Tag Archives: Nepal

Bangladesh and India: move towards open borders

As I’ve previously written, South-South migration — migration within and between poor countries — deserves attention in our understanding of global migration, and we can be inspired by scholars of migration and development who have worked hard on this. India is an important example: it is a large, fast-growing, but quite poor country (in per capita terms) surrounded by neighbors who are somewhat poorer in per capita terms and much smaller in size. I previously wrote part 1 of a two-part series on open borders within India and also looked at the existence of open borders between India and Nepal.

The topic of this post is more difficult: policy options for migration between Bangladesh and India. The lack of easily available public material on the subject, combined with my relative ignorance, make me an inappropriate candidate to delve into the relevant empirics and historical details. I’ve asked some others to do guest posts for the site on the subject, and these will hopefully materialize later this year. But, given that I consider the case for open borders to be universal, I should be able to provide an approximate contour of how I believe the case applies to India and Bangladesh. That’s what I try to do here.

Chickensneckindia
“Chickensneckindia” by Ankur; Additions to original map by uploader. Licensed under CC-BY-SA from http://commons.wikimedia.org/wiki/File:Chickensneckindia.jpg, used in the Wikipedia page Bangladesh–India border

Population and income differences

India’s population is about 1.21 billion and per capita GDP (PPP) estimates range from $4000 to $6000. Bangladesh has a population of about 157 million and per capita GDP (PPP) estimates range from $2100 to $3300. The estimates co-vary, i.e., the sources that estimate higher GDP (PPP) per capita for India estimate proportionately higher GDP (PPP) per capita for Bangladesh. You can see a few lists at this page.

Essentially, Bangladesh has about 13% the population and 60% the per capita GDP that India does.

The population ratio seems huge but not terribly so: even if all Bangladeshis migrated to India over a period of a decade, the effect on the Indian population as a whole would not be huge. On the other hand, current patterns of Bangladeshi migration, whereby they settle primarily in border states, may not be scalable to very large migration levels. I believe it is likely that, if borders were more formally opened, migrants from Bangladesh would move out farther to other parts of India, rather than primarily landing up in the nearby states of West Bengal and Assam. Also, I don’t think Bangladesh would empty out. Rather, the situation would probably be similar to that of Nepal and India: no immediate large-scale exodus, but over a longer timeframe, the “Bangladeshi diaspora” in India would grow to a size comparable with the population that is left in Bangladesh. Note that GDP (PPP) per capita in Nepal is lower than in Bangladesh, so if anything, pure economic pressure to migrate should be lower from Bangladesh. But there are some other differences, that we turn to next.

India-Bangladesh-border-map
India-Bangladesh border map, source India’s second most dangerous border? by Martin W. Lewis, May 26, 2011 GeoCurrents

The GDP per capita differences with the bordering states are not so severe. According to a list of gross state domestic products for Indian states, West Bengal is close to the national average and does reasonably better than Bangladesh. Assam does only slightly better than Bangladesh, and the other North-Eastern states do about the same or worse. Ironically, part of the reason for the relative underdevelopment of these states is their relatively poor land connectivity with the rest of India, and that poor land connectivity is because of the geographical location of Bangladesh. As I mention later in the post, allowing freedom of movement through Bangladesh can facilitate greater economic integration of these states.

Why do Bangladeshis migrate to nearby states despite small income differences? I suspect there are many reasons, including long-term cultural connections, but there is also the advantage of being part of an economy that is on the whole larger, faster-growing, and more promising. Once they are in India, they can more easily move to other parts of India — even if most of them don’t avail of the opportunity. Another factor could be weather-related problems leading people to migrate temporarily or permanently out of where they live in Bangladesh.

Differences with Nepal

I’ll repeat some differences I listed in my post on open borders between India and Nepal:

  • Population: Bangladesh has a population of 150 million, about 5-6X the population of Nepal. So, having open borders with Bangladesh is (considered) less feasible, or at any rate, would be a bigger and more transformative change.
  • Greater cultural similarity propelling more migration: Bangladeshis share close cultural roots with West Bengal (indeed, Bangladesh and West Bengal were both part of the state of Bengal in British India). Thus, there is likely to be much greater migration of Bangladeshis since they may have more confidence they’ll be able to adjust to life in West Bengal. (In practice, due to reasons of geography and the strength of border security, many Bangladeshis migrate to Assam rather than West Bengal).
  • Religion: Bangladesh is an officially Muslim country with a Muslim majority. Although not as hostile to India as Pakistan, it still has some hostility. Nepal is a Hindu majority country with small amounts of Buddhism and Islam — religious demographics very similar to India.
  • Historical accident: Bangladesh and India actually started off somewhat well, because India supported Bangladesh (then East Pakistan) in its struggle for independence against West Pakistan (~1971). But political changes in Bangladesh led to a worsening of relations.

Bangladeshi migration: raw numbers

As with most historical South-South migration, the current situation can be very open in practice for migrants. Or at least it has been until recent changes. An estimate of somewhere between 3 million and 20 million illegal immigrants from Bangladesh to India is a similar magnitude to the number of illegal immigrants from Mexico to the United States (about 7 million) and even comparable to the total illegal immigrant population of the United States (about 11-13 million being the median estimate, though there is again uncertainty). The number is smaller as a proportion of the population of India, which is more than 3.5 times the US population. This might explain the lower national salience in India of Bangladeshi immigration. On the other hand, the geographical concentration of Bangladeshi immigrants in West Bengal and Assam means greater regional salience of the issue.

Cultural camouflage

Here’s what Wikipedia says:

As per 2001 census there are 3,084,826 people in India who came from Bangladesh[1] No reliable numbers on illegal immigrants are currently available. Extrapolating the census data for the state of Assam alone gives a figure of 2 million.[3][4] Figures as high as 20 million are also reported in the government and media.[5][6] Samir Guha Roy of the Indian Statistical Institute called these estimates “motivatedly exaggerated”. After examining the population growth and demographic statistics, Roy instead states that a significant numbers of internal migration is sometimes falsely thought to be immigrants. An analysis of the numbers by Roy revealed that on average around 91000 Bangladeshis nationals might have crossed over to India every year during the years 1981-1991 but how many of them were identified and pushed back is not known. It is possible that a large portion of these immigrants returned on their own to their place of origin.[7]

According to one commentator, the trip to India from Bangladesh is one of the cheapest in the world, with a trip costing around Rs.2000 (around $30 US), which includes the fee for the “Tour Operator”. As Bangladeshi are cultural similar to the Bengali people in India, they are able to pass off as Indian citizens and settle down in any part of India to establish a future.,[8] for a very small price. This false identity can be bolstered with false documentation available for as little as Rs.200 ($3 US) can even make them part of the vote bank.

The cultural camouflage that Bangladeshi migrants can engage in reflects two truths: first, the absence of an all-knowing state that has documentation and records for all existing citizens (this might be changing, though, with new identification and documentation schemes being implemented). Second, the genuine historical and cultural connection between West Bengal and Bangladesh, that were one Indian state under British Rule prior to the Partition of India in 1947 (in fact, an attempted partition back in 1905 by the British had to ultimately be reversed after significant opposition). To the extent that there are no obvious differences between Indian Bengalis and Bangladeshis, it would seem that this should point in the direction of officially recognizing the freedom of movement, since it seems to obviate concerns regarding assimilation. But political commentators, who are keen to enforce the sanctity of borders and the formal concept of citizenship, often bemoan rather than celebrate the difficulty of distinguishing Bangladeshis from genuine Indian Bengalis:

Commonality of language, culture and religion between the two countries emerged as a major challenge in identifying immigrants, making deportation extremely difficult. The immigrants speak the same language as many Indians, and often have familial connections that make it easy to assimilate with the local population. Bangladesh’s consistent denial that its citizens are illegally crossing the border also complicates matters. Even when Indian authorities have identified illegal immigrants, deporting them becomes almost impossible given the reluctance of Bangladeshi authorities to cooperate.

An underdeveloped deportation machinery

As I wrote in my South-South migration post:

In some ways, the current nature of South-South migration as well as the social and political attitudes to it closely resemble 18th and 19th century migration worldwide. People moved from very poor countries to less poor countries with more vibrant cities and growth opportunities. Natives weren’t exactly thrilled, but strong anti-migration sentiment, while often virulent by modern standards, was relatively localized and took a fair amount of time to translate to successful national movements to curb migration. I’m not aware of survey data similar to the World Values Survey for the 19th century, but my guess is we’d see a similar 25-25-25-25 split about migration despite more overtly prejudicial attitudes among the people (similar to the situation in India today).

This connects with my very first post on the Open Borders site, where I blegged readers on why immigration was freer to the 19th century USA. I had listed three potential reasons in that post: (1) wisdom/desirability, (2) technological/financial feasibility, and (3) moral permissibility. At the time, I had written that (1) was unlikely, and the likely truth was a mutually reinforcing loop of (2) and (3) (that did eventually get broken in the United States with the Chinese Exclusion Act). I think the same dynamic is at play in South-South migration, with the difference that South-South migration today has at least some nominal level of border controls, and there’s enough of a global precedent of strict border controls that the learning curve towards very strict border enforcement can be (and in many cases, is being) traversed a lot faster.

Indeed, we can see this in India’s case today. There have been occasional bursts of effort to round up and deport illegal immigrants, often by governments that are prepared to basically “deport them all” — at least in principle. But if you’re used to US deportation numbers, you might laugh at passages like this:

Yet deportation under the Foreigners Act is also problematic. In 2003, the then Home Minister L. K. Advani ordered all states to deport illegal immigrants. A few weeks later 265 people were sent to the border, but authorities in Bangladesh declined to accept them. In fact India’s Border Security Forces (BSF), and its counterpart the Bangladesh Border Guards (then called the Bangladesh Rifles), came to the point of violence over the issue.

The deportation rates do seem to be increasing over time:

At the end of 2012, for instance, 16,530 Bangladeshi citizens with valid travel documents were found to be overstaying in India—while 6,537 and 5,234 Bangladeshi nationals were deported in 2012 and 2013, respectively.

And while the move towards newer, more effective forms of identification will probably mean that previous migrants get effectively amnestied, it may well make things harder for future migrants.

Overall, the level of preparedness and competence of the interior enforcement and deportation machinery at present seems to be comparable to what the US had for Chinese immigrants around the passage of the Chinese Exclusion Act. In both cases, natives are very unsympathetic, and many of them are openly virulent, to the immigrants. But the enforcement machinery is sporadic and erratic, and its throughput is minimal. Contrast this with “pro-immigrant” Barack Obama, who deported over 30,000 people a month, not so much because he personally hated immigrants, but because the deportation machinery was so well-established and functional that trying to slow it down it would be an expenditure of political capital — one that Obama didn’t consider necessary.

The terrorism problem

In the wake of the October 2014 Burdwan blast, close to the border with Bangladesh, carried out by an Islamic terrorist (or, in their view, revolutionary) group called the Indian Mujahideen, concerns about border security and terrorism were revived. The blasts revived concerns about Muslim madrassas (training institutes) as breeding grounds for terrorism Interestingly, Indian Muslim religious leaders (who do not necessarily represent the views of all or even most Indian Muslims, but are considered widely influential) sought to deflect this by stating that the problem wasn’t Indian Muslims, it was Muslims coming from across the border (i.e., Bangladesh). For instance, NDTV reports:

“No madrasa in India is anti-national. No Muslim in India is anti-national,” said Jamiat’s Sidiqullah Chowdhury. “The ones who come from outside are anti-nationals. Indian Muslims or madrasas are not terrorists.”

I believe that the threat of terrorism is in general greatly exaggerated because of its greater political salience and visibility, but it is still a threat that deserves to be taken somewhat seriously. Would an open border between India and Bangladesh lead to a dramatic increase in terrorist activity? I don’t know enough to offer a clear answer, though I’m hoping that posts later this year will explore the question more closely. But going by what we generally know about terrorism and its relation to migration policy, it seems that, to the extent that the threat of terrorism can be reasonably contained, it can be done through better targeted policy, and closing the border to economic migrants can in some ways complicate it. Consider, for instance, this discussion in the Daily Mail:

“The advantages they enjoy here are innumerable: immigrant-pockets which have proved to be excellent hideouts; a big metropolitan, Kolkata offers them concealment and its railway stations, namely Howrah and Sealdah, easy connectivity with the rest of the country,” the official added.

In other words, those with terrorist ambitions can conceal themselves among economic migrants, who are also undocumented and seek to evade detection. What would happen if the border were officially opened? Things could move in either direction: terrorists would be hiding within a larger population, so would be harder to detect. On the other hand, if peaceful migrants did not need to hide from the law, a documentation or identity scheme could be more effectively enforced, so that one could more reasonably presuppose that those who did not seek to get appropriate documentation had nefarious intentions. Also, the cooperation of the Bangladeshi government in combating terrorist activities could be enlisted more effectively if the Indian government weren’t getting in the way of peaceful migrants from Bangladesh. How do these competing considerations balance out? It’s hard to know a priori, and it’s possible that there will be an increase in terrorist activity, but I don’t think that it will be a significant increase.

This is similar to the point that my co-blogger John Lee made about the US-Mexico border, where he drew on statements by officials who actually work in law enforcement at border towns:

Simply put, if you want a secure US-Mexico border, one where law enforcement can focus on rooting out murderers and smugglers, you need open borders. You need a visa regime that lets those looking to feed their families and looking for a better life to enter legally, with a minimum of muss and fuss. When only those who cross the border unlawfully are those who have no good business being in the US, then you can have a secure border.

Co-blogger Joel Newman made some related points when discussing open borders, terrorism, and Islam:

So one argument notes that, unlike our current restrictionist policy which devotes considerable resources and focus on keeping out unauthorized immigrants seeking to work in the U.S., resources under an open borders policy could be focused on screening out terrorists. Another argument is that the free movement of people between countries could lead to the spreading of ideas contrary to those which inspire terrorism; immigrants who move between the U.S. or other western countries and their native countries would share values such as individual rights, tolerance, and democracy with their compatriots who remain in the native countries. A third argument is that if terrorism grows out of weak economies in native countries, the free movement of people from those countries and the resulting economic benefit to those countries (through remittances and immigrants returning to their native country to establish new businesses) could help prevent terrorism.

There is another reason open borders could help combat terrorism. Kevin Johnson, author of Opening the Floodgates, notes that “carefully crafted immigration enforcement is less likely to frighten immigrant communities—the very communities whose assistance is essential if the United States truly seeks to successfully fight terrorism.” (page 35) Without the fear of being the targets of immigration enforcement, immigrants would be more likely to cooperate with authorities in identifying individuals who are potential terrorists in the U.S. and assist with efforts against terrorist groups abroad.

Narendra Modi’s election rhetoric

On February 5, 2014, Prime Ministerial Candidate Narendra Modi gives a Hindi speech in Assam from citizenist premises. He argues that Indian citizens should be given preference in jobs, and Bangladeshi immigrants to West Bengal and Assam have been stealing jobs from natives

Narendra Modi assumed office as the Prime Minister of India in May 2014. While campaigning for the election, Modi emphasized repeatedly that, once elected, he would aim to solve the problem of Bangladeshi illegal immigration. It wasn’t clear at the time whether his words, like most political manifestos, were mere promises, or whether he intended to follow through on them. Modi did make these pronouncements only when campaigning in Bengal and Assam, rather than using a national platform, suggesting that it might be more of a device to connect with and win over voters in the affected regions than a key component of his actual agenda.

Modi’s views on immigrants had some interesting twists. For instance, in February 2014, Modi made remarks to the effect that India’s borders would and should remain open for Hindus worldwide, but not for Bangladeshi Muslims. The “open for Hindus worldwide” idea would be similar to Israel’s Law of Return.

“As soon as we come to power at the Centre, detention camps housing Hindu migrants from Bangladesh will be done away with,” Mr Modi told a public rally at Ramnagar in Assam.

“We have a responsibility toward Hindus who are harassed and suffer in other countries. Where will they go? India is the only place for them. Our government cannot continue to harass them. We will have to accommodate them here,” he said.

Stating that this did not mean that Assam has to bear the entire burden, he said “it will be unfair on them and they will be settled across the country with facilities to begin a new life.”

Earlier, Hindus from Pakistan had arrived in Gujarat and Rajasthan, but Atal Behari Vajpayee during his prime ministership evolved schemes to accommodate them in different states, he said.

However, Modi has threatened deportation for the majority of Bangladeshi migrants, who identify as Muslim:

Narendra Modi has said that “Bangladeshis” will be deported if he comes to power, in his sharpest comments yet on illegal immigrants. They have been given the red carpet welcome by politicians just for votes, he said at a rally on Sunday.

“You can write it down. After May 16, these Bangladeshis better be prepared with their bags packed,” Modi said in Serampore in West Bengal, which shares a porous border with Bangladesh.

Modi accuses other political parties in West Bengal and Assam of encouraging such migration and helping the migrants obtain false documents so that they can vote — a variant of the electing a new people argument, a particularly extreme form of the general political externalities argument. I don’t know enough about the extent of actual voter fraud in West Bengal and Assam (although voter fraud in the US seems to be greatly exaggerated, the situation is likely to be quite different in India). I do think, though, that to the extent the problem is real, it is created to quite an extent by the illegal status that these people have. If one political party keeps announcing its agenda to deport you (even if it rarely executes on that agenda), and another political party, openly or tacitly, allows you to stay, who will you swear allegiance to?

Interestingly, even while disagreeing with specifics, most commentators have tacitly endorsed Modi’s overall frame of needing to restrict immigration from Bangladesh. For instance, the answers to a Quora question about Modi’s speech (YouTube video earlier in the post) defend a nation’s right to arbitrary selection of immigration policy, appealing to intuitive versions of the idea we here call citizenism and collective property rights. For instance, Syed Fuad, who identifies as Bangladeshi, writes:

I’m not an Indian, so it’s not for me to decide. But in my opinion, Narendra Modi shouldn’t take it easy. He, being the Indian Prime Minister, is accountable to Indian citizens. Addressing their issues should always come before anything else, even if it means taking strong and often unpopular stands on sensitive issues.

Narendra Modi’s proposed solution

For the first few months after being elected, Modi seemed to be quiet on the subject of Bangladeshi migration. I assumed that, like most campaign trail rhetoric, this too would not actually be executed.

However, around the end of November, Narendra Modi’s proposed solution was released. Quartz has a detailed review. Here are the highlights:

Prime minister Narendra Modi has indicated that his government is open to executing a land swap with Bangladesh that will iron out long-standing border disputes and help thousands of people who live along the 4,096.7 kilometer-long common land boundary.

The deal, once ratified by the Indian parliament (PDF) will redraw India’s boundary with Bangladesh, where New Delhi will cede 17160 acres of land, in return for 7110 acres and swap enclaves. Enclaves are areas which are surrounded from all sides by foreign territories. India currently has 111 enclaves belonging to Bangladesh, while Bangladesh has 51 such areas.

Modi, in a speech in Assam on Dec. 01, also assured that the land swap—which his own party had previously vehemently opposed—would stop illegal Bangladeshi migrants from entering into India.

“The government will utilise the India-Bangla land transfer agreement to seal all routes across the international border through which illegal Bangladeshi migrants have been entering Assam and creating havoc in the state,” the prime minister said.

Quartz notes many problems with Modi’s solution in terms of the stated aim of reducing the illegal immigrant population, but does not question the goals themselves.

What would I suggest?

For good or bad, Narendra Modi, thanks to his generally hardline reputation, has more leeway to make genuine progress with migration liberalization than most other prime ministers. Given his past record of rhetoric and action, he is relatively insulated from the charge of being soft on Bangladeshis or on Muslims. This gives him a Nixon goes to China opportunity.

Modi has made some surprise moves in that direction. I don’t know about the wisdom of the land swap per se, but insofar as it contradicts his own rhetoric and at least apparently concedes land to the other country, it shows how, as somebody with a hardliner image, he is able to take actions that people with a softer image might be afraid to take as it would make them look weak.

But the land swap does not solve the fundamental need for free movement: even after all these years, the villages of Bangladesh and West Bengal are intertwined. People have extended families across the border. People seek economic opportunity across the border (my co-blogger John Lee made a related point about the borders of South-East Asia and the Nusantara a while back).

Modi can take a bold step forward by proposing a free migration zone with Bangladesh of the same sort that India has with Nepal. If Bangladeshis can come and go as they please, they have few incentives to pretend to be Indian citizens or to vote for parties using fraudulent documentation. Most people from Nepal who come to India are secure in the knowledge that they are free to go back and forth, and feel little need to become Indian citizens because it makes very little material difference to them (of course, there will be some who want Indian citizenship after living in India for a long time, or if they want to travel to third countries). Bangladeshis could get to the same point.

Modi could combine the creation of legal channels for migration with user fees that are slightly greater, but not much greater, than the cost of migrating illegally and getting false documentation. He could also come up with creative ways of encouraging greater geographical spread of Bangladeshi migrants. He’s already given the matter some thought with regards to Bangladeshi Hindu refugees. I don’t know offhand what the ideal solutions would be, or even if the problems faced by the states adjacent to Bangladesh are serious enough to warrant action, but it might still be politically expedient for Modi to show he is doing something in that regard. For instance, there could be special trains for immigrants that, at a relatively low cost, transport the immigrants to specific states, and where the immigrants formally enter the country after getting off the train at the new state. (Incidentally, concerns that immigrants who land at a particular part of the country may just stay there rather than migrating to other parts of the country were also voiced by some officers at Ellis Island).

There is also the question of whether the Bangladeshi government will agree to a free migration agreement with India. If it doesn’t, the Indian government can still do something similar unilaterally, but perhaps with fewer bells and whistles, so as to encourage the Bangladeshi government to reciprocate. Overall, I believe that the case for free migration doesn’t depend on reciprocity, but it may still be politically expedient to negotiate the deal that way, to placate voters that India is getting something from the deal. Independently, there is probably some value in making it easier for Indians to move to and from Bangladesh. There are also trade and transportation advantages: reducing border tensions with Bangladesh can allow for easier transportation of goods and people between the North-Eastern states and the rest of India. Currently, due to the way the borders are structured, the North-Eastern states are connected to the rest of India via a very narrow region of land, making economic integration harder. The free migration agreement can accompany greater ease of movement of goods and people through Bangladesh between the North-Eastern states and the rest of India.

Why does this matter?

The place premium between Bangladesh and India is probably not large (it would approximately equal the GDP per capita ratio, which is less than 2). And the absolute gains per migrant aren’t large either. Why, then, is this important?

  • The absolute population sizes in question are big enough. Allowing the 150 million Bangladeshis to move to India, seasonally or permanently, is a big deal even if the per capita gains aren’t huge. It creates a larger, more flexible, integrated labor market.
  • There may also be a peace dividend: with less border tensions, the residents of the countries have more opportunity to collaborate, and the governments can better negotiate on other issues.
  • The Indian government saves on some border and interior enforcement costs, though there may be some costs to setting up an efficient free movement system. But I suspect that those costs are less than the costs of setting up a border and enforcement process that actually works at the level that the US system does.
  • The free movement zone can create a precedent for a larger free movement zone. Other countries like Sri Lanka and Burma could be encouraged to join at a later stage. And in the longer run, perhaps Pakistan could be part of the zone as well. Open borders between India and Pakistan are unlikely to happen in the near future, because of the usual problems facing open borders between hostile nations. I think a free migration zone offers the best bet.
  • This is somewhat niche, but allowing free movement creates an insurance of sorts against adverse weather events, something that Bangladesh in particular is susceptible to because much of its land is low-lying and flood-prone. It is believed that climate change will exacerbate the problems that Bangladesh is facing. Free migration can possibly help avoid disaster striking suddenly. Similar points has been made by co-bloggers Joel and Nathan.
  • From the open borders perspective, I believe that this is a critical time in the history of India’s immigration enforcement. This is a time when Indian record-keeping is finally getting good enough that the country can start building a systematic enforcement and deportation apparatus. If we start traveling down this road, it can be quite hard (though not impossible) to reverse or change course. I believe that the years immediately before and after the passage of the Chinese Exclusion Act had considerable importance in terms of the development of the basic immigration enforcement apparatus, as well as the legal precendents they created. India could be going through a similar phase. Putting a brake on the process could yield larger-than-meets-the-eye dividends in terms of an undesirable road not taken.

Related reading

Some of the links in this section are also present in the body of the article.

South-South migration and the “natural state”

This blog post builds upon an Open Borders Action Group post of mine and the comments on it.

In an earlier post on what open borders advocates and scholars of migration and development can learn from each other, one of the things I had said open borders advocates can learn from scholars of migration and development was the importance to give to forms of migration that currently exist, as opposed to what might exist in a hypothetical open borders world:

More focus on intranational migration, migration between low-income countries, and migration from low-income to middle-income countries: […] [I]t might be worth looking at the huge amount of migration that already exists and understanding its implications. While still arguing morally for open borders worldwide, we can focus on understanding what already exists and making changes to it. Often, there is little reliable data and little interest among readers in such matters (such as Nepal and India, or North Korean refugees), simply because blog readers are highly likely to be in First World countries and are more aware of First World issues. But I think that pushing more in the direction of better understanding migration as it’s actually happening is worthwhile, even if it doesn’t make us popular. We can be inspired here by migration scholars, who have worked very hard to compile data and collect anecdotes to further the world’s understanding of migration.

World Press 2014 Signals from DjiboutiWorld Press 2014 photo: Signal from Djibouti, source National Geographic. The photo shows people from Somalia living in Ethiopia trying to catch Somali cellphone networks at the border of the country so as to talk cheaply with their families.

This post can be considered a partial attempt to put that learning in practice. Here are some examples of “South-South migration” that I have in mind when listing my general observations. Each of these should deserve its own post. For those that don’t already have posts I link to a relevant news article or paper:

Some of the salient features of much of this South-South migration:

  1. In most of the cases, the destination countries of migration are large and somewhat heterogeneous economically. The average GDP per capita in the destination may be somewhere between 2 and 5 times that in the source country (with the exception of the somewhat special case of migration from North Korea to China, the range is more like 2 to 3 times). However, this hides a large degree of intranational variation in the destination country. The destination countries, despite their poverty and Third World status, generally have greater scope for people to become rich and successful. They have bigger cities with more opportunities. Compare, for instance, Afghanistan with Pakistan. Pakistan scores pretty poorly in terms of GDP per capita or HDI. But it has cities like Karachi and Lahore, that are (relatively speaking) thriving centers of commerce. Similarly, Indian cities offer opportunities that most Bangladeshis can’t access in their home countries. Even if the migrants don’t initially move to cities, the promise is there.
  2. Large parts of the destination country are rural, and the rural-urban gap on many development indicators is huge. Moreover, the rural areas may not really have much affiliation with or integration into the national identity. Many people in rural areas may not even have any form of documentation establishing citizenship or national membership. Thus, many natives are also “undocumented” and in some ways indistinguishable from migrants. The role of ethnicity as betrayed by appearance and accent is therefore greater than the role of formal citizenship.
  3. Migrants tend to move to border towns and to some large cities, generally those with pre-existing diasporas (cf. diaspora dynamics). These are the places where the issue of migration has the greatest salience, and anti-migration sentiment may be more common, and expressed more openly and virulently than in most developed countries.
  4. There is usually no pro-migration or pro-migrant movement per se, though there may be NGOs focused on providing services for migrants.
  5. If anything, intranational migration might be more salient in many parts of the country. In fact, intranational migration may also quantitatively swamp international migration, as is the case in China and India (here’s a blog post on intranational migration within India and a blog post discussing large-scale migration within India and China). But insofar as there are no real constitutional ways of restricting intranational migration, it might never become a politically important issue at the national level. In many regions, on the other hand, intranational migration may take on more significance than international migration in political rhetoric, even if politicians have little power or little interest in actually curbing such migration.
  6. At the national level, the importance of migration is minimal. This is partly because the destination countries have many more pressing problems. Anti-migration movements are relatively localized, and pro-migration movements are negligible.
  7. For many people in such countries, the issue of open borders and migration restrictions is a largely theoretical one, and their answers to it might represent generic ideas of human fairness untainted by personal interest, so to speak. This might explain why India, despite not being known for having a high degree of tolerance and welcome for foreigners of different races and ethnicities, had a roughly 25-25-25-25 split in the World Values Survey question of how open migration policy should be.

In some ways, the current nature of South-South migration as well as the social and political attitudes to it closely resemble 18th and 19th century migration worldwide. People moved from very poor countries to less poor countries with more vibrant cities and growth opportunities. Natives weren’t exactly thrilled, but strong anti-migration sentiment, while often virulent by modern standards, was relatively localized and took a fair amount of time to translate to successful national movements to curb migration. I’m not aware of survey data similar to the World Values Survey for the 19th century, but my guess is we’d see a similar 25-25-25-25 split about migration despite more overtly prejudicial attitudes among the people (similar to the situation in India today).

This connects with my very first post on the Open Borders site, where I blegged readers on why immigration was freer to the 19th century USA. I had listed three potential reasons in that post: (1) wisdom/desirability, (2) technological/financial feasibility, and (3) moral permissibility. At the time, I had written that (1) was unlikely, and the likely truth was a mutually reinforcing loop of (2) and (3) (that did eventually get broken in the United States with the Chinese Exclusion Act). I think the same dynamic is at play in South-South migration, with the difference that South-South migration today has at least some nominal level of border controls, and there’s enough of a global precedent of strict border controls that the learning curve towards very strict border enforcement can be (and in many cases, is being) traversed a lot faster.

In many ways, both current South-South migration and historical migration are closer to the “natural state” of migration and the responses it engenders. All is not hunky-dory with this natural state. The occasional outbreak of riots against immigrants, while quantitatively negligible, as well as the more frequent displays of overt private prejudice, are disconcerting. But for all that, the system is still a bigger win-win for migrants and natives than the strict border controls that much of the developed world has successfully implemented, and that the developing world is rapidly building out.

Nepal and India: an open borders case study

This blog post builds upon this Open Borders Action Group post and its comments.

Nepal is a small mountainous country, with much of the Himalayas running through it, and is home to some of the world’s tallest peaks, including Mount Everest. It is bounded by China on the north, though the Himalayas form a fairly impenetrable barrier for getting to China by land. it is also bounded by Tibet to the northwest and by India on all other sides. The border between Nepal and India is relatively easy to cross by land (compared with the border with China).

Map of India, Nepal, and other nearby countries
Map showing India and Nepal. Source Himalayan Homestays

According to many indicators, Nepal is somewhat less developed than its southern neighbor India:

  • GDP (PPP) per capita in Nepal is in the $1300-1500 range, relative to India, where it is in the $3800-4000 range. Thus, GDP estimates suggest that Indians are 2-3X as well off as Nepal.
  • Literacy rate in Nepal is 66% versus 74% for India.
  • The population of Nepal, according to the official 2011 census, is a little over 26 million, or about 2% of the Indian population (over 1.2 billion). The CIA World Factbook from July 2011 estimates the population at a little over 29 million.

Nepal-India land borderThe open border between India and Nepal, source Nepal Mountain News
Nepal and India have good diplomatic relations. The aspect that interests us most, obviously, is that the two countries have had open borders throughout their history (since long before Indian independence from the British). In fact, the border is literally open — it’s largely unmanned, people can cross any time, and natives of the two countries are not even required to have a passport to cross by land. Here’s what the description says (same as previous link):

Nepal-India border is unique in the world in the sense that people of both the countries can cross it from any point, despite the existence of border checkposts at several locations. The number of check posts meant for carrying out bilateral trade is 22. However, only at six transit points out of them, the movement was permitted to nationals of third countries, who require entry and exit visa to cross the border. As the whole length of the border except police does not patrol the checkposts or paramilitary or military forces of either country, illegal movement of goods and people is a common feature on both sides of the India-Nepal border.

So, the puzzle: given that India has 2-3X the per capita GDP of Nepal and substantially more economic opportunity, and given that the countries have open borders, why hasn’t there been labor market convergence between the countries? Why hasn’t a larger fraction of the Nepalese population moved to India in search of economic opportunity?

Here are a few other articles on Nepalese migration to India:

Open borders is a political non-issue

Indians aren’t necessarily welcoming of Nepalese as individuals: attitudes to Nepalese range from welcoming to hostile. But the total size of the Nepalese population is small relative to the Indian population (about 2%, and certainly not more than 3%). So far, nothing terrible has happened in India due to Nepalese migration. Even if the entire country of Nepal were to move over a decade to India, it wouldn’t be noticeable to most Indians. For the most part, therefore, Nepalese migration seems a non-issue.

There are a few issues. Smuggling of goods along the border could lead to a tightening of border security, and that might get in the way of peaceful migration. Maoists in Nepal have connections with Maoist-Naxalites in India. This too could lead to tightening of borders for security-related reasons. Overall, however, I don’t expect the open border between the two countries to be closed even in the face of Nepalese migration picking up significantly. This is in sharp contrast with Bangladeshi migration, which has caused a nativist backlash particularly in the state of Assam.

Are the GDP numbers reliable?

The GDP estimates carry a lot of uncertainty. In particular, they can be bad measures of standards of living in cases where:

  • A lot of economic activity goes unreported, or
  • government is a large share of the economy and spends the money on things that people don’t really benefit from.

Both problems exist to some extent in India and Nepal, but not to the same extent as they would in a country such as Somalia (the first problem) or North Korea (the second problem). Further, there’s no reason to believe that the figures are unreliable in a direction that would overstate the disparity between the two countries.

Reasons for economic disparity

Why is India so much richer than Nepal? A number of reasons suggest themselves:

  • Better geography: Nepal is covered with mountains that are unsuitable for many economic pursuits. It’s also landlocked. India is geographically diverse, and has large areas of flat plains suitable for agriculture, industry, and dense urban life. Moreover, India has a long coastline with great sea ports.
  • Larger population: Having a larger population allows for a larger diversity of activity and an economy with a greater level of specialization.
  • Historical advantages: The British developed a number of institutions in India that were bequeathed to the Indian government at the time of Indian independence. Even prior to the British, India, though very poor, had a diversity of historical institutions. These historical advantages gave India more growth potential, even if the country failed to realize that potential until very recently.
  • Economic freedom: Economic freedom in Nepal has been somewhat lower than India. According to the Economic Freedom of the World report, Nepal has had economic freedom in the 5-6.5 from 1980 to now, and even though its absolute level of economic freedom has increased, its ranking has fallen, as the rest of the world liberalized and it didn’t. As of 2011, it ranks 125th out of 152 countries with a rating of 6.19/10. Economic freedom in India started out lower than Nepal in the 1970s and 1980s, but with liberalization, overtook Nepal. In 2011, it ranks 111th out of 152 countries with a rating of 6.34/10. The difference isn’t huge, and arguably within the measurement error, but Heritage Foundation’s Index of Freedom reports a similar small gap in India’s favor. This probably accounts for only a small fraction of the economic difference, though.

How many Nepalese are there in India?

The size of the Nepalese population in India is highly unclear, both due to the fact that migration between the countries is not properly recorded by either, and because of the lack of clarity in definition. Table 3.3 in the Being Nepali Without Nepal chapter estimates that, as of the 1981 Indian census, there were 500,000 Nepal-born people in India, whereas the size of the Nepali community construed more broadly (to include descendants of Nepalese who spoke the Nepali language) may be as large as 2.25 million. Wikipedia claims that there are 4.1 million Nepalese Indians, without citing a source, but the data on the page seems to be taken from the 2003 CIA World Factbook (I cannot find a direct link for the specific claim). If correct, this is a huge increase relative to 1981, and more recent estimates may yield even higher values. The World Bank data for 2010, included in the Migration and Remittances Factbook, states that there are 564,906 Nepalese in India. It’s unclear what to make of these varied figures: the number of Nepal-born in India may be anywhere between 500,000 (~2% of the Nepalese population) to 5 million (~20% of the Nepalese population). The size of the Nepalese diaspora more broadly defined as descendants of Nepalese is probably at least 5 million, though probably not more than 10 million.

Recency of divergence

Indian GDP (PPP) per capita has been modestly higher than Nepal for quite a while, but the difference was fairly small until the Indian economy began liberalizing. As recently as 1992, one data set shows Nepalese GDP (PPP) at $800 per capita and Indian GDP (PPP) as $1200 per capita — hardly a big difference. India’s geographical and historical advantages, and more importantly, its much larger population, have always given it more growth potential, but the potential started getting unleashed only with liberalization measures in the 1990s. This caused the Indian economy to gallop ahead in relative terms, leading to the current situation where Indian GDP (PPP) per capita is $3800-4000 while that in Nepal is $1300-1500.

The recency of divergence might explain why Nepal hasn’t caught up with India and also why there aren’t more Nepalese in India: the Nepalese haven’t yet fully adjusted to the knowledge of India being a more attractive destination for economic opportunity. In other words, although India and Nepal have had open borders since forever, the “open borders with a large income disparity” is a relatively recent phenomenon, about as recent as the addition of poorer member states to the European Union. If Indian economic growth continues, we should expect to see Nepalese migration pick up, and I’d also expect that the gap in GDP (PPP) per capita is unlikely to widen beyond the 3X level.

Paul Collier’s diaspora dynamics model suggests that migration flows at any given point in time are far less than what either polling data on migration or economic models (such as John Kennan’s) predict.

However, if borders have been open for a sufficiently long time, then the fraction of the population with genetic origins in the source country that is in the target country does come close to the high levels suggested by polling data and economic models. Now, Nepal and India have had an open border for quite long, but the recency of divergence suggests that the “open border with huge economic disparity” is a relatively recent phenomenon for the country pair, and the size of the Nepalese diaspora in India at present is not inconsistent with the levels predicted by diaspora dynamics models (I’m just using a crude sense of the numbers and haven’t done any formal quantitative checks).

Prosperity of nearby states

One possible explanation for the wide disparity and low migration rates despite the disparity is that the Indian states near Nepal aren’t that prosperous. Consider this list of gross state domestic products. The five Indian states that share a land border with Nepal are Uttarakhand, Uttar Pradesh, Bihar, West Bengal, and Sikkim. Of these:

  • Bihar, one of India’s worst performing states, has a per capita GSDP pretty similar to Nepal’s (emigration from Bihar to the rest of India is significant, though I haven’t compiled quantitative measures). The state has also historically been linked with violence based on caste and class lines. It’s understandable that this is not an attractive target for Nepalese migrants.
  • Uttar Pradesh, one of India’s largest states, with a population of about 200 million (so only four countries have a higher population) has per capita GSDP about half India’s national per capita GDP. Assuming the same PPP adjustment factor, this would put the per capita GSDP at about $1900-2000, somewhat but not significantly more than Nepal. This is not a very attractive target for Nepalese migrants.
  • West Bengal and Uttarakhand have per capita GSDPs fairly close to the Indian average. These states should at least in principle be attractive targets.
  • Sikkim has a per capita GSDP about twice the Indian average, but it is a very small state, and India’s least populous. The most recent population estimate (based on the 2011 census) is 610,000, about 0.3% the population of Uttar Pradesh. It is the only state in India with an ethnic Nepali majority. The economy runs largely on agriculture. In recent years, tourism has been promoted. Despite being well-to-do, Sikkim may not offer many job opportunities to facilitate large-scale Nepali migration beyond the level that’s already occurred.

(The question of why significant economic disparities between Indian states persist despite internal open borders will be the subject of another blog post).

Apart from going to the immediate bordering states, Nepalese also go to other parts of the country, including the major metropolitan cities such as the capital cities, as well as the north-eastern states. But the lack of economic attractiveness of the states very close by could be part of the reason why Nepalese migration isn’t greater in magnitude.

Lack of information

Until I looked up the data, I wasn’t aware of the huge GDP (PPP) per capita disparity (I had been aware of the level for India, but not for Nepal). And I’m relatively well-informed about economic matters. Is it possible that many Nepalese simply aren’t aware of, or haven’t given active thought to, the prosperity levels in India? The recency of divergence is important here: information may take time to percolate.

Possible reasons for information not having flown fast enough within Nepal:

  • Low population density, making it more difficult for information to flow quickly (the population density is actually not all that low — at about 180/km^2, it is in the mid-range, but Nepal doesn’t have dense population concentrations where information flows really quickly)..
  • Low teledensity (less than 50%) suggesting that information about the outside world and about friends and relatives living abroad may not flow in as quickly and in as much detail as it needs to for people to be inspired to migrate. For comparison, teledensity in India is about 70% and teledensity in First World countries is close to 100%.

Explanations for differing attitudes in India to immigration from Nepal and Bangladesh

The following explanations have been posited for why India has an open border with Nepal, with little political resistance to it, yet a lot of unrest over illegal immigration from Bangladesh, despite Nepal being poorer than Bangladesh.

  • Population: Bangladesh has a population of 150 million, about 5-6X the population of Nepal. So, having open borders with Bangladesh is (considered) less feasible, or at any rate, would be a bigger and more transformative change.
  • Greater cultural similarity propelling more migration: Bangladeshis share close cultural roots with West Bengal (indeed, Bangladesh and West Bengal were both part of the state of Bengal in British India). Thus, there is likely to be much greater migration of Bangladeshis since they may have more confidence they’ll be able to adjust to life in West Bengal.
  • Religion: Bangladesh is an officially Muslim country with a Muslim majority. Although not as hostile to India as Pakistan, it still has some hostility. Nepal is a Hindu majority country with small amounts of Buddhism and Islam — religious demographics very similar to India.
  • Historical accident: Bangladesh and India actually started off somewhat well, because India supported Bangladesh (then East Pakistan) in its struggle for independence against West Pakistan (~1971). But political changes in Bangladesh led to a worsening of relations.

We’ll explore Bangladesh and India more closely in a subsequent post (UPDATE: The post is here).

Nepalese in Bhutan

India also has open borders with a very small country called Bhutan (population ~750,000). Bhutan has a much higher GDP (PPP) per capita, somewhere in the $7000 range, but the economy runs largely on tourism, suggesting that it may not scale well with increased population. A large number of Nepalese have migrated (via India) to Bhutan, and Nepalese form a significant minority in Bhutan. Although richer, Bhutan also has more political oppression, so many Bhutanese (including some of Nepalese origin) have sought refuge in Nepal (see here).

Related reading