VoxEU Column Migration

August 1914 and the end of unrestricted mass migration

When nations declared war in 1914, migration policies changed. This column describes some of these changes and how they affected later migration incentives and patterns. Before 1914, migration had been peaceful and driven by market incentives. Since 1914, it has been shaped by politically determined quotas, legal restrictions, and flights from wars and oppression.

The war declarations of August 1914 spelled far-reaching alteration to the fundamental character of modern long-distance international mass migration. For most of the preceding century, in the majority of big economies international human relocation had been largely peaceful, voluntary, and motivated by market incentives. Since 1914, it has been mostly shaped by politically determined quotas and legal restrictions, or driven by flight from war, oppression or similarly fearsome dangers and disasters (Hatton and Williamson 2005, Massey 2000).

These changes in migration policies and patterns had precursors before 1914, and were to be further propelled by warfare and economic upheaval in the 1920s, 30s, and 40s. There had been significant refugee flows in the Franco-Prussian War and the Balkan wars of the late 19th and early 20th centuries. After the 1918 armistice, the Russian Civil War, ethnic cleansing in the Near East, and the rise of fascism sent millions more into flight. Nearly all migration from China to America had been halted by ethnically explicit legal restriction already in the 1880s, although permanent quotas and passport requirements were widely applied to non-coerced migration only from the 1920s on.1

Changes in migration policies following 1914

The events of late 1914 nonetheless mark a central turning point. The long-term demise of widespread open borders to peaceful human relocation was signalled almost as soon as the ‘lamps went out all over Europe’ in August. When Europe’s dominant naval power and its foremost land power embarked upon all-out conflict with each other, peacetime commercial competition between Britain and Germany for transporting the transatlantic core of global migration gave way to blockades, torpedoes, and troop ships. Mass international labour migration, which had bolstered urban and industrial growth across the North Atlantic basin for decades, declined as millions of mobilised soldiers marched to war. At either end of the dominant European front, neutral Netherlands and Switzerland became prime migration corridors for homebound expatriate workers and war refugees in unprecedented volumes.

At Ellis Island, arrivals fell to their lowest levels in many decades, while returnees rushed back to Europe in record numbers. The giant immigrant-carrying vessels of the North Atlantic commercial shipping lines were interned, impounded and/or converted to military use, and transatlantic passenger services were cut back. By 1915, voyages from Europe to the US had fallen by more than 70% from 1913’s level; steerage passenger arrivals had dropped by over 90% (Keeling 2012b).

A partial resurgence of transatlantic migration to the US after 1918 was short-lived because the war also significantly changed the American economy and US public opinion. Support for unrestricted international migration declined and opposition to it increased. German-American voters –influential supporters of open borders before 1914 – were less active politically after the US entered the war against the Kaiser in 1917. After the war, foreigners in general were viewed with greater suspicion in America. US businesses, meanwhile, had developed lasting alternatives to overseas immigration, relying more on migrants from Canada, Mexico and rural US states, on female employees, and on increased machinery itself, which – it was pointed out – was not at risk of going on strike, moving to the next town, or relocating back to Europe (Higham 1988). In the 1920s, American laws changed. Limited qualitative exclusion of migration was permanently replaced by strict quantitative limits.

Cross-border labour migration resumed again in Europe after 1918, but on a more limited scale. The war precipitated by an ‘incident in the Balkans’ had led to economic balkanisation in Central Europe. Sizable refugee flows continued and the new League of Nations established the first High Commission for Refugees in 1921.

Global migration, overwhelmingly governed by labour markets and family networks before 1914, has not returned since. Voluntary transnational migration nowadays is also sizable, and co-evolves with business patterns and network feedbacks as before, but does so under mostly restrictive policies rendered less predictable by the politics of ‘strange-bedfellow’ coalitions. Free-market advocates typically join multicultural progressives in seeking relaxation of border restrictions, while labour unions, nativists, and environmental groups are generally more reluctant to see those controls attenuate. Multinational corporations are laboratories of legal alternatives to mass labour relocation, while undocumented migration tends to weaken the efficacy of quotas and restrictions without undoing them (Massey 2000, Facchini and Testa 2011, Long 2013).

As salient as the 1914 sea policy change undoubtedly was for the long-term history of international migration, however, this should not obscure important influences running in a reverse causal direction. Because migrant workers are often close to the pulse of an economy, their movement can at times serve as a barometer and bellwether of major social, economic, and political trends.

Modern labour migration is the change in the stock of a segment of the workforce typically overrepresented in temporary and cyclical employment, and migrant networks are usually attentive to hiring trends for those sorts of jobs. Transatlantic shipping line agents and analysts in the early 20th century, for example, often took the level of prepaid tickets as a leading indicator of cyclical fluctuations, which their cyclical steerage business tended to reflect in magnified form. In the US recession of 1907-08 – the sharpest downturn of the decade prior to World War I – factory production was about 20% lower compared to the year earlier, and employment fell roughly 10%, but net migration dropped by about 150% as a large net outflow of migrants returned to Europe (estimates for 1907-08 based on Keeling 2012b, Keeling 2012a).

The leading feature of international mass migration was evident in the abrupt curtailment of immigration in August 1914, and in the prompt and concomitant new mass flows of repatriates and refugees, particularly out of Belgium. These were early signals that the new European War had the potential to lead to considerably more radical economic disturbance than prior wars had.

In his famously prescient 1919 book, The Economic Consequences of the Peace, John Maynard Keynes described the open borders of the then bygone first age of globalisation before WWI:

“The inhabitant of London could order by telephone, sipping his morning tea in bed, the various products of the whole earth, in such quantity as he might see fit, and reasonably expect their early delivery on his doorstep, he could at the same moment and by the same means adventure his wealth in the natural resources and new enterprises of any quarter of the world, [and] he could secure forthwith, if he wished it, cheap and comfortable means of transit to any country or climate without passport or other formality. (Keynes, quoted in Paxton, 3 [emphasis added]).”

A quarter of a century and another world war later, Keynes helped establish an institutional framework to gradually restore key components of the pre-1914 international free trade in goods, services, and finance. Returning to free cross-border movement of labour has been more recent, limited, and tentative, however. There are powerful forces today pushing migration hurdles lower or aside, not least the ongoing information convergence in the ‘global village’, but opposite forces seem nearly as strong. Unlike goods and money, international flows of people are accompanied by social, political and cultural baggage, and the act of relocation transforms migrants, their sending locales, and their destinations, and engenders feedbacks, side effects, and second thoughts.

Migration processes are often controversial, difficult to predict, and dependent upon fragile political compromises. Especially in a world of considerable and increasingly undeniable inequality, economic self-improvement is a powerful motivator, but fear can often trump it, particularly during periods of economic uncertainly and cyclical joblessness.

Concluding remarks

By most indications, globalisation in some form seems likely to persist for the foreseeable future, although the golden age of open borders described by Keynes does not look retrievable any time soon, if ever. Compared to their early 20th century counterparts, early 21st century societies face a quite different constellation of vulnerabilities with a quite different complex of resources and opportunities. We can, however, look back before the shock to globalisation in 1914 with a sober hindsight that might contribute, if only indirectly, to improved foresight beyond 2014.

In the long arc of world history, a half millennium of population relocation centred on movement from Eurasia to the Americas was liable to be a one-time demographic reallocation regardless of geopolitics. But, as it actually happened, unfettered transnational migration – driven overwhelmingly by economic opportunity and operating on a massive worldwide scale – did not gradually phase out (as it had developed) over decades and centuries. To a considerable and lasting degree, it was lost, and in short order, with the generation lost following the guns of August 1914.

Editors' note: This is the third in a series of Vox columns by leading economic historians on the First World War, which will be collected in a Vox eBook at the end of the year: "The Economics of the First World War", edited by Nicholas Crafts, Kevin O'Rourke and Alan Taylor.


Facchini, G and Cecilia T (2011), “The rhetoric of closed borders”, VoxEU.org, 28 April.

Gould, J D (1979), “European Inter-Continental Emigration, 1815-1914: Patterns and Causes,” Journal of European Economic History 8, 593-679.

Hatton, T J and J G Williamson (2005), Global Migration and the World Economy: Two Centuries of Policy and Performance, Cambridge: MIT Press.

Higham, J (1988), Strangers in the Land: Patterns of American Nativism, 1860-1925 (2nd edition), New Brunswick: Rutgers University Press.

Keeling, D (2012a), “Return Migration from the United States to Europe During the Recession of 1907-08,” In Krisen-Ursachen, Deutungen und Folgen, edited by Thomas David, Jon Mathieu, Janick Marin Schaufelbuehl und Tobias Straumann, 245-60, Schweizerisches Jahrbuch für Wirtschafts- und Sozialgeschichte 27, Zurich: Chronos.

Keeling, D (2012b), The Business of Transatlantic Migration between Europe and the United States, 1900-1914, Zurich: Chronos.

Long, K (2013), “When refugees stopped being migrants,” Migration Studies 1(1), 4-26.

Marrus, M (1995), The Unwanted: European Refugees in the Twentieth Century, New York: Oxford University Press.

Massey, D (2000), “To Study Migration Today, Look to a Parallel Era,” Chronicle of Higher Education, August 18, p. B5.

Paxton, R (1985), Europe in the Twentieth Century, Fort Worth: Harcourt.


1 Marrus,(1995), indicates that, in Europe alone, World War II generated some sixty million refugees, a tenfold increase from the numbers in World War I, and well above the circa 50 million peacetime overseas migrants (Gould 1979) during the entire century, 1815-1914. Between 1815 and 1935, Europe’s population grew roughly two and a half fold.