Dr Will Jones, Senior Lecturer in International Relations, Royal Holloway University of London – Written evidence (ZAF0055)


1. What are the main drivers of irregular migration (both forced and voluntary) in Sub-Saharan Africa? How significant as factors are conflict and climate change?


It is important to remember that most African migration is within Africa. In 2019, over 21 million Africans were estimated to be living in another African country:


Another thing this makes clear is that migration within and from Africa is rising. There is no reason to believe that is going to change.

Why are people moving then? Identifying single drivers is notoriously difficult, and almost all migration flows are ‘mixed’ in character (i.e. a combination of economic, security, and personal motivations). As some countries becomes wealthier, migration becomes easier for some. Migration is always a combination of negative factors ‘pushing’ people, and material resources enabling them. For example, one of the strongest predictors of ongoing migration destinations is previous migration to that destination (for the simple reason that over time connections, networks, knowledge, and organisations emerge to help people migrate). This implies (and the research confirms) that attempts to alter migration routes through policy instruments is an extremely blunt instrument at best.

What is also clear from these figures is that forced migration is a clear minority of cases, and asylum seekers from Africa to Europe are a drop in the bucket in policy terms. This does not mean that African asylum seekers are largely mala fide. They are not. It means restricting asylum will have a negligible effect on migration overall.

Importantly, much of this migration is also extremely temporary (i.e. under two years, often much less than that) because regional economies produce cyclical demand for labour. For example, the South African agricultural sector is sustained by huge communities of irregular migrant labourers, particularly from Zimbabwe and Mozambique. Research has often shown that Somalis in Kenya also move in large numbers back and forth as conditions of physical safety and labour demand wax and wane. This is a good thing. A relatively flexible open border where individuals are able to move across safely without fear of losing hard-won status (which they will preserve as a back-up at all costs) prevents short-term displacement turning into long-term exile and humanitarian dependency. Counterintuitively, therefore, keeping a border open usually reduces the long-term costs to the hosting state.

The quality of the statistical data we possess connecting migration to government policy has undergone a quantum leap in recent decades (I am thinking particularly of the DEMIG: Determinants of Migration project at the University of Oxford, which brought world-leading expertise to bear on the best data we have ever assembled on migration and public policy. If you have not read it you should). Crucially, it suggests that changes in government policies in destination countries have at best a marginal effect. There are some effects: for example, the announcement of tightening of migration rules produces rushes of individuals keen to get in before the rule change takes effect, and stops people returning (because they fear losing status). This is compounded a further marginal effect of tightening controls: migrants engage in categorical substitution (i.e. they use another legal status such as family reunification, or simply migrate completely illicitly). This is particularly damaging when restricting safe legal routes for asylum simply produces a bonanza for organised crime, and increases mortality rates on journeys, without significantly reducing figures for entry.

Therefore, it is relatively clear that restricting access to asylum will have a negligible effect on migration, and focussing on asylum to Europe at the expense of considering other much larger forms of African migration would be a mistake.


Secondly, most current migration within Africa is not (directly) conflict-driven. African migration has risen dramatically since the ‘90s, despite a continent-wide decrease in intra-state and inter-state violence. Nonetheless, conflict-migration is a paradigmatic low-probability high-impact event. When it does happen, we see refugee flows which are rapid and vast in scale (e.g. the millions who crossed into Congo in the wake of the Rwandan Genocide). Such flows are particularly unmanageable, particularly brutal in terms of human suffering, and particularly destabilising to the regional theatres in which they occur. Conflict migration is normally largely regional in character, unless there is a breakdown in humanitarian and peacekeeping infrastructure, or large-scale repression, which (obviously) leads people to move on. For example, we have clear evidence that what prompted Zimbabwean refugees in South Africa to move in larger numbers to Europe and North America, or Eritrean refugees in Ethiopia to attempt extraordinarily dangerous journeys to Europe through the Sahara and the Mediterranean, was not changes in their home country, but worsening conditions in their country of residence. Supporting hosting countries in Africa is therefore the simplest, cheapest, and most effective way to decrease onward migration.

Where this is not possible, the alternative is large-scale resettlement of such individuals out of the region. This has been practiced before with impressive results (as when the Comprehensive Plan of Action for Indo-Chinese Refugees resettled over three million refugees from South East Asia into the West, largely without negative effects on security, the public purse, or social cohesion).


Thirdly, it is clear that there is already a significant degree of ‘climate migration in Africa driven by depleted land, irregular rainfall, soil degradation, and the ensuing fall in agricultural yields. This will often form part of the bundle of motivations which lead individuals or families to migrate. Fortunately, climate change has not yet produced the truly terrifying levels of displacement we have seen in other regions, notably South East Asia. Because it is obvious that climate change will get worse, this will get larger.

Some state policies will make this worse. Industrial development plans are often extremely expensive in terms of water consumption and disruption to land (e.g. the Sudanese government’s extremely aggressive plan of industrialisation built on dam construction). This will obviously increase ‘push’ migration pressures.

It is also worth noting that Africa is particularly vulnerable to large-scale ecological collapse produced by comparatively minor rises in global temperatures, for structural reasons to do with Africa’s geography. Obviously, there will be considerably more than minor rises in global temperatures. Furthermore, climate change and conflict are not unrelated. There is now an extensive scholarly literature showing how climate change exacerbates social conflict, weakens governance, and leads to violence (which undoubtedly leads to large-scale migration).


2. Which countries are the most vulnerable to these drivers, and which routes would this migration be likely to take?


The figures make clear that migration is increasing and will increase everywhere. Focussing on particular areas is probably therefore not that helpful, except with respect to particular categories of migrants, such as refugees. For refugees, barring large-scale violence or democratic backsliding emerging elsewhere, the overwhelming majority of forced displacement will continue to be in the zone demarked by the Great Lakes region in the south and Somalia in the north (i.e. from East-Central Africa to the Horn).


More generally, migration routes are driven by availability, and the resources and risk-aversion of migrants themselves.

On resources: for example, when extremely poor families have to move very rapidly (most obviously in cases of conflict migration), most of it will be within region: when violence in Congo flares up people run to Uganda, Rwanda, or Tanzania; those in South Sudan walk south to Uganda, during the Zimbabwean crisis some three million moved to South Africa; violence in Somalia produces larger numbers of registrations in camps in Kenya; and so on. Those routes are relatively fixed, because they are picked by individuals in exceptionally straightened circumstances. However, again, that is the minority of cases. Most migration is not conflict-driven. Therefore such migration is conducted by individuals with the capacity to – at least to some degree – wait and see, consider options, and plan.

Furthermore, one’s first destination is often not one’s last. Onward (or ‘relay’) migration is common, and is generally driven by a deterioration of conditions in the host country (e.g. the decisions of Eritreans to move onwards from Ethiopia to North Africa or Europe). Those conditions can be economic (availability of land or jobs), or political (large-scale repression, withdrawal of rights, etc).

It is important to also return that return can be considerably harder than onward movement. A significant example of this is the situation of sub-Saharan migrant workers in Libya. The relative economic dynamism of Gaddafi’s Libya had meant there was a considerable population of migrant workers, largely from West Africa, at the point of the 2011 civil war, intervention, and subsequent collapse of the country. Libya at this point witnessed large-scale xenophobic violence directed against ‘foreign black Africans’ which led many to attempt entry to Europe via the Central Mediterranean sea route. Such individuals were not committed necessarily to entering Europe no matter what, but that route was considerably less dangerous than attempting to return to their country of origin. At the time much effort was expended by European states on attempting to deter and control entry, whereas at it at least possible that an approach which focussed on assisting and protecting populations in-country, and facilitating genuine voluntary return would have done more to decrease numbers of irregular arrivals in Europe at lower cost and with considerably more regard to the rights of the individuals involved.


On availability, migration routes follow networks and institutions. Networks of individuals (particularly kinship) are the key channel whereby people acquire information they trust about the society to which they are considering moving (e.g. can my qualifications be used? What are the schools like? How racist is it?), crucial practical information about how moving works (e.g. can you help me with this form? Should I apply for this status or this?), and also the resources at the other end which they will rely on in the first stage after they have moved to their destination country (e.g. a place to stay in the first instance, or a pre-made network of friends who speak one’s first language, etc). Over time, migration from a particular origin to a particular destination will also lead to a set of institutions springing up: travel agencies, credit transfer agencies, transnational religious, cultural, or social organisations, lawyers who specialise in migration from these two nodes, etc. This can also include informal organisations (e.g. Eritreans make extensive use of kinship networks to transfer debt and credit) or – where migration is criminalised – criminal networks (people trafficking organisations will normally have routes they specialise in). Once these networks and institutions are in place only extremely repressive actions can disrupt them.

This is why the statistical evidence is clear that one of the strongest predictors of migration destinations is the presence of prior migrants from that country. For example, it is largely down to luck that New York became the residence of the first Italian migrants to the USA, but after that it became a key nexus for Italian migration to the United States in the 19th century, because that is where Italian migrants could find the communities and institutions they needed to succeed. It is extremely hard for governments to convince individuals that the information they get from personal networks and institutions is not correct. Some governments have tried (for example, the Australian government ran public information campaigns seeking to deter Afghans from attempting to move to Australia), but I cannot think of any examples where this has succeeded.

If closing routes is hard, opening routes is comparatively easy. States can do much to provide alternative routes for safe legal migration. For example, the South African government in the 2000s consistently worked to support the government of Botswana to incentivise them to retain their (relatively) open policy towards Zimbabwean refugees, despite considerable public antipathy to hosting large numbers within Botswana. This usually amounts to supporting regional inter-state cooperation for enabling migration, and is the most effective way of preventing untenable concentrations of refugees in a single area which a host government cannot deal with. It is usually the state inability to provide for a refugee population, not factors about the population itself, which leads to violence, and large amounts of onward migration.


3. What is your assessment of the EU approach to irregular migration from Africa to date? Are there opportunities for a different UK approach now that the UK has left the EU?


It is telling that this question says ‘from’. I will answer that first, but I think a key missed opportunity has been to focus overwhelmingly on controlling entry at Europe’s borders, to the detriment of thinking about migration in Africa more generally. I will try and persuade you that considering African migration more broadly would make it possible for the UK to realise another set of benefits, but would also serve the goal of controlling entry better than our prior policies.


The question as written first: the European Union has, since the political crisis produced by high volumes of asylum seekers in the ‘90s, erected some of the most comprehensive and restrictive border controls on the planet, probably only exceeded by Australia. This is commonly known as the non-entrée regime. Better scholars than I have described this at length, but a basic list would include:

-          Militarised physical borders at Europe’s external land borders (Ceuta and Melilla).

-          A naval presence in the Mediterranean, particularly since the arrival of some 280,000 arrivals via the Central Mediterranean route in 2013, which saw the end of Mare Nostrum and its replacement with Triton (in essence, a shift from a small humanitarian operation to a larger one which conducts interdiction at sea).

-          The deployment of an increasingly recondite array of drones, satellite aircraft, remote tracking systems, and satellite remote sensing.

-          A complex series of legal reforms in most countries in Europe which narrowed the grounds for asylum, limited legal opportunities for appeal, increased scope for immediate rejection of ‘manifestly unfounded’ claims, expanded the list of ‘safe countries’ to which one could be summarily returned, expanded financial requirements for African migrants, narrowing definitions of family, limiting access to welfare and social support, etc. In the UK the key examples are the Immigration and Asylum Act of 1999, the 2002 Nationality, Immigration, and Asylum Act, and the 2004 Asylum and Immigration Act. Despite this enormous effort, this had very little effect on overall migration figures (not least because asylum seekers are such a small proportion of the overall number).

-          The creation of legally-grey ‘transit zones’ within European territory where individuals could not claim asylum.

-          Expansion of the bureaucratic institutions of border control, detention during processing (to prevent absconding), tracking visa-overstay, and deportation.

-          A system of ‘carrier sanctions’ which extend responsibility for preventing arrival via plane to the private companies which run those flights, effectively closing that route completely.

-          Numerous extremely costly ‘third party country agreements’ whereby a series of countries are paid large sums to detain migrants, accept deportees, etc (most obviously the €6bn EU-Turkey deal, although there are many many other examples).


This is in one sense extremely impressive. It demonstrates a degree of unwavering purpose, determination, and cross-party agreement that European policymaking is not usually credited with. There are three obvious criticisms of this approach: its consequences in terms of human rights and the rule of law have been subject to widespread condemnation, it has not succeeded in allaying public concern over irregular migration (indeed, many scholars have suggested it does the reverse, by implicitly signalling the problem must be enormous), and it is – by any stretch of the imagination – a truly fantastic outlay of public money (the 2021-’27 EU budget envisages spending some €34.9bn) on a panoply of notoriously ineffective policies.


This will be familiar to you, so I shall not belabour those points. A further consequence of this which is not normally considered is that European governments have overwhelmingly interacted with African governments with this agenda as their sole focus, to the detriment of both. Controlling migration ‘from’ Africa has largely structured the European policy approach more generally, which has become a problem. Prosaically, plenty of research now shows how migration management is now absorbing disproportionate amounts of the money governments have flagged to advance their wider set of aims in Africa. More generally, it means that we have not really spent much time or intellectual energy on the wider set of ways in which we could work productively with African countries on migration issues, many of which would be to our benefit if we did.


If this focus on control could be lessened, there is a lot the British government could consider: it could support regional structures for supporting migration and displacement. These schemes already exist within the SADC, ECOWAS, the EAC, and of course the AU. There is a good argument that Europe’s drumbeat promotion of externalisation (e.g. through bilateral repatriation agreements) has stifled rather than promoted those regional dynamics. This is despite such functioning schemes being ultimately the sort of thing most likely to reduce onward movement to Europe. Regional coping structures will be absolutely essential where there is conflict or climate degradation. What European states ought to want is therefore well-resourced and trained African structures which enable the continent to cope.

The British government is actually historically very good at supporting the technocratic side of these endeavours: it has received justified plaudits for its military cooperation schemes which foster knowledge-transfer and training, for its technical support of government ministries delivering social welfare in Africa, and so on. It has an excellent reputation for this with many African states. It has a set of institutions (such as the Commonwealth) which make it well placed to interact in a fashion that African governments will be receptive to. This is something we are good at. We have not used that advantage in supporting regional arrangements on mobility and migration. Instead, we have actually undermined regional mobility schemes in order to impose stringent controls further down the line. Given this means we have failed to support the kind of structures which might lead people not to attempt onward migration, this has been counterproductive.

So the British government has low-hanging fruit to grab in simply recognising the importance of these structures and supporting them. More concretely, what we are talking about here is regional inter-governmental mechanisms for the smooth regulation and enabling of:

  1. The liberalisation and management of regional migration.
  2. Facilitating safe and easy labour migration, and helping migration management bureaucracies in host countries manage and assist such migrant communities in such a way as to head off problems further down the line.
  3. Well-funded and thought-through crisis-response mechanisms which state bureaucrats are trained in (within this there is loads we could talk about: regional burden-sharing schemes, local resettlement, integration support, camp management, data-sharing and tracking, etc, etc).

Boosting that interim layer between national-level state agencies (without the capacities to respond to what are, after all, regional dynamics) and the continent in general is the simplest way for Britain to engage considerably more usefully and productively going forward.



4. How can the UK work more effectively with the AU and governments in Sub-Saharan Africa on issues relating to economic migration and refugees?


In my answer to the second part of the last question I have outlined an approach which I think the British government could take. As it happens that approach would involve working with the AU and African governments in the manner suggested by this question. So I won’t waste the committee’s time by repeating this.


Instead, I will confine my remarks to one observation about working effectively with African governments, which would be out of place in my answer to [3]. I have now spent almost all of my professional life interviewing African government elites from across the continent. It is difficult for me to overstate how dramatically we are damaging our international reputation, our commercial interests, and our relationships with countries who care deeply about our historical connections, and wish to preserve them. This is not a complaint I only hear from radical activists or those already antipathetic to Britain, but near-universally. I am referring to our rules, processes, and charges for entry. I cannot communicate strongly enough the number of conversations with government ministers and senior civil servants where this issue is raised, apropos of nothing, by individuals who want to like us, and want to work with us. Our rules are regarded as punitive (which makes us look unwelcoming), price-gouging (which makes us look cheap), and unevenly applied (which makes us look racist). Everyone has a niece who has applied for student visa, or an uncle who wanted to attend a funeral, or a friend who was considering a business trip to assess potential suppliers, and so on. It is deeply mistaken to think that such policies only affect the individuals directly named, rather than more generally trashing our reputation within the political and economic elite of a country, who expect to be treated better, and can and will look elsewhere. It is no longer the case, as it was occasionally during the early Cold War, that African elites are zealously committed to ongoing friendly relations with the former metropole. We face a diverse multi-polar donor environment where we will not be able to match the scale of commitment offered by the USA, the EU, or China. It is therefore key that we do not sacrifice what assets we do have, foremost of which is reputation. It is not clear to me that this important raison d’état has been internalised by the Home Office, but it is damaging and will continue to damage British interests abroad if it allowed to continue.

5. What role do diaspora communities play in supporting economic development and funding relief activities in Sub-Saharan Africa?


Diasporas of Africans (both within and outside the continent) play an enormous and well-documented role in supporting development, fostering the reconstruction of societies after conflict, and providing money transfers in emergencies which prevent widespread tragedy. However, I want to try and persuade you that thinking of diasporas solely in terms of economic development and funding relief activities is an error.

Nonetheless, just to recap: the sums of money alone are enormous. The World Bank has attempted to calculate the scale of financial transfers from the diaspora, which is of course extremely difficult, but even accounting for that, the picture is dramatic:

When we talk about diasporas we are talking about far more money than aid, far more money than is raised on the financial markets, and it could easily be more than FDI too in short order. Bear in mind these flows are also going to be localised to countries with particularly large and well-organised diasporas, so this will be even more crucial for some societies, particularly Somalia. Therefore, many have suggested that supporting diasporas is a way to generate outsize impacts for development from comparatively small investments. That apolitical ‘diasporas for development’ discourse is widespread, and is extremely dangerous.


First, there is plenty of evidence that diasporas have a particularly crucial role in post-conflict societies, far beyond just financial transfers, as either ‘peace-makers’ or ‘peace-breakers’. Returnees from the diaspora often possess skills, education, and international networks which individuals present in the home country during the conflict were unable to obtain. In some cases, returnees from the diaspora literally take over the apparatus of the state, as in Rwanda, where the President, much of the Cabinet, and senior civil servants are returned refugees, who brought back with them invaluable technocratic knowledge for the reconstruction. Diasporic returnees are often regarded as having better local knowledge and contacts than donors, and more ‘buy-in’ (after all, they cannot up-and-leave in the event of failure quite as easily as Western donors can), which makes them more committed. This suggests we should be in favour of supporting diasporic associations during post-conflict reconstruction.

But (second): supporting a diaspora may also destabilise politics back home, in a much broader set of cases than just ‘post-conflict’ societies. Diaspora organisations are rarely neutral or apolitical (after all, the scattering of diasporas is usually produced by some traumatic event in the homeland) and European governments have more than once unwittingly supported politically partisan diaspora organisations who presented themselves as politically neutral, to their cost.

This can generate conflict in a variety of ways. In Somaliland, affluent returnees monopolising senior positions and opportunities generated considerable tension and resentment, flaring out into violence more than once. There are now many such societies where there is a considerable social cleavage between returnee and non-returnee. Furthermore, diasporas can literally perpetuate conflict by funding ongoing rebellion, providing a safe harbour to militants, and facilitating the transfer of arms and other materiél. The clearest example (although there are many) of this is support for the LTTE (the ‘Tamil Tigers’) in Sri Lanka’s civil war. Within Africa, it is clear that both the Eritrean struggle for independence and the South African struggle against apartheid relied extensively on resources and capabilities provided to them by their respective diasporas.

But (third): as the Eritrean and South African examples suggest, it may well be that destabilising politics back home is desirable. More recently, the UK has been one of several governments which offered support (overtly and covertly) to Zimbabwean diaspora organisations with the relatively explicit objective of removing the government of Robert Mugabe and ZANU-PF. Although that support did not achieve its primary objective, it ought not be thought of as a complete failure. When states engage in repression, the diaspora often becomes a crucial space where opposition, political debate, and civil society more generally relocates. Foreign governments did much to assist Zimbabweans at a crucial point: they helped Zimbabwean journalists set up online versions of their newspapers in exile, they helped lawyers record and digitise evidence of torture, and – most simply – they provided a place of safety for opposition activists to continue their work without fear of extra-judicial violent retaliation. During the worst of the crisis, the Zimbabwean opposition survived because of the diaspora. We should not think of this as a failure: governments can do much to support pro-democracy forces in the diaspora, although they should be careful their overt support does not delegitimise such actors by making them appear to be the instruments of foreign powers.


Of course, the Zimbabwean government was exceptionally unhappy about this support. Parallel examples abound: the Rwandan government would like many governments to do more to expedite the return of individuals accused of acts of genocide, the Eritrean government would like overseas governments to crack down on criminalised opposition organisations, and so on. The point is the same: it is not possible for states to support or decline to support diasporas without getting involved in the messy business of politics. Going into this unwittingly (i.e. assuming or pretending that diaspora organisations are politically-neutral unequivocally-positive sources of funding) amplifies the risk that such support will blow up in our faces.


6. Are there any other issues you would wish to draw to our attention?


I think you are asking the right questions, and I am very grateful to be allowed to try to be helpful. Please do let me know if clarification of any of the above would be helpful.

Received 18 May 2020