Fertility Intentions in Rural Malawi After Cyclone Idai

Abstract

Environmental disruptions such as extreme weather events can intensify household uncertainty by creating financial, marital, and housing instability. In this paper, we draw on unique interview data to advance knowledge of fertility intentions and the role of environmentally induced uncertainty in shaping them. We use in-depth interviews collected in rural Malawi in July 2019 to examine the motivations behind fertility intentions at a moment of heightened uncertainty, the months following the destruction brought by Cyclone Idai. Among respondents with children, the majority planned to wait at least four years before having another child. From the narratives of parents’ reproductive plans, we develop a typology of respondents who prefer long inter-birth intervals. Analyses reveal that some respondents’ motivations are consistent with traditional spacing intentions, whereas others reflect their fundamental uncertainty about the future and the chances of regaining economic and household stability in a natural-resource contingent context. Our results are foundational for building theory to understand the relationship between environmental forces and fertility.

Studies of fertility intentions in sub-Saharan Africa have emphasized the flexibility and contingency of fertility plans in response to material, relationship, and health conditions and instability (Agadjanian 2005; Hayford and Agadjanian 2019; Johnson-Hanks 2005; Kodzi, Johnson, and Casterline 2012; Sennott and Yeatman 2012; Towriss et al. 2019; Trinitapoli and Yeatman 2018). Climate change introduces a new dimension of uncertainty to a region where environmentally-contingent livelihoods, especially natural resource-based livelihoods, remain common. Climate fluctuations influence short-run birth rates (Davis 2017; Sellers and Gray 2019; Simon 2017; Thiede et al. 2022) but may also reshape longer term fertility intentions and actions by influencing food security, resource stability, and income (Grace 2017). Climate-related extreme weather events, in particular, may create conjunctures that force parents to reevaluate their ideas about what the future will hold. By focusing on the narratives of parents in the aftermath of an extreme weather event, we can understand how environmental experiences are incorporated into reproductive plans.

In this paper, we explore the fertility intentions of parents in Malawi in the period following an extreme weather event. We conducted in-depth interviews in July 2019, four months after Cyclone Idai moved through Malawi and resulted in flooding, infrastructure damage, housing destruction, and severe crop loss (Devi 2019). In the wake of the cyclone, respondents’ reflections on their fertility intentions foregrounded their uncertainty about the future. Questions asking respondents how long they would like to wait until having an additional child generated narratives that centered concern about future rainfall patterns and the household’s prospects for economic recovery. By introducing an intense dimension of contextual uncertainty, Cyclone Idai brought into focus how perceptions of the future influenced fertility intentions¹ and revealed the time horizons over which respondents planned their childbearing.

At the time of data collection, the majority of respondents wanted long birth intervals and planned to wait until their youngest child reached five years of age before having additional children. From the narratives of their reproductive plans, we develop a typology of how fertility intentions were influenced by Cyclone Idai. For some respondents, the cyclone arrived in the middle of a multi-year birth spacing plan and had no impact on fertility intentions. For others, the hardships and uncertainty triggered by the cyclone motivated narratives of birth postponement. We explore these variations in why parents want to delay additional childbearing and discuss the implications of our findings for discerning trends and informing theories of fertility decline in natural-resource contingent contexts.

Birth postponement and contextual uncertainty

Recent theories of fertility intentions foregrounding contextual uncertainty emphasize the importance of imagined futures in guiding present reproductive plans (Vignoli, Guetto, et al. 2020; Vignoli, Tocchioni, and Mattei 2020). Perceptions of future uncertainty will vary from person to person, depending on current circumstances, past experiences with instability, and the set of schemas invoked to understand and make sense of social action. These ‘shadows of the future’ influence actions and intentions in the present, as individuals imagine and navigate possible trajectories (Bernardi, Huinink, and Settersten 2019). When envisioning future possibilities, individuals not only face uncertainty about the outcomes of current actions but also uncertainty about the contexts in which future actions will occur. In this study, we are not only interested in the ways that parents’ fertility plans directly respond to experiences with Cyclone Idai but also how these narratives reflect perceptions of future contextual uncertainty, specifically uncertainty invoked by environmental forces like extreme weather events.

Theories of fertility postponement in sub-Saharan Africa center the uncertainty of everyday life as a motivation for couples to delay their next birth. Birth postponement is distinct from birth spacing (Timæus and Moultrie 2008). Birth postponement is grounded in an evaluation of present circumstances and whether now is the right time to have another child. Current material difficulties and perceived insecurity lead couples to decide they do not want additional children now; the decision to have additional children can be revisited at some future time as conditions and priorities change (Timæus and Moultrie 2020). In contrast, decisions about birth spacing are independent of these contextual factors. Birth spacing is based on the age of the youngest surviving child and is related to controlling the length of the birth interval itself.

In the imagined futures framework, fertility preferences are recognized as uncertain and flexible (Morgan 1981; Trinitapoli and Yeatman 2018), shifting and taking shape in response to evolving life circumstances. Empirical studies of fertility intentions in sub-Saharan Africa suggest that even when respondents think the present is a bad time to have a child, the right moment is often ambiguously framed (Agadjanian 2005). Even when economic and marital instability are perceived as temporary, whether the next child will directly follow an improvement in circumstances is not certain (Garver 2018). Persistent uncertainty may interfere with the ability to imagine the future or make longer-term reproductive plans, requiring the continual re-evaluation of desires and circumstances (Johnson-Hanks 2005). Such ambiguities suggest long birth intervals emerge inadvertently, as couples repeatedly defer their next birth to an imagined better future. Decisions to postpone additional childbearing may also result from attempts to minimize the impact of future shocks and contextual uncertainty. If parents anticipate that closely spaced births may constrain the ability to achieve other household or personal goals, then longer birth intervals may be an intentional strategy to minimize the consequences of potential future instability.

By examining parents’ fertility intentions in the period following Cyclone Idai, we use the imagined futures framework (Vignoli, Guetto, et al. 2020) to integrate the schemas parents invoke to understand and make sense of environmental forces (i.e., extreme weather events) as they envision their family’s future composition and wellbeing. Our approach is distinct from recent research linking fertility intentions and short-run climate anomalies or subsequent food security. These studies analyze isolated disruptions in near-term reproductive behaviors and outcomes and find a mixed pattern of association that varies across contexts (e.g., Brooks et al. 2023; Eissler, Thiede, and Strube 2019; Gray and Thiede 2024; Somefun, Banougnin, and Smith-Greenaway 2024).² More importantly, none of these studies operationalize uncertainty for fertility intentions nor do they examine how the experience of a short-run anomaly is understood by parents in terms of family planning. A small number of qualitative studies have drawn attention to heightened desires to delay or stop childbearing during periods of climate-induced food insecurity (e.g., Grace et al 2017; Rosen et al 2021), but these studies also underplay how environmental disruptions generate uncertainty.

In contrast, we foreground the role of contextual uncertainty in our respondents’ reproductive plans, with a particular interest in the time horizons over which parents jointly evaluate their environmental and reproductive futures. Within the imagined futures framework, the immediate context of environmental instability generates an anticipated future characterized by uncertainty, which in turn informs reproductive actions and intentions in the present. By centering uncertainty in both fertility intentions and perceived weather and economic futures, we recognize the important role that uncertainty plays in the lived experience of our respondents (e.g., Trinitapoli 2023).

Our study is situated in communities dependent on natural-resource contingent livelihoods, primarily influenced by rainfed agriculture. Consequently, environmental fluctuations like extreme weather events disrupt household income and increase household costs. For instance, heavy rains that wash away crops and flood fields decrease a household’s anticipated food supply for subsistence and market and, at the same time, introduce the unanticipated costs of replanting and purchasing food for consumption. Additionally, significant rainfall paired with high winds can damage or destroy housing, which imposes unexpected costs of rebuilding. Cyclone Idai generated a conjuncture that forced families to evaluate their fertility plans (e.g., Johnson-Hanks et al. 2011): households confronted income disruptions and unanticipated costs as they balanced their desires to grow their family against their anticipated capacity to maintain living standards and support another child.³ Within this context, we examine how experiences of economic uncertainty induced by an extreme weather event inform visions of the future and how these visions shape fertility intentions and actions, in particular the time horizons invoked as parents develop plans for future births.

Study Context

The inter-annual variation in rainfall in southern Africa has increased since the 1960s— droughts have become more intense and widespread, while rainy periods have become shorter and with increased rainfall (Fauchereau et al. 2003; Shongwe et al. 2009; Williams, Kniveton, and Layberry 2010). Countries in the region are particularly vulnerable to climate variability, given the paucity of resources and infrastructural constraints that exist even in the absence of extreme weather events (World Meteorological Organization 2020). Malawi has one of the most erratic rainfall patterns in Africa, with high spatial variability in monthly and annual rainfall across districts, by topography and location (Kumbuyo et al. 2014; Ngongondo et al. 2011). The number and severity of climate-related disasters has increased in recent decades, including 16 major flooding events between 2010 and 2022 (World Bank 2022). These challenges are exacerbated by widespread deforestation related to agricultural expansion in the colonial period and continued reliance on fuelwood or charcoal for domestic fuel production (Bone et al. 2023; Davies et al. 2010).

In this study, we focus on fertility intentions in the months following Cyclone Idai, one of the most destructive tropical cyclones to impact southern Africa to date (Devi 2019). Cyclone Idai began as a tropical depression off the coast of Mozambique and first made landfall on March 4, 2019. The storm moved inland and circulated over Malawi before moving back out to sea on March 9. Cyclone Idai then re-energized and made secondary landfall in Mozambique on March 15–19. An estimated 975,000 people in Malawi were directly affected by the storm, including 60 deaths, almost 87,000 displaced people, and over 288,000 collapsed or partially collapsed homes (Government of Malawi 2019). Areas with limited resilience to extreme weather were hardest hit, including smallholder farmers reliant on rain-fed agriculture (FEWS NET 2020; International Food Policy Research Institute 2019). Our primary study sites of Balaka and Machinga districts experienced substantial differences in the total amount of rainfall during Cyclone Idai (Figure 1). Although damage in Balaka was primarily due to strong winds, areas of Machinga received greater total rainfall during the storm (176 mm) relative to Balaka (133 mm). Despite substantial crop loss, Machinga was largely spared the deaths and displacement experienced by communities in southern Malawi.

Figure 1.

Map of total rainfall during Cyclone Idai highlighting study sites, May 4–9, 2019, Malawi (Data Source: CHIRPS)

Over the past 30 years, Malawi has also experienced substantial fertility decline, from 6.7 births per women in the early 1990s to 4.1 by 2020 (National Statistical Office/Malawi and Macro International 1994; National Statistical Office 2021). Median birth intervals in Machinga (35.6 months) were slightly shorter than in Balaka (40.1 months) and Malawi as a whole (41 months) in 2015–16 (National Statistical Office (NSO) [Malawi] and ICF 2017); these values are comparable to those found in other African countries at the same level of fertility (Casterline and Odden 2016). The prevalence of birth intervals of 60 months or longer almost doubled from 10.7 in 1992 to 20.4 percent in 2015–16 (National Statistical Office (NSO) [Malawi] and ICF 2017), a trend interpreted in previous studies as evidence of widespread and increasing birth postponement (Timæus and Moultrie 2020).

Data and Methods

This study draws on 65 in-depth interviews collected in rural Malawi in July 2019.⁴ Respondents were purposively selected from past participants of the Malawi Schooling and Adolescent Study (MSAS), a longitudinal survey that followed a cohort of youth from 2007 to 2013 in Balaka and Machinga districts (Hewett and Mensch 2015). The present sample consists of 35 women and 30 men ages 26–29 years old at the time of the interviews.

We used a purposive quota sampling procedure to identify respondents for the in-depth interviews. We selected MSAS clusters based on two criteria. First, the cluster must have had at least 10 female respondents and 10 male respondents in residence at the time of the 2013 survey round. Second, we focused on clusters identified by local key informants as having experienced extremely high rainfall in March 2019 when Cyclone Idai hit the region. Our final sample contained five clusters in southern and western Machinga where the rainfall was highest and two clusters in southwestern Balaka along the border with Machinga. Relative to the main MSAS sample, clusters chosen for this study were more likely to be rural than peri-urban. We contacted MSAS respondents until at least three and no more than six respondents of each sex had been interviewed within each selected cluster. Due to high levels of out-migration, the interview team often contacted all male respondents who were living in a cluster in 2013 to complete three interviews. In contrast, the majority of female respondents were still living in the same area as they had lived in 2013.

All interviews were conducted in Chichewa, the national language, by experienced local interviewers of the same sex as the respondent. The interview guide (Appendix 1) included questions about the respondent’s background, recent experiences with extreme rainfall, livelihoods, fertility plans, and fertility outcomes, and interviewers were encouraged to probe beyond these prompts. The authors of this paper met with the interviewers daily to discuss interview content and revised the interview guide in response to emerging themes as the study progressed.

Recordings of the interviews were transcribed and translated into English by local transcribers immediately following the interviews. The authors read all transcripts immediately after they were translated to clarify the content with interviewers and transcribers. The interviews were analyzed using a flexible coding strategy (Deterding and Waters 2018). First, interviews were index coded by a research assistant in NVivo with categories corresponding to the major themes of the interview guide. Second, the authors re-read the coded units focused on the current open birth interval and inductively generated detailed analytic codes based on themes that emerged from the interview data. Our analytic coding process focused on the manifest reasons given to explain desired birth intervals, the certainty of childbearing intentions, and time horizons of fertility plans.

Results

Four months after the extreme rainfall associated with Cyclone Idai, our interviews reveal continued evidence of the physical devastation resulting from the storm. Almost 40 percent of respondents reported that at least part of their home collapsed during the heavy rains and winds associated with the cyclone (Figure 2) and all respondents who engaged in subsistence agriculture reported extensive crop loss (Table 1). The average female respondent had 2.5 children and the average male respondent had 1.4 children, including two women and six men who were childless. A similar share of women and men attended secondary school.⁵ Twenty-seven women and 24 men were married at the time of the interview, seven women and two men were separated or divorced, and one woman and four men had never married. All never married respondents were nulliparous and unable to describe intended birth intervals beyond describing ideals. Therefore, this analysis will focus exclusively on the fertility intentions of ever-married respondents.

Figure 2.

Example of a home collapsed during a cyclone in Machinga district, Malawi (Source: authors’ photo)

Table 1.

Characteristics of respondents, aged 26–29 years old, Malawi, 2019

	Women (N=35)	Men (N=30)
Currently married (N)	27	24
Currently using contraception, self or spouse (N)	24	22
Injectable or pill (N)	16	15
Implant or IUD (N)	7	5
Condoms (N)	0	2
Not currently using contraception, self or spouse (N)	3	1
Currently pregnant, self or spouse (N)	1	1
Numeric time until next birth (N)	23	16
Desired time until next birth, years (mean)	4.30	3.81
Non-numeric time until next birth (N)	3	7
Currently divorced or separated (N)	7	2
Never married (N)	1	4
Current number of children (mean)	2.54	1.36
Attended secondary school (N)	13	11
Home partial or full collapse (N)	16	10
Crop loss due to rains (N)	30	29

Ever-married men were more likely than ever-married women to provide a non-numeric response to questions about how long they would like to wait until having another child. Of respondents who provided a numeric response, the average married woman wanted to wait 4.3 years and the average married man wanted to wait 3.8 years until having another child. Divorced and separated respondents wanted to wait 5 years between births. Given the youngest living child for most respondents was at least two years old, enacting these fertility intentions would generate birth intervals longer than 5 years (National Statistical Office (NSO) [Malawi] and ICF 2017).

In the period following Cyclone Idai, three motivational narratives emerged. First, nearly half of women and one-third of men described their fertility intentions as undisrupted by the cyclone. A second group of respondents were certain that they wanted another child but planned to postpone their next birth because of uncertain conditions. These postponement intentions were grounded in economic uncertainties created by Cyclone Idai. Finally, the third narrative focused on uncertain fertility intentions. These respondents were not only postponing their next birth, but also the decision about whether to have additional children given uncertainties about the future. These narratives reflect mutually exclusive categories applied to current open birth intervals.

Undisrupted Fertility Intentions

Two narratives of undisrupted fertility intentions emerged from the data. The first group typically had a young child and pre-existing plans to wait until that child was five or six years old before having another; we categorize this narrative as extended birth spacing. The second group included six respondents who were eager to have another child in the next year.

Extended Birth Spacing

We categorized respondents as having an extended birth spacing narrative when their fertility plans specifically used the age of their youngest child as the most important determinant of birth intervals and were independent of current household circumstances. These respondents emphasized the importance of children’s self-sufficiency and health, such as wanting to wait until “after the child I have clocks six years” or that “my child should first grow up and start primary school” before conceiving the next child.

Jonathan⁶, a father whose two children were born five years apart, stated he waited “so that the first one should grow healthy…[and] be able to take care of the other”. Respondents with extended birth spacing narratives planned to have birth intervals of five or more years and their fertility intentions were unchanged or reinforced by the hardships created by Cyclone Idai. One mother of three children (a seven year-old daughter and 1.5 year-old twins) reflected:

When you don’t have anything children are the ones who suffer, it’s not good. I think I have to wait before I have another child…I want to take care of my children so that they can grow well, that they will be able to know what they are doing. At the end, I can have another child. (Ruth, three children)

Ruth lost most of her crops and was in the process of rebuilding her house after it collapsed during Cyclone Idai. Her comments emphasized the impact of current financial hardships on her children, expressing concern elsewhere in the interview about whether there would be sufficient food in the coming year. Consistent with other examples of extended spacing, she stated elsewhere in the interview that she waited until her first child began primary school before having a second pregnancy. Now she evokes a similar logic of waiting to have a fourth child so the twins will have time to “grow well”. Although her twins are too young for her to contemplate another birth, concerns for their health and well-being in the aftermath of the cyclone reinforce Ruth’s plans for extended spacing.

Other respondents emphasized maternal health concerns in their extended birth spacing narratives. According to Promise,

I would like four years to pass before having another one...I want my body to get back in shape. Bearing frequently can affect the health of a woman, since a woman loses a lot of things from her body (Promise, two children)

Promise’s rice and maize crops were severely damaged by the cyclone and she reflected that it would be a bad decision to have another child when the family was struggling with food insecurity. However, her reproductive plan was unaffected by these difficulties; her youngest child was less than two years old and she received a contraceptive implant the year before the cyclone that would enable her birth spacing plan.

Iness, a mother of two children, also invoked family planning as means of giving a child “space to grow”, but her reproductive plan was also dominated by concerns about her own health. Iness’s last pregnancy ended in a stillbirth and medical instructions not to engage in heavy labor. She currently has an intrauterine contraceptive device because of continued complications from her last delivery. Her husband left for South Africa a year before the interview and she supported her family by selling donuts and samosa at the local market. Most of the crops she grew for household consumption were lost in the cyclone, but her market stand earned enough money to buy food. The experiences of Ruth, Iness, and Promise were typical of respondents with an extended birth spacing narrative, such that they were not interested in having a child at the time of the cyclone and continued to maintain previously established reproductive plans.

Near-future Intended Childbearing

Five women and one man planned to have another child within the next year. Ester had already waited eight years since her last birth to have another child. She had recently returned from South Africa with her husband and used part of her savings to build a maize mill. Although they lost some of their crops during Cyclone Idai, income from the mill meant their household was financially stable and Ester reported she was trying to become pregnant. John was the father of a 1.5-year old child and was also financially unimpacted by Idai. He wanted his children to be born 3 to 4 years apart and planned to conceive a child with his wife within the next 1 to 2 years.

Judith was the only respondent who explicitly prioritized her desire for another child over the hardships caused by Idai. Her youngest child was four years old and she had recently remarried. According to Judith,

[The cyclone] didn’t affect my feelings, I just accepted it. I don’t care if we struggle as long as my children are growing…I stayed a long time [without a child] (Judith, 3 children)

Finally, two divorced women hoped to have a child as soon as they remarried, and one recently married woman planned to have a child within the next year.

Postponement and Uncertain Conditions

The second group of respondents were certain that they wanted another child, but their fertility intentions were not anchored on the age of their youngest child. The health and well-being of existing children and mothers remained central to their narratives, but respondents reflected openly about the contingency of their reproductive plans on future events, in particular the future economic well-being of their household given a changing environmental context (i.e., extreme weather events).

Indefinite Postponement

Friday, the father of a three-year old child, described how perceptions of economic uncertainty shaped his fertility intentions. As one of the few respondents not to provide a numeric waiting time until their next child, his rationale for postponing was grounded in the current state of uncertainty generated by the recent cyclone:

I cannot say that I want to have another child in this year or next year. I will wait so that I see what will be the future. I cannot choose to have a child now because I will not be able to take care of it because of the problems that have befallen us…I will wait up until I see that I am potentially able to take care of the child (Friday, one child)

What is striking about this quotation is the ambiguity of Friday’s future time horizon. His household was severely affected by Cyclone Idai, losing their crops and house to the storm. Given these hardships, he does not want another child in the near future and is unwilling to specify when he thinks his family’s situation will be secure enough to consider having additional children.

Harry also provided a non-numeric desired waiting time and elaborated on these contingencies:

R: I managed to have the [two] children but my intention was to have three children, so because of the circumstances [related to Cyclone Idai], I saw that it would not have been okay to have three children, so for now we agreed to stop at having two only. I will wait...

I: How long do you think you are going to wait in order to have the third child?

R: We have to get prepared for farming so that I will have enough budget and later on I will start that other budget, the child’s budget. But when you see that things are not okay, you wait for some time because things are not working properly, but if all is well, it is very okay to have the third child. It depends. (Harry, two children)

In this passage, Harry reflects that even though he intends to financially prepare for a third child, after regaining financial security things may not turn out how he expects. This quotation exemplifies the “wait and see” perspective voiced by many respondents, where the decision to have another child can be postponed for as long as necessary until conditions improve.

Other respondents provided extremely long desired waiting times as a response to the cyclone:

I: Did the rains make you change your mind about having another child?

R: Yes, I changed my mind, that’s why I am saying that I will have another child after some years. This is 2019, I think the year to have another child is 2025 (Mussa, three children)

Elsewhere in the interview, Mussa says he wants four children. His plan to wait six years before having another child suggests not only the time immediately after the cyclone is a bad time for childbearing, but conditions might not be stable for some period. A six-year interval may as well be indefinite given the potential for future extreme weather events to push the “right time” farther into the future. This long interval implies Mussa’s plans for a fourth child may be more uncertain than definitive and he may be at risk of becoming a “permanent postponer” where the continual postponing of the next child ultimately shifts into a curtailment of childbearing.

Defined and Extended Postponement

For other respondents, the uncertainty of when they would economically recover from Cyclone Idai led them to state relatively long intended waiting periods until their next birth. Blessings indicated he would like to wait at least three more years to have another child, until his youngest child is five years old. Although only 2.5 years elapsed between his first and second child, the impact of Cyclone Idai changed his plans for the future:

R: During the time we had flooding, our seedlings were washed away, which demanded we search for other seeds. It was a time we were worried that things weren’t okay...We just decided to have a good number of children so that we could manage them and should not cause any more troubles to the family that’s already affected by poverty, to keep a good number of family members was the remedy we took so that we were able to manage the family (Blessings, two children)

Blessings reflects his family is already facing poverty and spontaneously notes the birth of additional children would only exacerbate their current hardships. In this case, Blessings suspected his family would not recover sufficiently to support additional children for at least three years. References to a “good number” of children here equates to low parity, demonstrating how families facing poverty were intentionally, if temporarily, choosing small family sizes.

Other respondents noted that the uncertainty of when they would recover from Cyclone Idai led to responses walking the line between postponement and extended spacing. When asked what he would do if there was a successive year of bad rainfall that threatened his family’s food security, James responded:

I can tell my wife to be taking contraceptives, maybe for 3 years or 5 years to see what the future will be like. I want this [current] child to reach 5 years, it is when I can have another child...so that the newly born baby should be well cared for (James, one child)

Although James uses the language of extended spacing, his stated logic is not based on the age of his youngest child and he raises the topic in response to a question about how his family would handle the threat of food insecurity. James proposes they should wait until his youngest child is at least five years old because that should give his household enough time to recover from Cyclone Idai and make sure there are sufficient resources to financially support a new child. This longer time horizon not only allows for recovery from a bad agricultural season but includes additional time to “wait and see” whether conditions stabilize and the family can afford another child. As with other respondents already cited, the well-being of current and future children is paramount to James’s reasoning; his fertility plan centers the new challenges to maintaining child health and nutrition created by Cyclone Idai. It is also notable that James advocates for the use of contraception to achieve these goals.

For other respondents, the persistence of economic misfortunes that preceded and were compounded by Cyclone Idai led to the continued postponement of a desired birth. Eleni had two children aged 10 and 8 years old and wanted four children in total. She indicates that prior to Cyclone Idai she had been ready to conceive another child. Following the cyclone, however, she reported:

Yes, I changed [my mind] that now it’s not the right time to get pregnant because there is famine. Now I am still waiting but next year I will have [another child]. A pregnant woman can’t work to find food for her children so she has to wait, maybe in the future I can have one. I want to be prepared so that I should be able to take care of all the children (Eleni, two children)

Eleni emphasizes the current famine conditions created when the storm destroyed most crops in the area and the particular problems that these conditions generate for maternal and child health.

Many respondents, female and male alike, suggest women should try to avoid pregnancy during the most labor-intensive times of the year; these interviews were collected in the middle of the cool, dry season which many respondents indicated was the ideal time to be pregnant because the temperatures were low and there was less agricultural work to be done. For Eleni, however, the loss of her crops meant she needed to seek out piece work—paid work, such as collecting water for brickmakers or hoeing a neighbor’s field—to secure food for her children. The intense labor required to make ends meet given cyclone-related hardships, combined with concern about the well-being of her current children, prompted Eleni to continue postponing her next pregnancy. It is unclear from her interview whether she had originally planned to wait until her youngest child was eight years old before having another child, but the cyclone clearly led to the postponement of a wanted birth.

Postponement and Uncertain Intentions

Our third fertility narrative contained respondents who were uncertain about whether to have additional children. These five parents were not only pushing their next birth into the future, but also the decision about having another child at all. Uncertain conditions following Cyclone Idai contributed to uncertainty about future childbearing.

Whereas Chikondi used an extended birth spacing narrative to describe why seven years elapsed between his first and second children, his current fertility plan focused on his uncertain intentions:

I have two children…the decision to have another child came last year. What affects that is the financial status. When you make MWK500 [USD 0.66 at the time of interview], you use MWK300 and there are MWK200 that remains daily, so you think that even if you have another child, you will be able to take care of him because of the excess money you are having. So that’s when I started the decision to have another child. Right now, I have not yet decided whether to have three children or not because I feel like I have a big responsibility of taking care of these two children (Chikondi, two children)

Many fathers such as Chikondi discussed the financial margins on which their household existed; even when they were able to set aside a small amount of savings each month, they reflected it would not be enough to financially support a new child, sometimes going so far as to say under those circumstances the baby would “steal” from the older children and compromise the well-being of current children. Elsewhere in the interview, Chikondi said that “you cannot know about the future but you just make yourself ready for anything that can come.” His youngest child was only four months old and he had relatively stable income from his motorcycle taxi, but his narrative emphasized how the unpredictability of rainfall and food prices contributed to his uncertainty about childbearing.

Other respondents gave long desired birth intervals to leave open the possibility of future children. Alinafe had four children and stated:

R: At first I tried to have children with a gap of 2 or 3 years [between them] but this time around I have decided to wait for some years. And now I haven’t yet made the decision about when to have another child.

I: Now you have not yet made the decision to have another child because things are not okay?

R: Yes (Alinafe, four children)

Alinafe’s youngest child was five years old. The circumstances brought on by Cyclone Idai mean it is not only a bad time to have another child, but a bad time to make a decision about having another child. She goes on to say she intends to wait seven or eight years before making a decision about additional children. Alinafe planned to replace her current contraceptive implant with another when it expired the following year, suggesting a commitment to putting off the decision to the future.

Other respondents with extremely long desired waiting periods speak more directly to a pessimism that conditions are unlikely to improve. Thokozani, a mother of three children, wants a fourth child but is currently using a contraceptive implant to postpone additional childbearing into the distant future:

I: How long would you like to wait?

R: After the [youngest] child clocks 10 years…[she is now] two years old

I: You still have 8 years ahead? Why would you like to wait for such a long period?

R: Because we have insufficient food and lack support (Thokozani, three children)

Although Thokozani’s house was not damaged by Cyclone Idai, her crops were washed away. This quote, however, points to the poverty her family faced before the compounding hardships caused by the recent cyclone. Furthermore, the twin of her youngest child died around the time of Cyclone Idai from cerebral malaria. By pushing additional childbearing into the distant future, Thokozani leaves the possibility of additional childbearing open even as she voices a firm commitment to postponing a fourth child; elsewhere in the interview, she describes her plan to get another contraceptive implant inserted when her current implant expires.

Discussion

Few of our respondents were untouched by the economic hardships that followed Cyclone Idai, even though none were displaced or lost family members because of the storm. The cyclone itself came a year after their communities experienced a drought that also strained family incomes. In the wake of this extreme weather event, our respondents perceived ongoing uncertainty; one year of good rains would not be sufficient to ensure economic recovery and the following year’s conditions were unclear. Respondents who planned to postpone childbearing because they faced uncertain conditions proposed waiting multiple years—ranging from three or more years to indefinite periods—to be sure conditions were stable enough to not endanger the health of current and future children. These findings, situated in a period immediately following an extreme weather event, suggest our respondents were considering the cumulative experience of climate uncertainty over a longer time horizon when building their reproductive plans of action. These narratives inform our understanding of postponed fertility as an adaptation to climate change.

Economic insecurity experienced immediately following a destructive climate-related weather event such as Cyclone Idai shapes both the experience of the present (‘the shadow of the past’) and perceptions of how the future will unfold (‘the shadow of the future’) (Bernardi et al 2018). Many of our respondents describe imagined futures punctuated by uncertainty about future climate-related weather conditions, including the inability to trust one year of good rain as more than a fluke or the need to wait multiple years to be sure household conditions had stabilized. Some respondents described a ‘wait and see’ approach, where pressing desires to have more children may eventually outweigh the financial pressure of additional children in the face of ongoing economic instability. Far more respondents described plans of action using long birth intervals as a strategy to protect the household’s living standards against fluctuating economic conditions directly linked to extreme weather events. None of our respondents described a need to increase or accelerate childbearing in response to uncertainty about environmental conditions, nor did they invoke a child’s future labor as part of a risk diversification strategy, in contrast to the expectations of some economic theories of fertility (e.g., Cain 1981; Friedman et al. 1994; Dasgupta 2000; de Sherbinin et al. 2008). To the extent that children’s labor was mentioned, it was almost exclusively within the context of signaling a child’s self-sufficiency.

One limitation to our sampling strategy is that we are unable to compare these narratives to those that circulate in communities that were relatively unaffected by Cyclone Idai. We sought to understand how fertility intentions may be part of a strategy for adapting in place to extreme weather events. This approach allowed us to observe how the cyclone was integrated into the reproductive plans of parents in communities that continued to experience its consequences. The immediate context of the cyclone, however, may have led respondents to underplay other environmental factors, such as deforestation or ongoing food insecurity related to droughts in the prior year. These topics were mentioned in relation to constrained local livelihoods, but our respondents did not spontaneously connect these themes to their fertility intentions, and we did not probe around these topics. We hypothesize that droughts might increase uncertainty about childbearing, with respect to both the timing of the next birth and desires for additional children. In contrast, we would not expect deforestation or other forms of natural resource degradation to be associated with a postponement narrative because these conditions are unrelenting; it only makes sense to postpone childbearing if one believes that future conditions might be better.

In a region characterized by high levels of unintended childbearing (Palamuleni and Adebowale 2014; Bornstein, Gemmill and Norris 2023; Yeatman and Smith-Greenaway 2018), it remains to be seen whether these respondents will be able to enact the reproductive plans described in these narratives. The majority of unintended childbearing in Malawi takes the form of mistimed births, suggesting some respondents will fall short of the long birth intervals to which they aspire. Nonetheless, there is reason to believe many of our respondents will achieve these goals: at the time of the interviews, more than 90% of married respondents reported using modern contraception and all but one of the respondents not using contraception hoped to have a child in the near future. Although many respondents were using long-acting reversible contraceptives, we did not ask respondents about their family planning history and, consequently, we cannot know whether the cyclone affected the widespread commitment to contracept. In the 2015–16 Demographic and Health Survey, 54% of 25–29 year old married women in Machinga district and 62% in Balaka district were currently using modern contraception (National Statistical Office (NSO) [Malawi] and ICF 2017); the level of contraceptive use declared by study respondents is substantially higher than that measured in the same regions four years earlier, but not implausible for a population highly motivated to avoid childbearing in the aftermath of an extreme weather event. Recent evidence from Malawi also suggests Machinga district has one of the more robust family planning programs in the country (Peterson et al. 2022). Finally, median birth intervals in Malawi rapidly increased in the years preceding Cyclone Idai (Grant et al. 2023) to levels consistent with plans described by our respondents. Our respondents’ fertility intentions will evolve in response to subsequent life events and contextual circumstances, but these narratives lead us to hypothesize that women in communities exposed to an extreme weather event like Cyclone Idai will achieve longer birth intervals than women in unaffected communities.

Many respondents had precarious livelihoods prior to Cyclone Idai. Thus, the cyclone likely exacerbated earlier conditions and increased respondents’ uncertainty about the future. Prior work by Trinitapoli and Yeatman (2018) points to the flexibility of fertility intentions, both in timing and number of children. Our study suggests more research is needed on the intensity of parents’ attachment to their childbearing intentions with explicit attention to the environmental contexts in which parents make fertility decisions. Although our study cannot fully capture how plans for extended birth intervals may have changed in response to shifting conditions sparked by a recent climate-related weather event, we show the texture of how parents in their peak reproductive ages approach these decisions as they contend with the consequences of environmental disruption. Our results are generalizable to other communities with natural-resource contingent livelihoods, where environmental uncertainty shapes economic uncertainty and serves as an important frame for guiding fertility intentions.

Four years after Cyclone Idai moved through our field site, Cyclone Freddy impacted the same area. As extreme weather events become more frequent, how are they incorporated into local environmental schemas? Cyclone Idai was experienced as a destabilizing event that rendered the future less predictable. Plans to postpone additional childbearing—or decisions about childbearing—served as an adaptation to this heightened uncertainty. Repeated exposure to extreme weather events might be schematized as a type of reliable unpredictability (Cunfer 2005; Polsky and Easterling 2001), normalizing longer birth intervals for everyone and reinforcing continued fertility decline in the region. Alternatively, repeated exposure might lead contexts to be perceived as degrading (e.g., Biddlecom, Axinn, and Barber 2005; Carr 2005; Ghimire and Mohai 2005; Rosero-Bixby and Palloni 1998; Sasson and Weinreb 2017), which might shorten birth intervals and preserve higher fertility. The risk of unrealized fertility goals or mistimed births may also grow if families continually judge conditions as unsuitable for additional childbearing. Future research on how climate-related weather events are incorporated into narratives of the future will be critical for understanding decisions about fertility.