In the past, Kigali has frequently experienced heavy rain events. These have often led to flooding, which also affected businesses. In the face of climate change, such events can become more frequent and can threaten economic development. To determine if more action is needed to protect businesses from flooding, we assessed how many businesses have suffered damages from floods in certain years in a certain area of Kigali. We also gathered information on how businesses were affected, how they are preparing for flooding and what support they are seeking. We developed and piloted a survey, a standardised questionnaire for gathering information on the relevance of flooding for businesses. The survey was then conducted among 350 businesses in Kigali asking business owners about their experiences with flooding in recent years. Eighty-one per cent of businesses have been affected by floods in 2013 and 2014. The annual damage costs resemble 22% of the total net profit of the businesses in the area. The most common damages were damages to goods that were to be sold and damages to buildings. The extent of past flood damages warrants action on flood risk management, both by businesses and citizens, as well as by city officials. Suitable actions range from increasing awareness about suitable protection measures to upgrading the sewage system.
Worldwide, extreme precipitation events, which can lead to landslides and floods, are one of the most severe hazards to human settlements and have caused considerable economic and social costs. In the light of progressing climate change, the likelihood of such events will further increase (IPCC
The economy of Rwanda is vulnerable to the impacts of climate changes whose additional net economic costs are estimated to be almost 1% of gross domestic product (GDP) each year by 2030 (SEI
Thus, analyses of the extent of flood damages for businesses are of importance as they can provide policy makers and entrepreneurs with information which can be used as a basis for reducing economic loss from flood damages, for example, via improved flood risk management in affected locations (Merz et al.
Against this background, the article aims to quantify business-related damage costs incurred from flooding and the determinants of impacts on businesses in the catchment area of the Nyabugogo River in Kigali in recent years. These investigations can benefit science and policy-making in different ways: they can add to methodology development for assessing damages costs for businesses, provide policy makers with a baseline of damage costs against which flood risk management options can be selected and inform entrepreneurs about the most effective measures for dealing with floods in the area.
In a developing country context, SMEs can be credited with the potential to increase economic growth, alleviate poverty, foster development and decrease income inequalities (USAID
There are a range of characteristics of SMEs that point to higher vulnerability
However, it can also be claimed that the characteristics of SMEs, like their flexibility and limited capital needs for operation and recovery, entail an inherent resilience (Ballesteros & Domingo
[
Other concepts go beyond this bounce-back concept and look at what capacities are needed for businesses to return stronger than before. Most reviewed literature though indicates that this inherent resilience does not outweigh the vulnerability factors of SMEs in a developing country context.
Therefore, there is a need for SMEs to prepare for disasters through identifying their vulnerabilities and enhancing their resilience. This process ‘requires partnerships and cooperation among the firms, public and other private organizations’ (Ballesteros & Domingo
In the aftermath of a flood event, it is important to minimise the recovery time of SMEs to avoid further marginalisation of affected people and sustain income opportunities. Asgary, Anjum and Azimi (
Extreme weather events, such as heavy rain events triggering floods, can lead to a variety of damages and resulting costs (
Typology of damage costs.
Variable | Tangible | Intangible |
---|---|---|
Direct damage | Physical damage to assets such as buildings, infrastructure, machinery or inventories | Loss of life, health effects or loss of ecosystem services |
Indirect damage | Disruption of production processes at supply firms or traffic disruptions | Inconvenience of post-flood recovery, reputational damages and increased vulnerability to future hazards |
The most immediate physical effects of extreme events on assets, human beings or the environment are characterised as direct costs. The results of damages to buildings, infrastructure, machinery or inventories are among the most visible impacts of extreme weather events. Indirect damage costs are costs that arise from reductions in demand if customers are directly affected by an extreme weather event or from the interruption of production processes because of a lack of production inputs provided by supply firms that are either directly affected by an extreme weather event or that cannot supply production inputs to the business under consideration because of traffic disruptions caused by damages to transport infrastructure.
Direct damage costs are often measured using susceptibility functions (Meyer et al.
Overall, assessments of direct and indirect costs of damages from extreme weather events can be conducted on different geographical as well as temporal scales. The methodological approach used for cost evaluation is determined by the selected spatial level. While macro approaches of damage evaluation on a national or even international level (Bizimana et al.
Data to be collected for the determination of the value of assets at risk at a micro level require a categorisation of the building type and usage. This can further be divided by subcategories such as the business sectors, age of the building, inventory of machinery and other equipment, type of materials or goods, area and use of the ground floor and/or basement, or elevation based on threshold for floodwater intrusion, as well as height above ground level (Messner et al.
Unlike direct damage costs, the assessment of business disruption costs is subject to a less straightforward approach. Business disruption costs have to be assessed by either using sector-specific reference values, by comparing production outputs for years when the business was affected by an extreme weather event with a year without such an event or by calculating production losses using a fixed share of direct damages (Mechler
In conclusion, it can be said that there is a wide range of theoretical approaches and tested methodologies for estimating damage costs of extreme weather events, such as floods triggered by heavy rains. However, it is apparent that specific circumstances, objectives, spatial scales and time frames shape the selection process of methodologies. Data availability is among the key determinants for selecting a suitable methodology.
Against this theoretical background and the reviewed empirical assessments, it is apparent that conducting a survey on a single unit of analysis (businesses) is an adequate methodology for the purposes outlined above that allows gathering data on damage cost estimates and insights into entrepreneurs’ behaviour in the Rwandan context. There are three limitations of this approach: (1) Few business owners possess written records of flood damages and thus had to rely on their memory and estimations of costs to reply to the questions. (2) Business owners do not have much time in their daily work to respond to questionnaires; thus, some answers might have been rushed. (3) The exact locations of the responding businesses were not recorded as this would have made it impossible to ensure anonymity to the respondents. Anonymity, however, was important because sensitive information like revenues were gathered. To address the first two limitations, the survey only focuses on damages from the years 2013 and 2014.
The field research was carried out in the area around the Nyabugogo River within the catchment of Nyabugogo–Gatsata–Kimisagara–Giti Kinyoni, City of Kigali, Rwanda (
Map of the surveyed area. Map credits: ©OpenStreetMap-contributors (SRTM); Map design: ©OpenTopoMap (CC-BY-SA), changes made: Buildings added, individuals buildings are aggregated, area encircled in red shows surveyed area.
The data collection was undertaken in a two-stage process: (1) transect walks around the study area to observe enterprise styles and construction features, such as used materials, as well as geographical features such as slopes and swamps (propensity to flooding for example); and (2) a quantitative survey, for which structured interviews with SMEs based on a standardised questionnaire were conducted. The structured interviews for the survey were conducted between 19 August 2015 and 07 September 2015.
A range of studies successfully applied comparable firm or household-level surveys to estimate flood damage costs and impacts (Khandlhela & May
For the estimated number of SMEs in the Kigali river catchment, the sample size needed, according to Slovin’s formula, should include at least 286 businesses to allow drawing representative conclusions (with a confidence level of 95%) for all businesses in the area (Guilford & Fruchter
All of the businesses in the sample have less than 100 workers, and thus can be classified as being SMEs (Ayyagari, Demirguc-Kunt & Maksimovic
The methods of survey research in the form of a structured interview of individuals were applied during the investigation as the method is versatile and efficient (Schutt
Furthermore, the questionnaire was carefully designed to avoid inaccuracies and measurement errors, keeping in mind that questions should be clearly phrased and neutral (Gray
When designing the interview schedule, refining and testing the questions included is a crucial step (Schutt
The questionnaire was then administered face to face by three trained interviewers in the preferred language of the respondents, who were asked to respond on behalf of the owner of the enterprises. Three call-backs were made before an enterprise was recorded as a non-response.
The damage cost estimation in this study focuses on a micro (local) level. The questionnaire was designed to gather baseline information on the socio-economic characteristics of the sampled enterprises, flood-related damages (both direct and indirect damages, and business interruptions) in 2013 and 2014 as well as information on flood protection measures the enterprises implemented, or not, to enhance their resilience against flooding. This information enabled an investigation of damage types on all three broadly defined stages of a value chain, namely procurement, production and sales (
Damage types and assets at risk included in the survey along the value chain.
The damage costs themselves are separated into direct and indirect damage costs. As it was expected that the interviewees would not be able to precisely reflect estimates on all costs for their enterprise triggered by floods in recent years, questions about damage costs always referred to the most severe flooding event in the two years between 2013 and 2014 that the respondents could recall. This method of self-reported costs enabled a rough estimate of the costs of high-impact events on businesses in the area for each year. The estimates of indirect damage costs are based on the number of days of business interruptions after the strongest (most severe) flood event, in 2013 and 2014, and the estimates by the interviewees of revenue loss per day. Further investigation of the data set, utilising the analytical software SPSS, was undertaken in order to receive insights into the relation of business type and size, damage experienced and disaster risk reduction measures implemented or interested in.
This article followed all ethical standards for research without direct contact with human or animal subjects.
The following section provides insights on the overall socio-economic characteristics of the sample of 353 SMEs surveyed in Nyabugogo, Kigali, Rwanda. In the sample, among the most dominant types of enterprises were car item retailers with 40.2% (142 SMEs), which were followed by retailers of daily use items with 35.1% (124). The other types of enterprises were far less in number. These were wholesalers of food (7.6%, 27 SMEs), manufacturers (6.8%, 24 SMEs), hotels and restaurants (4.2%, 15 SMEs), service providers (3.4%, 12 SMEs), and hardware stores (2.5%, 9 SMEs). This distribution reflects the general image of the area as a main marketplace, particularly for car parts, in that part of Kigali.
Among these businesses, only 2.5% (9) own the premises of the enterprise. The other 97.5% are renting the premises. Ownership of property of the enterprise can be a determining factor in the motivation of entrepreneurs to engage with disaster risk reduction measures. Compared with renters or leasers, property owners may feel that their assets are at higher risk, and thus they may be more concerned about vulnerability and disaster reduction (Han & Nigg
The size of the enterprises (in terms of number of employees) ranged from one employee to more than six employees, with the majority of enterprises having two employees (47%, 166 SMEs, including the owner), followed by one employee (31.4%, 111 SMEs), three to five employees (16.1%, 57 SMEs) and more than six employees (5.4%, 19 SMEs). Thus, the vast majority of enterprises surveyed (78.4%) operate with no more than two workers and could be classified as being small, or even micro enterprises.
All surveyed SMEs were asked to reveal their net profits. More than half of the sample (51.3%) earned less than 100 000.00 RWF (Rwandan franc) (about $140.00) per month. Another 41.4% earned a monthly profit between 100 000.00 RWF and 500 000.00 RWF. Only seven enterprises earned more than 500 000.00 RWF (around $700.00). Based on these self-declared net profits, the overall monthly profit for all businesses surveyed is estimated to be 53 425 000.00 RWF (around $75 000.00). The combined
The entrepreneurs in the Nyabugogo catchment area recognise flooding as the main hazard for their businesses: when asked which of the four disasters they consider most dangerous for their operations, 88.7% mentioned flooding, followed by fire in the building (9.3%), windstorms (1.1%) and landslides (0.8%). The recognition of this hazard seems to be in line with the experience of businesses regarding flooding in the area. Eighty-one per cent (284 businesses) were affected by flooding at least once between 2013 and September 2015. In the period between 2013 and 2014, the surveyed enterprises were most often affected (i.e. suffered any damage) by flooding in the year 2013 (
Frequency of enterprises being affected by flooding between 2013 and 2014.
Frequency | 2013 | 2014 |
---|---|---|
Zero times | 147 | 192 |
One to two times | 181 | 154 |
Three to four times | 25 | 7 |
These findings correlate with the answers given by the interviewees about the year in which they experienced the most severe flood impact (
Year of the most severe flood impact on the enterprises between 2013 and 2014.
Year | Frequency | Percentage |
---|---|---|
2013 | 189 | 53.5 |
2014 | 95 | 26.9 |
Total | 284 | 80.5 |
Missing | 69 | 19.5 |
Based on these most severe flood events experienced by 284 businesses of the sample, the direct and indirect damage costs for the enterprises were calculated (see below). Among the most common direct flood damages were damages to items that were supposed to be sold (experienced by 74% of affected businesses, multiple answers possible), damage to building or premises (e.g. the door or walls of the building were damaged; 36.8%), damage to equipment or machinery (e.g. cash register or tools; experienced by 18.8% of affected businesses) and damage to production inputs (e.g. raw materials like wood; 3.1%).
The estimated damage costs of these single most severe flood events in 2013 and 2014 for businesses in the area amount to 144 800 000.00 RWF (around $200 000.00) (
Estimation of direct (physical) damage costs based on the most severe event for each enterprise by year cross-analysed with estimated direct damage costs for those events (all amounts in Rwandan franc).
Direct damage in RWF | 2013 | Costs 2013 |
2014 | Costs 2014 |
Total cost |
---|---|---|---|---|---|
< 100 000.00 | 64 | 3 200 000.00 | 57 | 2 850 000.00 | 6 050 000.00 |
100 001.00–500 000.00 | 48 | 12 000 000.00 | 13 | 3 250 000.00 | 15 250 000.00 |
500 001.00–1 000 000.00 | 20 | 15 000 000.00 | 8 | 6 000 000.00 | 21 000 000.00 |
1 000 001.00–5 000 000.00 | 34 | 85 000 000.00 | 4 | 10 000 000.00 | 95 000 000.00 |
5 000 001.00–10 000 000.00 | 1 | 7 500 000.00 | 0 | - | 7 500 000.00 |
Not declared | 22 | - | 13 | - | - |
RWF, Rwandan franc.
, Estimated damage costs are based on the middle amount within each category (e.g. 750 000 for the category 500 001–1 000 000) and their related frequencies per year, whereas for the lowest category, an amount of 50 000 was used.
Additional to these tangible direct damages, anecdotal evidence shows that some people were injured during the flood events in the area. Furthermore, the floods negatively affected ecosystem services, such as the provision of local raw materials like wood.
Businesses were asked to estimate the profit losses through business closures after the flooding event that was most severe to them between 2013 and 2014 (
Estimation of indirect damage costs based on the most severe event for each enterprise by year cross-analysed with estimated loss of profits during business closures after flooding (all amounts in Rwandan franc).
Indirect damage in RWF | 2013 | Costs 2013 |
2014 | Costs 2014 |
Total cost |
---|---|---|---|---|---|
< 1000 | 17 | 8500.00 | 29 | 14 500.00 | 23 000.00 |
1001–10 000 | 42 | 225 000.00 | 27 | 135 000.00 | 36 000.00 |
10 001–50 000 | 60 | 1 500 000.00 | 17 | 425 000.00 | 1 925 000.00 |
50 001–100 000 | 25 | 1 875 000.00 | 15 | 1 125 000.00 | 3 000 000.00 |
100 001–500 000 | 37 | 9 250 000.00 | 1 | 250 000.00 | 9 500 000.00 |
500 001–1 000 000 | 1 | 750 000.00 | 0 | 0.00 | 750 000.00 |
> 1 000 000 | 2 | 3 000 000.00 | 0 | 0.00 | 3 000 000.00 |
RWF, Rwandan franc.
, Estimated damage costs are based on the middle amount within each category (e.g. 5000 for the category 1001–10 000) and their related frequencies per year, whereas for the lowest category, an amount of 500 was used and 1.5 million for the highest category.
The total estimated direct and indirect damage costs for the businesses surveyed add up to 139 308 500.00 RWF (around $194 000.00) in 2013 and 24 049 500.00 RWF (around $33 500.00) in 2014 – based on the most severe flooding event that each affected business experienced in these two years.
To interpret these figures, it is important to take into account that the estimated damage costs are solely based on the most severe flood events experienced by the businesses surveyed. Considering the frequency of businesses being affected by floods, the real damage costs are very likely to be significantly higher than these estimates. This leads to the conclusion that in the years 2013 and 2014, floods had seriously affected the economic welfare of businesses in the surveyed area, reducing profits and causing damages in a way that was threatening the bottom line of many businesses in a serious way: For 2013, the total damage costs resemble 22% of the total annual net profit of the interviewed businesses in the area. On average, each affected business in 2013 suffered direct and indirect flood damage costs of around 737 000.00 RWF (around $1030.00) which is more than the annual net profit of around 25% of the businesses in the area.
It is remarkable that the total damage costs in 2014 were 83% lower than in 2013. There are two likely explanations for this difference: after a flood on the main road from the city centre to Nyabugogo in September 2013, local authorities took action between October 2013 and April 2014 and unblocked important waterways in the area and reconstructed a major drainage channel connected to the Nyabugogo River, spending about 265 000 000.00 RWF (around $370 000.00; Kubwimana
The majority of businesses (71.3%) in the sample have implemented measures to reduce expected flood damages. The most common action was creating a flood barrier (160 SMEs), and moving important equipment and items in the store to higher grounds when flood water approached the premises (149 SMEs). Only 34 out of the 353 businesses (less than 10%) have an insurance cover that covers flood damages. One hundred and one SMEs did not implement any disaster reduction measures, while floods between 2013 and 2014 spared only 69 enterprises. This means that at least 32 enterprises affected by flooding did not prepare for or react to frequent flooding of their premises.
There were 101 businesses that did not undertake any measures for reducing flood damage costs for a variety of reasons: 18 of them mentioned that measures were too expensive, eight businesses said that they lacked the time to deal with these issues and only three SMEs reported that they lack information on how to protect their businesses. The main reasons for not taking action lie beyond lack of information or resources: 57 businesses (42.2%) said that they were not taking any actions because they do not expect any future flooding and 25 (18.5%) consider the impact as not being significant enough to prepare for. It should be noted that from the 57 enterprises that do not expect any future flooding, 31 businesses (54.4%) had previously been affected by a flood. When taking into account the above-mentioned evidence on the likely increase in extreme precipitation events, this resistance to taking action can be seen as problematic for the bottom line of these businesses and development in the surveyed area. There are two likely explanations for this resistance to taking action: respondents might not be aware that the likelihood and intensity of extreme weather events are projected to increase; or they might be overconfident, thinking that future floods will not affect the area where their business is based (Kind & Savelsberg
When asked about what the local or national government should do to reduce damage costs from flooding, 78.3% (278 businesses) responded that more information should be provided on how one can effectively protect one’s business from flood damages. Thus, there is a veritable demand for more information on how to properly deal with floods on an individual level. Among these 278 businesses, more than 50% have implemented flood protection measures. Hence it seems that there is a larger group of businesses that is very willing to take action and would like to have more information on how to make these actions more effective. On the other hand, there is the group of 77 businesses (21.7%) that did not reply that more information should be provided. Most businesses in this group have not implemented any flood protection measures and do not foresee any future flooding. This second, smaller group can be characterised as unworried about possibly increasing flood risks and uninterested in information on how to deal with floods. Again, it should be noted that a large share of this group has previously suffered damages from floods.
The most common request with respect to governmental action is the improvement of the drainage systems (mentioned by 85.1% of the surveyed 353 businesses). Making insurance products more affordable is requested by 43.1% businesses in the sample.
The topic of insurance also features highly among the measures that businesses are considering to implement in the future (mentioned by 60.6% of the businesses surveyed). The second most frequent action mentioned was relocating one’s business to a less flood-prone area (mentioned by 29.2% of the businesses surveyed). The later point is of particular interest, considering the frequent controversies about relocation of businesses in Kigali (see Barigye & Rutarindwa
With respect to the methodology, it can be concluded that the approach of asking businesses only about the damage costs of the most severe floods made it possible to arrive at a consistent data set that did not overstretch respondents’ memory and time available for responding to the survey. At the same time, this approach only allowed it to determine the absolute minimum of possible damages costs that businesses experienced from flooding in recent years. The actual annual costs will be significantly higher. However, the approach still allowed detecting relevant changes in damages’ costs between years.
The results of the survey showed that flooding is the most relevant natural hazard to businesses in the Nyabugogo area. In 2013, the annual damage costs for businesses amounted to more than 139 308 500.00 RWF (around $194 000.00, sum of direct and indirect damages) which poses a serious threat to economic development in the area as these damage costs resemble 22% of the total annual net profit of the interviewed businesses in the area. On average, each affected business in 2013 suffered direct and indirect flood damage costs of at least 737 000.00 RWF (around $1030.00), which is more than the annual net profit of around 25% of the businesses in the area. Given that extreme precipitations are likely to become more frequent and intense with the changing climate, action needs to be taken by public authorities and by businesses to ensure climate compatible growth in Nyabugogo and similar areas in Kigali.
Between 2013 and 2014, after demands from the national government, the local government took action and cleaned smaller drains and rehabilitated a large drainage channel in the area at a cost of around 265 000 000.00 RWF (around $370 000.00; Kubwimana
Regarding the provision of information, there are two different demands among businesses: The majority of enterprises would like to receive information about how to protect their businesses effectively, while a small group of businesses are unwilling to take action because they do not see flooding as a relevant risk and thus also do not see a need to receive information on flood protection. To reduce future damage costs from flooding it thus seems advisable for public authorities to provide more information on effective protection measures to a larger group of businesses. Regarding the smaller group that does not see a requirement for taking action, there seems to be a need for raising awareness on the actual relevance of flooding, resulting costs and the effect of climate change on the frequency and intensity of heavy precipitation events. Possible overconfidence among business owners with respect to their enterprises’ vulnerability should also be addressed in these awareness-raising activities if one wants to increase the level of preparedness among businesses. Providing additional financial resources for flood protection, however, do not seem to be required.
Only a minority of less than 10% among surveyed businesses have an insurance that covers flooding. As around 60% are thinking about obtaining insurance and many are requesting affordable insurance products from the government, it should be investigated by public authorities, business associations, insurance providers and researchers in how far new and affordable insurance schemes could benefit businesses in the area.
Furthermore, nearly one-third of the enterprises would consider relocating their business to less flood-prone locations. This high willingness of business owners should receive notice in development planning processes; however, it needs to be taken into consideration that for a relocation to take place in an economically viable way, business owners need affordable and attractive alternative locations for their enterprises.
This article was written as part of the research project ‘Building resilience in Rwanda through business collaboration’ that was kindly financed by the Climate & Development Knowledge Network (CDKN). The authors would like to thank the funder CDKN, as well as REMA, MIDIMAR, the City of Kigali and the businesses that provided the information used to produce this article.
The authors declare that they have no financial or personal relationships that may have inappropriately influenced them in writing this article.
A.T. was responsible for research design, data collection, data preparation, data analysis and drafting text. C.K. was responsible for research design, data analysis and drafting text. J.S.H. was responsible for data analysis and drafting text. R.M. was responsible for quality control and data preparation. A.R.B. was responsible for the research design.
The project was funded by Climate and Development Knowledge Network (CDKN).
Data sharing is not applicable to this article as no new data were created or analysed in this study
The views expressed in the article are the views of the authors and not an official position of the institution or funder.
There is no standardised definition of vulnerability. In this article, vulnerabilities are understood as ‘the characteristics and circumstances of a community, system or asset that make it susceptible to the damaging effects of a hazard’ (UNISDR 2007).
Using exchange rates from September 2015 all throughout this article.
This is an agency in the Department of Commerce that maps the oceans and conserves their living resources; predicts changes to the earth’s environment; provides weather reports and forecasts floods and hurricanes and other natural disasters related to weather.