rob Ford. This isn't limited to the position of mayor: in the Canadian Parliament, MPs from visible minority groups are consistently only half as numerous as suggested by visible minorities' share of the Canadian population, with similar or greater under-representation in the Ontario Provincial Parliament and even on the most recent Toronto City Council but one. The Canadian experience is not so different from the French, where until recently descendants of the immigrants from North Africa (not, as Sarkozy's existence shows, immigrants from Europe) were excluded. Elsewhere, too?

    Todd's almost certainly right in saying that it's only a matter of time before Vancouver gets a Chinese mayor, that it's simply a matter of immigrant groups becoming more comfortable with the Canadian system and second- and third-generation community members moving up through the ranks, perhaps starting low and gradually plugging into networks. I hope so: the scale of this lag is embarrassing, and only partially mitigated by the fact that Canadian multiculturalism is almost entirely a post-Second World War creation. This lag is also potentially problematic, inasmuch as excluding entire population groups from political power can bring obvious negative consequences onto the polity.

    Friday, April 08, 2011

    Tuberculosis, Canadian first nations, and pandemics

    Over at my blog, I linked to a startling news item pointing to the Proceedings of the National Academy of Science paper "Dispersal of Mycobacterium tuberculosis via the Canadian fur trade". The abstract?

    Patterns of gene flow can have marked effects on the evolution of populations. To better understand the migration dynamics of Mycobacterium tuberculosis, we studied genetic data from European M. tuberculosis lineages currently circulating in Aboriginal and French Canadian communities. A single M. tuberculosis lineage, characterized by the DS6Quebec genomic deletion, is at highest frequency among Aboriginal populations in Ontario, Saskatchewan, and Alberta; this bacterial lineage is also dominant among tuberculosis (TB) cases in French Canadians resident in Quebec. Substantial contact between these human populations is limited to a specific historical era (1710–1870), during which individuals from these populations met to barter furs. Statistical analyses of extant M. tuberculosis minisatellite data are consistent with Quebec as a source population for M. tuberculosis gene flow into Aboriginal populations during the fur trade era. Historical and genetic analyses suggest that tiny M. tuberculosis populations persisted for ∼100 y among indigenous populations and subsequently expanded in the late 19th century after environmental changes favoring the pathogen. Our study suggests that spread of TB can occur by two asynchronous processes: (i) dispersal of M. tuberculosis by minimal numbers of human migrants, during which small pathogen populations are sustained by ongoing migration and slow disease dynamics, and (ii) expansion of the M. tuberculosis population facilitated by shifts in host ecology. If generalizable, these migration dynamics can help explain the low DNA sequence diversity observed among isolates of M. tuberculosis and the difficulties in global elimination of tuberculosis, as small, widely dispersed pathogen populations are difficult both to detect and to eradicate.

    This finding documents any number of things, such as the underlying and continuing vulnerability of Canada's indigenous peoples to epidemic disease, the long-standing ties between French Canadians--then, as commenters at CBC point out, simply Canadiens--and First Nations, the highly contingent nature of the transmission of pandemic diseases, and the extent to which these pandemics can remain below public attention for decades or even centuries. Parallels with the the evolution and spread of HIV, reconstructed from fossil viruses and genetic data, are entirely merited.

    The data point to 1908 as the year that HIV group M (which now infects more than 31 million people worldwide) began its assault — somewhat earlier than the previous best estimate of 1931. Though 1908 is an approximation, the evidence suggests that the true date almost certainly falls sometime between 1884 and 1924.

    When such evolutionary studies are overlaid with the history of human societies in Africa, a detailed picture of the origins of HIV group M comes into focus. Historically, chimpanzees in west-central Africa have been hunted for food. Many of them are also infected with the virus that HIV evolved from, Simian Immunodeficiency Virus (SIV). Butchering chimps probably repeatedly exposed local hunters to SIV. The virus may have made the leap to infect people many times — but only at the turn of the century did this viral invasion gain a foothold in the population. Around that time, a hunter seems to have picked up the virus from a chimp in the southeast corner of Cameroon and carried the pathogen along the main route out of the forest at the time, the Sangha river, to Leopoldville (modern-day Kinshasa). Mirroring the growth of the cities in Africa, the virus spread slowly in Leopoldville until around 1950, when it began to proliferate rapidly. Still undetected, the virus continued to evolve and to diversify, leapfrogging through burgeoning cities. With the increasing ease of global travel, HIV was carried out of Africa and around the world — and the rest, as they say, is history.

    This reconstruction of HIV's origins certainly satisfies our curiosity — but it also serves as a practical reminder of the conditions that foster the emergence of new diseases. We cannot stop evolution. Pathogens regularly make the leap to infect new hosts, and we increase our chances of being victimized by one of these host switches, when we take on lifestyles that put us in close contact with other species — especially ones closely related to us — like chimpanzees. The early history of HIV also illustrates that the virus is not invincible. For more than 50 years, HIV infected human populations but had such a small impact that it wasn't noticed against the backdrop of other diseases. In comparison to pathogens like malaria (which is carried by mosquitoes) and the common cold (which can travel through the air), HIV is pretty terrible at getting from one person to the next, relying on the direct transfer of body fluids. The virus only got its start in humans through a confluence of opportunity and history — the practice of hunting chimpanzees, the rise of densely populated cities in Africa, and a correlated increase in high-risk behaviors involving the exchange of body fluids (e.g., injection drug use, prostitution). The fact that changes in human societies were so critical in the rise of the virus suggests that changes in human societies could snuff it as well.

    The lead author notes that in the case of tuberculosis, isolated early cases produced an epidemic only when living conditions deteriorated sharply from the late 19th century on, as traditional lands were confiscated, children sent to residential schools, and living conditions on reserve became--and remained--Third World. If these conditions didn't occur, then presumably tuberculosis would be much less of a problem on Canada's reserves.

    Thursday, April 07, 2011

    How migration contributed to the meltdown in Côte d'Ivoire

    The civil war in the West African country of Côte d'Ivoire is entering its final stages (I hope).

    Forces loyal to Ivory Coast presidential claimant Alassane Ouattara laid siege to incumbent leader Laurent Gbagbo's residence on Thursday, after an attempt to pluck him from his bunker met with fierce resistance.

    Fighting continued in the economic capital Abidjan as Ouattara's forces tried to unseat Gbagbo, who has refused to cede power after losing a November election to Ouattara according to U.N.-certified results.

    Sporadic explosions broke the silence in one of the lighter nights of fighting since Ouattara's soldiers arrived in the city a week ago, a Reuters witness said.

    A Gbagbo advisor based in Paris told Reuters Ouattara forces had renewed an assault on Gbagbo's residence late on Wednesday with support from U.N. and French helicopters. His statement could not be independently verified.

    Ouattara forces had attempted to storm the residence in the upscale Cocody neighborhood earlier on Wednesday after talks led by the United Nations and France to secure Gbagbo's departure failed, but they were pushed back by heavy weapons fire, a western diplomatic source who lives nearby said.

    [. . .]

    Analysts said Ouattara forces, who swept south last week in a lightly contested march toward Abidjan, could struggle to best Gbagbo's remaining presidential guard and militias.

    "Just like in Libya, it's going to take both the rebels and outside forces to push Gbagbo out," said Sebastian Spio-Garbrah, analyst at DaMina Advisors in New York.

    This conflict has its roots in the country's demographics, particularly in its immigrant population. Landinfo's Geir Skogseth has a extensive chronology of the crisis and its background. The migration from the Sahel to the West African coast that I blogged about back in June 2006 was one of the most important theme in Ivoirien demographic change.

    During the French colonisation, new migration patterns evolved due to the economic development of the southern part of the country. This led to two main migration flows:

    1. Urbanisation and rapid population growth in Abidjan fed by extensive migration from the whole country (as well as neighbouring countries – especially Burkina Faso, Mali and Guinea). Thus, Abidjan has grown from a tiny village to a city of more than three million in less than a century.

    2. Expansion of the plantation economy in the south-western part of Côte d’Ivoire
    led to the movement of large numbers of farmers from the north and southeastern
    regions (as well as the neighbouring countries mentioned above –
    especially Burkina Faso).

    As is common all over Africa, several ethnic groups live across borders, and may be known under different names in different countries. For example, the Guéré/Wè are known as Krahn in Liberia, and the Yacouba/Dan are known as Gio.

    The migrants going to Abidjan mainly work in commerce, transport, industry and as
    artisans, whereas the migrants moving to the southwest mainly work in agriculture –
    either as farmers or working on plantations. Most migrants from outside today’s Côte
    d’Ivoire came from neighbouring French colonies – Burkina Faso was even
    administered together with Côte d’Ivoire as a single entity. Comparatively few
    migrants came from Liberia (independent in principle, but under strong American
    influence) or from the British colony Gold Coast (Ghana since independence). The
    migration to Côte d’Ivoire from other French colonies was encouraged by the French
    authorities, and this policy was continued by president Houphouët-Boigny after

    People with Kwa origin, especially Baoulé, were given preference in the colonial
    administration, the bureaucracy and in the plantation economy, and this situation
    continued after independence. A smaller migration flow in numbers, but with great
    political importance, consisted of educated Baoulé and other southerners who moved
    from the southeast to other parts of the country to take up administrative posts – in
    the educational system, as civil servants and as administrative staff on plantations.

    As mentioned above, the migration from outside Côte d’Ivoire has been substantial,
    and roughly a quarter of the population of Côte d’Ivoire are people with origin in
    neighbouring countries. Relatively few of these migrants have been naturalised, not
    the least since this generally was a non-issue until the end of the 1980s. Only then
    did a serious debate evolve on whether there should be a difference in rights between
    Ivorian citizens and others living in the country. During the colonial period and the
    single-party system under president Houphouët-Boigny, there was no difference in
    rights to speak of between Ivorian citizens and immigrants, accordingly immigrants
    had little incentive to apply for naturalisation, and very few did.

    Sadly for Côte d'Ivoire, the strong post-independence economic growth that attracted these millions of migrants disappeared by the 1980s, and in the democratic transition following the death of Houphouët-Boigny, ethnic nationalism via the concept of Ivoirité took hold.

    One key to exacerbating ethnic divisions here is the concept of Ivoirité, which means the state of being a true Ivorian. The term manifests itself throughout all levels of society, and is held up by many observers as a root cause of the country's violent downward spiral from its status throughout the 1970s and '80s as the most prosperous, stable country in volatile West Africa.

    Many residents from the government- controlled southern part of country say those from the rebel-held north (often identifiable by their names) are not true Ivorians because many have lineage originating in poorer, neighboring countries such as Mali or Burkina Faso. Some southerners also resent that their northern neighbors support northern political figures.

    [. . .]

    The Ivoirité concept emerged here in the 1970s when many nationals from neighboring countries flooded into southern Ivory Coast to work the manual labor jobs in the coffee and cocoa sectors. Many Ivorians became resentful, feeling the newcomers were coming to take advantage of the financial boom the country was experiencing.

    In the 1990s, former President Henri Konan Bédié brought the term to the national stage when he used Ivoirité to gain an advantage over former Prime Minister Alassane Ouattara in the 1995 presidential election. He accused Mr. Ouattara, who is from the north, of not being 100-percent Ivorian.

    Benoit Scheuer, a Belgian sociologist who made a documentary film called "Ivory Coast: The Identity Powder Keg" in 2001 in which he filmed acts of vandalism and physical violence stemming from Ivoirité, says it was the intellectuals around Mr. Bédié who brought Ivoirité back to into the picture. Mr. Scheuer compares Ivoirité to exclusionary tactics used in other conflicts such as Rwanda and the former Yugoslavia.

    The country, he writes in an e-mail interview, has a "political elite that wants power or is in power but has a legitimacy problem (this was the case of [former President] Habyarimana in Rwanda and Bédié in Ivory Coast). In these cases, the elites are going to manipulate the spirit and the mentality [of the citizens] and are going to develop a discourse, a rhetoric of 'them and us.' "

    The only virtue that this conflict has is that if Gbagbo is decisively defeated, Côte d'Ivoire might be able to move beyond its sterile north-south conflict and evolve in a positive direction. This, alas, might be a lost cause, with Ivoirité having caught on. In the meantime, as the Inter Press Service's Fulgence Zamblé noted, the Côte d'Ivoire that was once a magnet for migrants from across West Africa has now become a major contributor to refugee flows within West Africa.

    As many as a million people have fled Côte d'Ivoire's commercial capital, Abidjan, due to intensified fighting. Many people are fleeing to areas in the north, centre and east of the country as thousands of youth answered a call to join forces loyal to Laurent Gbagbo; others are trying to leave the country.

    Saturday saw a rally of up to 15,000 youth supporters of Gbagbo outside the presidential palace in Abidjan, but the exodus from the city began accelerating following a mortar attack on a market in the northern suburb of Abobo on Mar. 18, which killed 30 people,and wounded 60 more.

    UNOCI, the U.N. peacekeeping force in Côte d'Ivoire, said the shelling of the market was carried out by soldiers loyal to Gbagbo, who has refused to concede that his rival, Alassane Ouattara, was the winner of presidential elections last November.

    The death toll since mid-December is put at 462 by the United Nations, rising quickly as Gbagbo supporters and loyalist soldiers fight their way into areas of the city identified as Ouattara strongholds.

    An estimated 100,000 people have sought refuge in camps and villages in Côte d'Ivoire's neighbour to the west, Liberia. The deteriorating situation will likely also see an increase in refugees fleeing east, across the border into Ghana.