Climate Change, Poverty, and Migration, Part 2: The Evidence

by Quentin Wodon

How devastating are weather schocks associated with climate change for the populations affected? Are the populations able to cope with and adapt to the shocks? Do they migrate away from the affected areas? These are the questions that were considered in two recent studies, one for the Middle East and North Africa (MENA), and the other for the Sundarbans in South Asia. The book for the MENA region was published last year and is available electronically free of charge here. The book on the Sundarbans was published last month – it is not yet open access, but I received quite a few copies, so if you would like a print copy, please send me an email with your address through the Contact Me page.

Recall that the argument outlined in the first post in this series has three parts: (1) households often suffer from large losses when affected by weather shocks; (2) they have limited ways to cope and adapt; and yet (3) hosueholds affected more do not have higher migration rates away from the areas. In this post, I will share  evidence that backs up these conclusions. In the next post, I will discuss implications of the findings for humanitarian organizations and service clubs.

Impact of Weather Shocks on Households

Surveys were implemented in climate-affected areas of the MENA region and in the Sundarbans. In the MENA surveys, households could identify droughts, floods, storms, mudslides, excessive heat, excessive rain, pest infestation, crop and livestock diseases, and other events as having affected them. In the Sundarbans, the questionnaire identified cyclones, droughts, floods, and other events. Almost all households said that they had been affected by adverse weather events in the five MENA countries as well as in the Sundarbans (the exception was Egypt where the proportion was smaller, but still above two thirds). When asked which type of adverse event had the largest negative consequences for them, households in the MENA countries cited droughts first, followed by excessive heat and floods. In the Sundarbans, households cited mostly the cyclones.

In the MENA region, more than half of the respondents said that changes in weather patterns had led to a loss of crops, and more than a third reported a loss of income. About a fourth reported a loss of livestock. Poorer households were more likely to be affected than richer households, which is not surprising given that poorer households are more likely to be farmers and rely on rainfed agriculture. In the Sundarbans, more than a third of households declared that their dwelling had been completely destroyed by cyclones. Another third said that their dwelling had been partially destroyed. All these losses are rather large, showing that the populations in the areas surveyed are highly vulnerable to extreme weather shocks that indeed have dramatic consequences.

Ability to Cope and Adapt

How do households cope with losses generated by weather shocks? Questions were asked about what households did in the recent past and what they would do in the future in case of a new shock. In the MENA region, almost two thirds of households declared that they have used or would use their savings to cope. Close half have sold or would sell assets, have asked or would as for loans, and have sold or would sell livestock. More than a third have withdrawn or would withdraw their children from school. In the Sundarbans, responses were similar, but the proportion of households declaring that they had withdrawn their children from school or would do so in case of a new shock was smaller. Overall, these coping strategies make sense, but they also show how households often must take desperate measures – such as selling productive assets – to survive. Such coping strategies may have long term negative consequences for the ability of households or their children to emerge from poverty. In addition, in a separate survey for Morocco households declared that they had not been able to recover from losses associated with the weather shocks years after the events.

The data also suggest that households are often not able to implement adaptation strategies to deal with extreme weather shocks and a changing environment. For example, some twenty different adaptation strategies were listed in the MENA questionnaires, and only a minority of households declared implementing these strategies. Community and government programs were also scarce, with the exception of a few government-led safety net programs in the Sundarbans.

Decisions to Migrate

One would expect that if households are repeatedly affected by weather shocks, and if they have limited ways to cope and adapt, younger household members would migrate away. The evidence however suggests that households more severely affected by the shocks are not much more likely to send members away than other households.

  • When households are asked in the surveys about the reasons why some of their members migrate temporarily or permanently, reasons directly related to climatic factors such as droughts, floods, or cyclones do not come up much –socio-economic factors, such as better opportunities for employment at the place of destination, are cited much more often.
  • Regression analysis suggests that the impact of perceptions regarding changing climatic conditions as well as actual weather shocks on migration is small. Households who perceive worsening climatic conditions do not have much higher permanent migration rates among their members than households who do not perceive such changing climatic conditions to the same extent. The same holds for the severity of the impact of whether shocks on households – temporary migration is affected (more severely affected households have more temporary migrants), but this is less the case for permanent migration.
  • Analysis of census data suggests that “push” factors such as climatic conditions at the place of origin (temperature and rainfall) affect migration patterns only in a limited way, with again socio-economic “pull” factors at the place of destination playing a much larger role in migration movements.
  • In the MENA region, qualitative focus groups help explain why permanent migration does not seem to be higher amomg households affected more severely by extreme weather events and changes in the climate: many migrants are simply not doing very well at their place of destination.

Overall, these various strands of evidence provide back-up for the argument that many household members in climate-affected areas may be unable or unwilling to migrate away, at least under present conditions. Many households may be trapped where they are with all the risks that this entails as climatic conditions continue to worsen in the future. In the last post in this series, I will discuss the implications of such findings.

Climate Change, Poverty, and Migration, Part 1: A Major Risk

by Quentin Wodon

Members of service clubs involved in international projects (global grants in the case of Rotary), as well as those interested in development and poverty eradication more generally, need to be aware of broad trends affecting the livelihood of the populations they are trying to serve.

In this series of three posts, I will share results from recent research on climate change, poverty, and migration. In this first post, I will outline a basic argument about the risk that the poor may not be willing or able to leave areas affected by the extreme weather events that climate change will exacerbate in the future. In the second post, I will provide evidence to back up this argument from two regions: the Middle East and North Africa (MENA) and the Sundarbans in South Asia. Finally, I will discuss in the third post the implications of the research for policies and programs. For those interested, the details of the analysis are available in two books, one on the MENA region published last year (open access), and the other on the Sundarbans published last month.

Let’s start with the big picture. Climate change is one of the most significant threats faced by the developing world (see for example the Turn Down the Heat reports). By the end of this century, global mean annual temperatures may increase by 4°C.  In many countries climate change will manifest itself through reduced rainfall, greater seasonal temperature variability, a rise in sea levels, and a higher frequency of extreme weather events. These events constitute threats to people’s ability to continue to live where they are living today, their economic security, and political stability.

Declining precipitation will affect the availability and usage of water, causing agricultural productivity to decrease and poverty to rise. Climate change will also lead to a higher likelihood of extreme temperatures, floods, droughts, and cyclones, depending on the geographic area affected, and thereby risks of substantial displacement as well as deteriorating environmental conditions. How climate change will affect specific regions remains debated. But threats to the MENA region – the first area considered here – include more severe droughts and floods as well as rising temperatures and water scarcity. In the Sundarbans, the second area that I will consider, threats include a sea level rise and more frequent and severe cyclones.

Do households living in climate-affected areas in both regions believe that changes in climate patterns and the environment are taking place? To what extent have households been affected by extreme weather events and what has been the impact of these events on them? What are the mechanisms that households use to cope and the strategies that they rely upon to adapt? Do households benefit from community and government programs that can help them cope and adapt? Are they warned ahead of time of weather shocks? To what extent do remittances reach households living in climate-poor areas and what is their impact on poverty and human development indicators? Finally, to what extent are perceived changes in the climate and weather shocks leading to more temporary and permanent migration? These are the questions I will explore.

But first, what does the literature suggest? A recent Foresight report suggests that while environmental change will lead to an increase in migration, because of the complexity of so-called “push” and “pull” factors involved in migration decisions, in most cases it will not be feasible to identify “pure” environmental migrants. The report also suggests that some vulnerable groups may be either unwilling or unable to migrate away from affected areas, among others due to the cost of migration and the fact that environmental change and weather shocks may result in large losses in income and assets, rendering migration unaffordable. Thus, some population groups may well be trapped in climate-affected areas with progressively deteriorating conditions for their livelihood.

In order to explore the relationship between climate change, extreme weather shocks, and migration, we implemented at the World Bank new households surveys in areas affected by weather shocks and changing climatic conditions in seven countries: Algeria, Bangladesh, Egypt, India, Morocco, Syria, and Yemen. The five MENA countries were chosen in order to be illustrative of the region as a whole, with some countries affected mostly and severely by droughts, others by both milder droughts and floods, and still others less affected until now

While the MENA countries are affected mostly by droughts, and to a lower extent floods, the Sundarbans in Bangladesh and India (West Bengal) are affected by cyclones, sea water rises, and salinity intrusion in arable land. The contexts in which households are affected by changing climatic conditions and extreme weather events are thus very different in the MENA and South Asia regions. But as it turned out, while circumstances and contexts differ, many of the findings were similar in the two sets of countries. The main conclusions from the analysis can be summarized along three main points as follows:

  • The areas surveyed tend to be repeatedly exposed to weather shocks. Households do perceive a worsening of their environment in terms of the frequency and severity of those shocks. Most households in the areas surveyed have personally been affected negatively and substantially by those shocks, in terms mostly of losses of crops, income, and livestock in the MENA region, and losses in dwellings in the Sundarbans. The poor in both regions typically suffer the most.
  • The ability of households to cope with those extreme weather events and the losses in income, crops, or dwellings that they cause is limited. Relatively few households engage in medium- and long-term adaptation strategies, and remittances, while useful, do not necessarily reach climate-affected areas more than other areas. Support received from communities and government programs is limited, with the exception of safety nets in West Bengal and Bangladesh. Finally the extent to which households are warned ahead of time of imminent shocks differs between areas.
  • As a result of limited ways to cope and adapt, temporary and permanent migration could represent a strategy for household to deal with difficult climatic conditions and extreme weather shocks. The data suggest that today temporary migration is indeed used by affected households to cope and adapt. However the share of the observed temporary migration that can be directly attributed to climatic conditions tends to be small and the link between climatic conditions and permanent migration is even weaker. Essentially, whether this is due to an attachment to their area of origin or a lack of good options at destination for migrants, households who are affected the most by poor climatic conditions and weather shocks do not appear to be sending members away permanently much more than other households living in the same areas but affected less by those shocks. This does not mean that in the future climate-induced migration may not be much higher.

This basic argument, if valid, has implications for development policy in climate-affected areas, and likewise it has implications for the types of projects that service club organizations should implement in those areas. In the next post in the series, I will provide more details on the empirical evidence about the argument outlined above. In the third post, I will discuss implications of this research for development policies and programs, including for service clubs.