1. Introduction
Risks and uncertainties are closely related to agricultural activities due to the uncertain economic and biophysical circumstances according to which farming operates [
1]. In agriculture, risk and uncertainty are inseparable. Risk is imperfect knowledge where the probability of possible outcomes is known, and uncertainty exists when the probability is unknown. Uncertainty can be defined as inadequate information whereas risk implies unfavorable consequences to the economy or some adversity or losses which have negative effects on an individual’s well-being [
2]. Farming is a risky business because forces which are beyond the farmer’s control affect the profitability of farming [
3]. Risk, uncertainty, and agricultural production are inextricably related to income (variable returns) and have consequences for decision making in a specific time period [
4]. Risk and uncertainty are typical features of agricultural production [
5] which can harm production and cause extensive losses [
6]. In farming, it is therefore important to be aware of risks, uncertainties, and the available options for mitigating these conditions [
7]. Examples of risks and uncertainties include variations in frequency and intensity of rainfall, temperature fluctuations, crop diseases, and more. The prevalence of these risks makes the agriculture sector more risk-prone than other industries [
8], and also affects other sectors of the economy [
9]. In agriculture, the five most significant risk sources are production risks, marketing risks, legal risks, financial risks, and human resource risks [
10,
11]. Several types of risk management (RM) tools are available to farmers. The socioeconomic characteristics, losses by natural calamities, and perceptions of risk and risk attitudes of the farmer are considered to be important factors in the making of RM decisions [
7,
12]. Moreover, the relationships among various RM tools are also important considerations, as farmers commonly adopt multiple RM tools across a given time span [
13]. The common risk-reducing strategies include crop diversification, off-farm and on-farm diversification, forward hedging, contract farming, crop insurance, and precautionary savings, each of which has been used by farmers in Bangladesh for managing the risks and uncertainties of agriculture.
Bangladesh, which is the focus of this study, is one of the countries that is most affected by natural calamities, which have previously caused massive damage to its agricultural sector. Bangladesh is an agro-based economy in which agriculture contributes 14.2% to overall GDP and employs 48% of the country’s total workforce, including almost 65% of the rural population [
14,
15]. Agriculture sector in Bangladesh has experienced serious challenges from various catastrophic events. Hence, an adaption of RM tools in agriculture for reducing the effects of catastrophic impacts is an effective and necessary measure to protect the livelihoods of farmers. Most Bangladeshi farmers usually adopt one or more RM tools to protect against losses. Contract farming and precautionary savings are the most popular RM tools in Bangladesh.
Contract farming involves coordinating the production and supply of agricultural and horticultural products between sellers and buyers [
16]. Contracts may differ based on existing circumstances, but generally, the amount and price of the product are specified [
17]. Risk sharing is the most common reason to establish the contract [
18]. A predetermined price provides a chance to reduce maximum risk in contract farming [
19]. When contract farming is competently organized and well managed, it lessens risk and uncertainty, in contrast to purchasing and selling at open markets [
20,
21]. Some researchers have also studied collective economic organizations that share risk and mitigate moral hazard, comparing them with the relative performance of contracts, and found that contracts are attractive alternatives to risk sharing [
22].
Precautionary saving is a self-insuring strategy for risk reduction used by farmers [
12]. Precautionary savings include accumulation of liquid, semi-liquid and fixed assets, as well as assets in the form of hard cash, crop or yield inventories, livestock, farming tools, farming equipment, and additional useful assets [
23,
24]. Generally, these are extensively used by small farmers as ex-ante shock absorbing tools [
25]. Additionally, age, training or education, the size of the household, earnings, and microcredit are all noteworthy factors which respectively determine farmers’ attitudes towards risks handling [
26,
27,
28].
The maize market in Bangladesh has enormous potential as a subsistence and cash crop, especially for small farmers [
26,
29]. The estimated maize area and production level are projected to increase to 448,000 hectors and 3.54 million metric tons (MMT) respectively in the 2018–2019 period. In 2018–2019, the total consumption of maize is predicted to rise to 4.4 MMT [
30]. The demand for maize has grown in the poultry and fishery sectors as well as human food sectors [
31,
32]. As maize contributes 60% of the animal feed, it is estimated that in 2020 feed demand in the poultry sector alone will be 6.5 MMT, and maize will account for 3.9 MMT [
30]. Also, in Bangladesh, the nutritional requirements of the increasing population are not being met by the traditional crops, which include rice and wheat. This has led to a widespread turn away from traditional rice–rice and rice–wheat cropping systems and toward rice-maize systems [
33], however, natural calamities reduce the profit as well as the attraction of maize production among farmers. It is important to understand the RM strategies of maize for farmers as well as for policymakers. An understanding of farm characteristics, risk attitudes, risk perceptions, and RM usage policies is crucial to fill the demand gap for maize and enable sustainable income for farmers. Therefore, the prime concern of study is to elaborate the information related to the RM strategies of farmers, especially the participation in contract farming and adoption of precautionary savings, and their behaviors. Thus, the main objectives of this study are: (1) evaluation of the impact of farm and farm characteristics, farmers’ risk perceptions, and their risk attitude on their decisions to adopt the two RM strategies; (2) to investigate the linkage between farmers’ decisions for adopting the two RM strategies.
This study also reveals that both RM strategies have a positive impact and most of the farmers are risk averters. This infers that strategies cannot be evaluated solely in terms of expected output, but risk must be considered. Lastly, the findings of risk perceptions and attitudes of farmers regarding RM strategies from bivariate and the multinomial analyses enables us to formulate better policies for agricultural development.
2. Literature Review and Theoretical Background
In this section, some related literature about the different causes of agricultural risk and different factors affecting the decisions of farmers for adopting the risk management (RM) strategies for reducing risk at the farm level are presented.
Hardaker et al. [
2] made a distinction between two important forms of risk in the field of agriculture; firstly, the risk in business posed by the risk in production, and secondly, the financial risk that remains a result of various systems of funding to different farm businesses. As stated by Drollette [
6], in the agriculture sector, production risk is the most important risk. But, according to Harwood et al., the climatic risk is paramount among all production risks for developing countries [
34]. In the field of agriculture, along with the features of the farm and risk situation, RM strategies differ from one another, as presented by Hope and Lingard [
35]. The choices and communications associated among the farmers are independent of risk attitudes, risk perceptions, and purposes, in addition to the existing resources [
36,
37,
38]. A group of researchers similarly found that farmers risk choices differ significantly with their age [
25,
39], literacy level [
40], earnings [
41], farming practice [
42], farmstead size [
43], and land proprietorship status [
44].
Another group of researchers discussed the influencing factors of a single adoption of an RM tool, rather simultaneous adoption, and their linkages [
45]. The implementation of the RM strategies depends on various factors, including irregular conditions of weather, inconsistency in output and input prices, rapidly advancing technology, and farmer indebtedness. Besides these factors, features of farms, farm families, risk observations, and attitudes of farmers considerably affect farmers risk management choices [
46]. The bivariate and multinomial probit models calculated the impacts of different risk managing approaches which are concurrently accessible and usable to the growers [
1,
13,
23,
47]. In the existing literature, the theoretical causes and effects of farmer’s choices about RM are properly acknowledged. However, the simultaneous adoption of multiple risk-coping strategies, i.e., contract farming and precautionary savings, are less studied. Most importantly, a few existing available studies were conducted for different economies, except for Bangladesh. To fill this gap, we investigate the simultaneous adoption of two RM tools among maize cultivators in Bangladesh by considering the bivariate and multinomial probit models. We also measure the influence of several issues on the implementation of these two tools for risk reduction.
The relationship between different risks and associated RM tools are presented in
Figure 1.
6. Conclusions
In modern times, there are several options for managing agricultural risks. Agricultural producers use several RM strategies simultaneously. However, most previous studies have ignored the correlation between farmers’ adoption of RM tools and prospects of using RM tools simultaneously. Therefore, this study was designed to investigate the impact of farm and farm characteristics, as well as farmers’ perceptions of different risks, and their risk attitudes in making decisions about the two selected RM tools—contract farming and precautionary savings. The current research work also describes the link between farmers’ decisions for adopting the two RM tools through bivariate and multinomial probit methods.
The results of this study confirm the correlation between farmers’ adoption of contract farming and precautionary savings in order to manage farm risks. It also concluded that the adoption of one RM tool would encourage farmers to use another RM tool simultaneously. The results indicate the significance of household head’s age and education, the experience of farming, monthly income, and ownership of land, and the influence of risk-averse attitudes of farmers on their decisions in adopting RM tools. However, we recognize that the existence of reverse causal relationships may lead to estimation bias. Although the study covers only four districts of Bangladesh, the results can be generalized in the contexts of all developing countries, specifically in countries where formal RM tools such as crop insurance are ineffective or absent. By analyzing RM choices, it can be understood that better explanations, better speculation, and additional factors which could develop the acceptance of farmer’s RM choices can be provided through both of bivariate and multinomial probit methods. The information provided in this study will help the government understand farmers’ risk behaviors and develop better policies for farmers to improve the performance of risks management tools. The results of this study will help stakeholders who need information to devise better RM tools for minimizing farming risks.
Future research will also benefit from this study as it helps to identify the different sources of risk typically associated with agricultural crops, especially considering the occurrence of different sources of risk and the accompanying problems. The government should make necessary arrangements for conducting training programs through developed extension services for spreading awareness with respect to the adoption of crop insurance as the main risk-reducing tool.