Migration and Demography: Impact on Latvian Sovereignty?

The Baltic Times ran an interesting article recently on a speech given by Latvian President Andris Berzins. He gave the speech on March 25 at the Freedom Monument in honor of the victims that were deported to Siberia during Soviet rule in 1949. This interesting bit is that the speech itself partially (at least) concerned Latvian emigration from the country, with the President arguing that if emigration wasn’t checked the Latvia’s independence would be in doubt in 10 years.

The article reports that over 40,000 Latvians were deported from the country by the Soviet authorities between March 25th and 29th, about 2.3% of the country’s population at the time.

The deportations are distinct to emigration in a number of ways, one being the forcibly nature of the activity. While deportations are, strictly speaking, “emigration” (out-migration) they are coerced and usually distinguished with the qualifier “forced migration”. Similar processes are also attributed to refugees, internally displaced persons, and asylees. Each of these persons are “migrants” but of a forced, rather than willed nature. Of course, that brings us to a philosophical debate on how much “free will” one has in the first place. There are indirect coercions, in terms of gentrification, being priced out of your neighborhood or home. Should we consider this type of migration a “forced migration”?

Another difference is the temporal nature. As the article shows, forced deportations can happen quickly (especially with governments without independent judiciaries). In Latvia, 40,000 people (the size of a town) disappeared within one week. Typically, migration data is presented over the course of a year (since its a fairly rare occurrence when viewed through the lens of an entire country). If this 40,000 in a week held as the average for the year (which thankfully it didn’t), over 2,000,000 Latvians would have been banished to Siberia. Note that would have been more than the country’s population at the time (it was between 1.7 and 1.8 million persons).

With this in mind, let’s try to evaluate the respective outflows. The Census Bureau estimates a NET migration of just over -5,000 people. This net number is the total of immigrants minus emigrants. So there’s likely to be a few more total emigrants, to account for the overall negative migration total. However, 5,000 emigrants represents less than 1% of Latvia’s total population of over 2 million people (2013 estimates). Since I’ve already noted that Latvia’s population is currently declining, its useful to add that the natural increase (in this case decrease) of the population is a larger magnitude than the emigration. To give you numbers, the 5,000 (or more) emigrants is a smaller loss compared to the (about) 8,000 people lost due to the death rate being higher than the birth rate. In other words, the disparity in birth/death rates is having a larger impact on population decline than the emigration rate.

By 2025, since the Latvian President mentioned a timeframe of 10 years, the Census estimates that the net migration rate will remain the same (about 5,000 people leaving Latvia that year). However, the magnitude of the natural decrease will grow to over -12,000 persons. The Census then predicts a worsening disparity between births and deaths, driven by an aging Latvian population (where births will fall even lower as there are less young people, and deaths increase since there are more elderly). Despite the population decreasing to about 1.9 million persons, the loss of 5,000 (or so emigrants) is still less than 1% of the total population.

Though emigration is an obvious “problem” in the sense that it contributes to population decline, the main driver is the very low birth rate. Since I’m not sure what Latvia’s President was referring to in terms of loss of independence, from a demographic standpoint, the low birth rate would be the prime suspect if Latvia lost its independence.

From an economic geography standpoint, however, emigration can cause at least three potential problems. First, there is the loss of labor and revenue from Latvia to another country. Latvian and multinational businesses in Latvia would lose out on increased productivity and not to mention potential sales in Latvia. Moreover, Latvia also loses individual taxes. Of course, the loss of potential sales and income taxes is offset if the individual is able to remit money back home (remittances). Remittances are the sending of money (often through the wire or informal networks) from an emigrants location to the home country to support a family or community. In this way, the home community still receives some economic benefit from the lost labor and productivity. But this brings about the second potential problem, Latvia may grow to become increasingly dependent on remittances for economic well-being. Although figures for remittances are difficult to come by (primarily because it can be under-reported or not reported at all), some countries (like Bangladesh, some Central American, and Sub-Saharan African states) are heavily dependent on remittances. If remittances disappear, whether by financial collapse in the emigrant’s country or loss of work or network, the home country will suffer. Finally, emigrants are often able, through themselves or their families, to maintain property in their home communities. Given that they typically earn higher wages abroad, these properties can earn the resentment of non-migrating neighbors or family members who are often contracted (or asked) to look after a place, work it,

By the same token, however, emigration can offer future opportunities for Latvia. I’m not a globalization booster, but if the returning migrants are able to find jobs or create their own (both necessitate Latvia having the economic space to accommodate this) they can bring skills and perspective that may not be available, making the Latvian economy dynamic in the long run.

Then there’s the political geography standpoint. As a well-functioning democracy, I assume that Latvia provides opportunity for Latvian emigrants to cast their vote during elections. Under the independence argument, perhaps the Latvian President is referring to a situation in which the numbers of Latvians abroad can significantly influence the outcome of an election, even though they don’t live, work, or play in Latvia. In the United States, we don’t have to address or think about this problem since the country attracts many more immigrants than it sends away. However in countries with significant numbers of citizens abroad, perhaps they can influence an election?

And finally, there’s the social aspect of sovereignty. Moving changes a person. Move from one town to another and you’ll be exposed to new people, new ways of doing things. Now move to another province, perhaps another region, maybe you’ll notice cultural differences (I certainly did moving from Texas to Washington, D.C.). Now think about moving abroad for years. You might not know the language (or have only a rudimentary knowledge) and you certainly won’t be acculturated. Chances are, also, that you’ll end up living in a city in the new country which is full of people, some of whom are native urbanites, some of whom are from more rural areas, some from different provinces or regions, and some (like you) from another country. Yes, moving changes you. And then maybe you’ll go back.

With a country like the United States, where a relatively small (I assume) proportion of Americans live abroad for extended (read: over 5 years) periods of time and then come back to the United States, the “political” and “social” shift is negligible. But imagine a country like Latvia where between 25,000 and 34,000 people emigrated to Ireland and the United Kingdom between 2004 and 2007 (based on a blogspot post). The 59,000 Latvians represent 2.8% of the 2013 population of 2.1 million. Of course, that includes children and the elderly (those least likely to emigrate for work). The U.S. Census estimates there 1.5 million Latvians between the ages of 15 and 64, making the 25,000 emigres almost 4% of the population. As the blogspot post observes, during the 2006 Ireland Census only 62% of the Latvians that had arrived remained. The rest may have gone to another country for work or returned home.

This post presented a meandering discussion of whether emigration in Latvia could, in the near future, impact that country’s “sovereignty”, as prophesied by the country’s President. As always, the answer depends on the meaning behind the terms. Will Latvia be at the mercy of a foreign power ( Russia ) because of emigration? Unlikely. Emigrants aren’t going to Russia, they’re going to western Europe. Latvia isn’t likely to lose its political and economic independence with a relatively small outflow of about 5,000 people a year (which might be undercounting). While Latvia isn’t at risk of being dominated by “foreign” interests, a growing proportion of Latvians living abroad, or influenced by their time abroad, can certainly have a noticeable political, economic, and social impact – at least in the medium term. Perhaps this is what President Berzins foresees? Perhaps he believes there’s a coming internal conflict (not necessarily violent) between “native” Latvians and returning or emigrated Latvians. The sovereignty issue, then, isn’t about foreign domination its about a (necessary) political debate on the future of the country and how its people interpret their experiences. If we were to identify a “real” threat to Latvian sovereignty, in the traditional sense, it would be demographic aging and low birth rates. Coupled with Latvia’s poor economic situation (compared with Western Europe) and we find that emigration from Latvia for higher wages is a safety valve for the country. Rather than degrading sovereignty, emigration might actually be protecting it.

List of Countries Currently Experiencing Population Decline

In an effort to provide a quick reference for those interested in stage 5 of the demographic transition model (where the number of births fall below the number of deaths resulting in negative natural population growth), I put together a quick list of countries (map coming in an update!) based on U.S. Census projections. The countries below are experiencing population decline from 2013 to 2014. However, I based this list on growth rates (which includes migration) so some countries may have sustainable fertility rates (like South Africa) but out-migration (emigration) from the country is high enough to cause a decrease in that country’s population from 2013 to 2014. Despite this caveat, the primary reason for population decline is low birth rates. With that aside here is the list:

Belarus, Bosnia and Herzegovina, Bulgaria, Cook Islands, Croatia, Cuba, Czech Republic, Estonia, Georgia, Germany, Guyana, Hungary, Japan, Latvia, Lithuania, Maldives, Micronesia, Moldova, Montenegro, Poland, Puerto Rico, Romania, Russia, St. Vincent and the Grenadines, Serbia, Slovenia, South Africa, Trinidad and Tobago, and Ukraine

South Africa is expected to experience a negative growth rate of 0.4 from 2013 to 2014. It is not due to low fertility (which is estimated at 2.2 children per woman), but due to large migrant outflows in 2013 and 2014 (over 300,000 people each year). With low number of births, the net effect on the population is decline (decreasing from 48,601,000 to 48,376,000).

A surprising addition to this list (for me) was Germany. I had assumed the country would continue to experience population growth for sometime because of its immigration. Not so, total fertility rate in Germany is 1.4 children per woman in 2013 (well below replacement level of 2.1). With an aging population, the 679,000 babies expected won’t be enough to replace the 906,000 deaths expected. These deaths aren’t due to an expectation of virulent disease or war, but simple old age. Moreover the 72,000 immigrants expected in 2013 isn’t enough to replace those German citizens that are dying.

This brings up a hidden facet to population decline, which I’ve touched on previously. These statistics treat “Germany” and “Russia” as homogeneous population groups. As I just said, 72,000 immigrants are expected (estimated) to enter Germany in 2013, many might become German citizens. Immigrants, typically, have larger number of children than the “natives” whether by bringing over their existing families when they naturalize or when they “settle down” and begin having families in their new homes (sometimes both!). When evaluating these (or any) statistics, its always important to keep in mind what’s being analyzed and to question what’s being left out. In these statistics we’re missing valuable ethnic, religious, and other important “identity” data. Like most things, concentrations develop as you narrow your focus from the countrywide to statewide level or from the nation to population groups.

Two general geographic trends are worth pointing out. The first is the prevalence of eastern and southeastern European, former Communist, Eastern Orthodox religion countries. This line stretches from the Baltic states and runs south to the Adriatic Sea. Greece, as it turns out, is expected to begin population decline by 2015. Outside the scope of the study it fits the larger regional pattern (though not “former Communist” thanks to Eisenhowerian intervention [I believe]). This isn’t really the space to speculate on the reason for the prevalence of population decline in this area but I would bet it would have something to do with the influx of “Western” medicine and technology after the Cold War prolonging lives (now leading to “larger” crude death rates) and lingering effects of the “Communist” social experiment that gave women more “freedom” (read: treated the same in regards to their labor as men) than in more “traditional” (read: not Communist) societies. The other general geographic trend is the presence of relatively small islands and island chains. These islands may actually represent the realization of a neo-Malthusian nightmare world. The “carrying capacity” of these islands are tapped, but not due to food, probably water or jobs, forcing a outflow of people. Combined with just below (St. Vincent and the Grenadines is 1.8 children per woman) or at-replacement level fertility and we have population decline in relatively small populations.

And then there’s Japan. While I’ve already discussed Japan in a previous post, its worth reiterating the multiple causes identified (or at least hypothesized) by scholars as to the reason for its population decline (really its falling fertility rates). Some point to ever-expanding educational and economic access for women, which leads to delay in having children (if having any at all). Others point to a wider trend of industrialization and modernization, breaking down traditional family structures and lifestyles – emphasizing “Western” ideals like individual gain and happiness. And of course, there’s the lack of immigration. The truth, as you probably could guess, is likely a combination of all these things (and more). In fact, one could point to the importance that “traditional Japanese culture” places on elders and the influence that longer lifespans has had in reinforcing this value. I used ironic quotes because I think every culture places high value on elders. At any rate, in a world of economic cost perhaps its more important to the individual or the state to be able to care for the elderly rather than children? Just a thought.

Demographic Transition in Latvia: Stage 5

I mentioned in a previous post that Japan isn’t alone in experiencing population decline, that is the proposed fifth stage of the demographic transition model. To briefly recap, stage 5 is the next proposed stage in the demographer’s framework of fertility in a society measured by the total fertility rate and the total death rate. Stage 4, “modernity” (if you will) is characterized by both fluctuating birth rates and low death rates. Birth rates fluctuate with things like economic busts/booms and war/peace dividends. Stage 5 is when birth rates fall below death rates, leading to a slowing of population growth (a deceleration). Eventually, population growth begins to decline if there’s no immigration (or net emigration, that is, people leaving) and birth rates don’t increase to replace the elderly who are dying.

I have amended my earlier post to account for some information. Japan wasn’t the first state to experience sustained population decline, while I don’t know who is first since I haven’t reviewed all of the data, I can say that Latvia was before Japan. Based on data from the U.S. Census International Program, Japan’s population was still growing in 2000, while Latvia’s had decreased. In 2013, Japan’s population had increased over the 2000 population figure from 126.7 million to 127.3 million. In Latvia over the same time the population had decreased from 2.3 million to 2.1 million. By 2050, demographic estimates forecast the population of Japan at 107.2 million and Latvia’s at 1.5 million. The situation is so dire in Latvia, that a Latvian demographer (as reported in The Baltic Times) predicted only 300,000 people would remain in Latvia by 2100.

Breaking down the rates behind these numbers, the number of births per 1,000 people in Japan is 8.2 while the number of deaths per 1,000 is 9.3 in 2013. The reflects a negative rate of natural increase, which is found by subtracting deaths from births and dividing by ten (or -0.11%). The net migration into Japan was negligible in 2013 so there is no effect from that activity on the population growth rate (which accounts for migration). In Latvia, births are 9.9 births per 1,000 people and deaths are 13.6 per 1,000, again suggesting a negative rate of natural increase (-0.37%). But, in addition, Latvia also loses 2.4 people per thousand persons in immigration. I discussed this briefly in a previous post on the uneven geography of minimum wages. Immigration bumps up this negative natural growth to -0.61% growth rate.

Looking into the future, Census estimates that immigration will increase in Latvia by 2050 while remaining negligible in Japan. In addition, birth rates are expected to fall further and death rates to spike dramatically. Why? The answer is the effect of population decline, barring any sort of short-term spike in births, in the relative proportions of the various age groups. As populations decrease naturally, there are fewer and fewer babies being born, meaning fewer potential children to be had. The children that are born are also less likely to have children meaning the situation will continue. On the other hand, the number of elderly increases in proportion to the number of children and middle-aged adults. With this larger number of elderly, the death rate looks like an increase but its really just the impact of having a “grey” population dying at the same rate. Essentially, the Census isn’t projecting a new disease or war will increase the death rate, just that a larger proportion of elderly will be dying off.

Consider the two population pyramids below for Japan (from the U.S. Census). The first is Japan male and female total population in 2000, the second is that total population in 2050. In 2000 the largest age groups are the 50-54 year olds and the 25-29 year olds. While the 50+ population is unlikely to have additional children, the population can conceivably still grow with the large numbers of youths, and as we saw above Japan’s population was still growing in 2000. Now consider the 2050 population pyramid. The two largest age groups are the 75-79 population group and the 70-74 population group. These groups are unlikely to have any additional children and to make it worse, there are fewer proportions of middle-aged adults and young adults to have children as well. Thus, barring any explosion in births population decline looks set to continue. The Latvian geographer bluntly said that Latvians would have to increase the natural birth rate to “African levels.”

Japan Population Pyramid, 2000 (via U.S. Census)

Japan Population Pyramid, 2000 (via U.S. Census)


Japan Population Pyramid, 2050 (via U.S. Census)

The demographer suggests that the government should do more to promote the birth rate including raising subsidies for young families and improving access to kindergartens. Unlike some governmental programs and policies which see immediate results, building roads, raising taxes, and so on, demographic policies take decades to produce results. Therein lies part of the problem, in the world of democracies measured by four, five, or six year electoral cycles, two-decade demographic plans are absent (or not discussed). Partly because each new government stops the policies of the outgoing government or funds different programs.

The potential economic impact of population decline is clear in the Japan 2050 population pyramid. Like developing economies today, these stage 5 economies will have too many dependents to support. In developing economies, the “youth bulge” burdens working adults with having to economically support large numbers of children either at home or through the state, limiting savings and consumerism. The situation would be similar in stage 5 economies when working adults must help support the “elderly bulge” either directly, by having parents and grandparents stay at home, or indirectly, by paying heavy taxes to care for them.

Uneven Geography of Minimum Wage and Cost of Living

A recent article in The Baltic Times inspired today’s post on minimum wages and the cost of living. According to the article, Europe’s Baltic states (Estonia, Latvia, and Lithuania) have some of the lowest minimum wage rates in the European Union. Latvia, in particular, has the third lowest rate (287.07 euros/month). The lowest rates are found in Romania (157.26 euros/month) and Bulgaria (158.5 euros). In contrast, tiny Luxembourg sets a standard of 1,874.19 euros/month. In U.S. dollars, the Luxembourg rate equals 2,510.29 U.S. a month (a nice sum of money). The Romanian standard is equal to 210.63 U.S. (a not as nice sum of money).

Naturally, I started thinking about cost of living between these countries and dug a little bit deeper and found a few surprises. I pulled data on cost of living (as of 2012) from numbeo.com, not familiar with this site but it offers a variety of indexes (rent, groceries, restaurant, purchasing power parity) in addition to an aggregated consumer price index. Each place or state is measured against New York City (set at 100%) so if Japan has a rent index of 105% then, on average, rents in Japan are 5% higher than in New York City. Minimum wage statistics (as of 2013) are from Eurostat, a part of the European Commission. Some results are shown in the table below:

Minimum Wage and Cost of Living in Europe and the U.S. (via numbeo and Eurostat)

Minimum Wage and Cost of Living in Europe and the U.S. (via numbeo and Eurostat)

The first interesting surprise is that not all European states have a country-wide standard for minimum wage. With Europe and its associated states, Norway, Switzerland, Denmark, Iceland, Sweden, Finland, Austria, Italy, Cyprus, Germany, Montenegro, and Macedonia don’t have standard minimum wages. Readers may be shocked to hear that the liberal Nordic states (Norway, Sweden, Denmark, Finland, Iceland) don’t have a national standard (I know I was). What makes it even more surprising is that the indices in these countries are high as well. Norway had the highest consumer price index in Europe (based on available 2012 data, which leaves out Luxembourg), 151.34%. Thus, consumer prices in Norway are over 50% higher than in New York City. Rent, while cheaper than New York City, is the second highest, 55.5%. The highest, Switzerland (also without a national standard), is 58.51%.

Some additional digging on Sweden revealed the likely reason that these states don’t have a set minimum wage – its established at local levels (hello devolution!). A Swedish news source noted in December 2012 that the setting of Sweden’s minimum wage has become “more decentralized,” while an academic paper from 2008 reveals that contractual wage increases and minimum wages are are collectively agreed to for a period of 2 to 3 years. Further, the news article points out that the ratio of average minimum wage to median income (or the hypothetical income of the worker in the middle of the lowest paid and highest paid worker) is the second highest in the Organization for Economic Cooperation and Development (essentially a rich states’ club that includes most of Western Europe, the United States, and Japan). Considering Germany’s federal structure, minimum wages are likely set at local levels there as well. Finally, I would assume that the rest of the Nordic states share a similar model to Sweden’s.

The other surprise is where countries fall in terms of balancing minimum wage with price indices. The outliers are interesting, France offers a minimum wage of over 1,400 euros/month with a combined consumer price and rent index of 75%. The United Kingdom, on the other hand, offers 200 euros/month less than France, despite a higher combined index of 77%. And then there’s the other end of the scale, Poland offers 376 euros as a minimum wage with a combined index of 38%. Bulgaria, noted above as having the second lowest minimum wage (159 euros), has a combined index of 37%. Romania, the lowest with 157 euros, has a combined index of 35%.

As for the Baltic states, Latvia’s 287 euros is balanced on a combined rent and consumer price index of 48% of New York City. By contrast, Croatia and Estonia offer 374 euros and 320 euros for similar combined indices. Lithuania’s 289 euros is not only higher than Latvia’s but its combined index is lower, 45%.

This information allows us to hypothesize why this phenomena exist and what the likely geographic impact will have. Most obviously, France’s higher minimum wage rate is likely a reflection of its more state-driven liberalism. On the other hand, the United Kingdom an early advocate of laissez faire economics has the lowest minimum wage standard of the countries with the highest indices. To a certain extent I would assume most countries and their constituent provinces would set competing standards of minimum wages. For instance, the United States maintains a federal minimum wage but many states (such as Washington) and municipalities (such as San Francisco) set minimum wages, higher than the countrywide average. Setting different minimum wages based on geography (specifically economic geography) makes sense, the cost of living varies from place to place. However, there can be unintended consequences.

Previously I posted on explosive rural to urban growth, minimum wage laws may play a small role into this growth as well – at least where minimum wages are set (primarily in the most economically developed countries). Since the “less” skilled and “less” educated workers (I use ironic quotes because it all depends on what you define as less) are likely to be the ones most like to need a guaranteed minimum wage, higher rates are likely to attract internal migrants to that area. Incidentally, this idea stumbles upon the push-pull model of migration (certain factors “push” migrants away, say lack of jobs or opportunities, while others “pull”, perhaps higher guaranteed wages). Assuming that a guaranteed minimum wage factors into the decision to relocate and that the potential migrant is willing and able to relocate and a host of other factors, perhaps he or she will. Like the attraction of the informal economy, a potential migrant may reason that she is more likely to earn a better wage (even if its the minimum) in a city or state with higher wage than staying at home.

While I have yet to come across direct evidence of this: namely, that workers relocate based on minimum wage laws. I do have some coincidental anecdotes. Take for instance another news report from the Baltic Times, aside from a wealth of geographic information, it discusses Latvian migration to Ireland. As I said earlier, Latvia has a minimum wage of 287 euros/month for a combined rent/price index of 48%. Ireland has a minimum wage of 1,461 euros/month for a combined rent/price index of 82%. In other words, the minimum wage of Ireland is 5 times higher than Latvia while the average prices are less than twice as high. The article is worth quoting at length:

‘Latvia has overcome the crisis only in terms of macroeconomic indicators; most residents are not seeing any improvements and are learning languages and trying to leave the country,’ Latvian President Andris Berzins said in an interview on Latvian Radio on Feb. 18. During his recent visit to Ireland, Berzins saw no point in urging local Latvians to return home. Ireland’s gross domestic product per capita is approximately three times higher than in Latvia, and Ireland’s social support system is much better as well. ‘The figures needed to bring them home are beyond our reach,’ explained the president.

Though I’m not suggesting that the uneven geography of minimum wages needs to be leveled, since Ireland could not (and should not) lower its minimum wage given its higher cost of living and Latvia certainly can’t raise its minimum wage (too much) given its small, but growing economy, more thought needs to be given to the potential impact that local policies in one constituency will have on other localities in a completely different state, often in another country. In the case discussed today, local minimum wage laws in Ireland (perhaps even at lower levels in Sweden or Germany) may influence migration patterns in other economies, notably Latvia or other Baltic or eastern European states. Developing a complex model of migration, based on much more than simple “push”/”pull” frameworks (but certainly including it), would leave to more effective management of people’s expectations, more reasonable growth, and, potentially, less strain on urban and state resources.

Note: If you’re wondering why the U.S. cost of living is so low compared with most of Europe, keep in mind that everything is being compared to New York City. The rest of the U.S. (mercifully) is not like that city.