Farmer Suicides across Indian States: What Do We Know? What Does it Mean?

October 3, 2015

By Debarshi Das and Deepankar Basu

1. Introduction

Suicides are extremely tragic events that are triggered by despair and hopelessness caused by material and psychological causes. Every suicide is a preventable death, a tragic event that points towards structural deficiencies of the society where it happens. The National Crime Records Bureau (NCRB) of the Ministry of Home Affairs, Government of India, has been collecting and publishing data on suicides in the country for many decades (going back at least to 1967). Analyzing these data can provide important clues about the causes and patterns of suicides across the country and can be potentially used in formulating policies for addressing these tragic events.

From 1995, the NCRB started publishing suicide data disaggregated by professions, a key category being “farmers”. Supplementing these data with reports from the ground, journalists and activists started drawing attention of the country to the specific problem of suicides among farmers. By the early 2000s these efforts had borne some fruits: the tragic phenomenon of farmer suicides in India had been recognized in national and international policy circles. Off and on, there have also been efforts by state and central governments in the country to address the issue.

The publication of the NCRB data in 2014 caused lot of surprise because it apparently showed that farmer suicides had declined dramatically in one year. Journalists, analysts and activists were quick to point out that this dramatic decline was more an artifact of definitional changes adopted by the NCRB than anything else. Once the numbers are calculated in a consistent fashion, the number of farmer suicides is seen to increase, rather than decrease, between 2013 and 2014 (as we explain in detail below).

While annual movements in the total number of farmer suicides are useful and should be monitored closely, it is also necessary to supplement this with more detailed analyses. There are at least two dimensions that would be useful to analyze more closely: (a) longer term trends, and (b) variation across states. In this article, we use data from the NCRB and the Indian Censuses to do both. Our analysis suggests that the general and justifiable concern with farmer suicides needs to be nuanced in various respects.

First, the problem of farmer suicides has not disappeared; it is very much still there. Second, although farmers’ suicide is observed all over the country, it is most concentrated in two states, Kerala and Maharashtra. This does not mean that there are no general patterns to the problems facing the agrarian economy all-through the country or that the spates of suicides that have often been reported from states such as Andhra Pradesh, Karnataka, Uttar Pradesh, Punjab, Madhya Pradesh or West Bengal are unimportant. However, a careful examination of the existing official data reveals that the farmers commit suicide at a far higher intensity in Maharashtra and Kerala compared to the rest of the country. This means that policy intervention to address the troubles affecting the agrarian economy of these two states would be effective in mitigating the problem to a great deal.

Before we present our analysis, a caveat about data sources is in order. The main source of data on suicides, including farmer suicides, which we use in this study is the NCRB. Many activists, journalists, and demographers have pointed out that the NCRB data seriously underestimates the number of suicides because these data are collected on the basis of police reports. Given the stigma attached to suicides and the possible legal problems that surviving family members might face, many suicides are liable to go unreported.

These data problems seem to be operative even in the case of farmer suicides. For instance, the sharp fluctuations in the number of farmer suicides in Chhattisgarh seem problematic. In 2010, Chhattisgarh reported 1126 farmer suicides; in 2010 it had fallen to 0! In 2013, it reported 0 farmer suicides; in 2014, the number had jumped back up to 755. These numbers certainly point towards problems of data collection. To take another example, West Bengal did not report any numbers for 2012 and then reported 0 farmer suicides in 2013. These are perplexing given that the state had been witnessing close to 1000 farmer suicides in the years before 2012.

The point in highlighting these data problems is not to suggest that we jettison NCRB data completely. That cannot be done because, with all its limitations, the NCRB data remains the only national level data source on suicides in India that cover a long time span. The point in highlighting the problems is to keep the caveats in mind while interpreting the results. It might very well be the case that the problem of farmer suicides is much worse than what is captured by the official NCRB data.

2. NCRB and Its Estimation of Farmers’ Suicides

For two decades National Crime Record Bureau (NCRB) has been publishing yearly data of suicides in India disaggregated by professions (the NCRB has been publishing data on suicides by sex, causes and means adopted since 1967; disaggregation by professions was added in 1995). While classifying the data under different professions, the NCRB reports suicides committed by a category of people called the “self-employed (farming/agriculture)”. The NCRB report does not define the category of a self-employed person. But parsing the report one gets an idea about the meaning of the category. “Self-employed” is one of the many categories of professions used by the NCRB, the others being “housewife”, “service”, “unemployed”, “student”, “retired person”, and “others”. Given these categories, it would be safe to assume that self-employed are those who have not been hired by other people. They would have been in the ‘service’ category if that had been the case. Moreover, they earn their living, unlike the unemployed.

In case of farming/agriculture it stood to reason that the self-employed are cultivators. That is, they earn their living from farming, but with the important qualifier that they own the land, or have leased in the land. Thus, agricultural labourers would not fall in this category because agricultural labourers do not own farm land; moreover, they are hired by others.

Earlier analyses of farmers’ suicides were made on the basis of this reasonable assumption. Although the above assumption regarding who is covered by the “self-employed (farming/agriculture)” category of the NCRB did not have much bearing on the content of the article, it is important to provide this clarification upfront because the publication of the NCRB report in 2014 indicated that this assumption was incorrect.

It now appears that the NCRB had been clubbing the suicides by cultivators and agricultural labourers in the “self-employed (farming/agriculture)” category, instead of taking the former category alone. This becomes clear once the data of 2014 is observed. The 2014 data has a different format than the previous years. It has subdivided the category of “self-employed in agriculture” into agricultural labourers, farmers owning land, and those who leased in land. So, there exists a category called the ‘self-employed persons (farmers)’ in the 2014 report. But if one compares the figures reported here with the figures under ‘self-employed (farming/agriculture)’ of previous reports, one would be surprised to find that suicides by farmers halved between 2013 and 2014.

Since this is clearly implausible, one has to compare data under the category of ‘self-employed (farming/agriculture)’ of 2013 (and earlier years) with the category of ‘self-employed persons [agriculture (total)]’ of 2014. Once that is done, the number of suicides is found to have risen from 11772 in 2013 to 12360 in 2014, an increase by about 5%.

 

3. Three Measures of Farmers Suicides

Although the number of suicides has risen between 2013 and 2014, once we look at suicides by farmers over the years, we do not find a clear trend. Data plotted in Figure 1 show that for the first ten years from 1995, suicides went on rising year after year. From over 10,000 per year in 1995 it went to over 18,000 per year in 2004. In the last ten years however there has been a reversal of this trend, with farmer suicides declining continuously. As we have just seen, in 2014 the number was a little over 12,000. So, although the total number of suicides in 2014 is more than in 1995, in the intervening years suicides had gone up sharply and then dropped.

Total number of suicides is a blunt measure to capture the severity of suicide deaths. This is because the number of farmers, some of whom were committing suicides, might be changing. Hence the same number of suicides could have a different meaning because the number of farmers is different. In other words, it may very well happen that in two years the same number of farmers has committed suicide but in the second year the total number of farmers is lower than in the first. This should mean that in a relative sense the suicides were higher in the second year, although in absolute sense suicide numbers did not change.

Figure 1: Number of farmer suicides in India, 1995-2014.

farm_suicides_fig1

In short, one needs a relative measurement to gauge the severity of the problem. That is why researchers estimate number of suicides per 1,00,000 population. This number is called suicide mortality rate (SMR). In table 1 we have reported suicide mortality rate of farmers in India over the last 20 years (column 2).

 

Table 1: Farmer SMR and the ratio of farmer SMR and non-farmer SMR in India, 1995-2014
YearFarmer SMRNon-Farmer SMRFarmer SMR / Non-Farmer SMR
19955.4210.870.50
19966.7510.140.67
19976.5211.000.59
19987.4511.670.64
19997.3212.220.60
20007.3411.710.63
20017.1711.500.62
20027.7611.330.68
20037.3711.250.65
20047.7011.270.68
20057.1511.220.64
20067.0411.510.61
20076.7711.860.57
20086.6011.950.55
20096.6011.960.55
20106.3912.590.51
20115.5712.690.44
20125.4112.500.43
20134.5812.430.37
20144.7611.870.40
Note: Farmer SMR is the suicides of farmers per 1,00,000 farmers; non-farmer SMR is the suicides of non-farmers per 1,00,000 non-farmers.

 

It is clear from table 1, column 2 that, like the total number of suicides, SMR also rose during the first ten years, and declined thereafter. We do not have the data of suicides disaggregated by professions before 1995. So, it is not possible to tell if the rising trend of suicide since 1995 was part of a tendency which had started earlier. In any case, the fact that it rose significantly for a decade after 1995 is borne out by the numbers we have reported in table 1.

But from these numbers can we infer that farmers in particular have been in the receiving end? It can very well happen that during the same time, that is, from 1995 to 2004, suicide mortality rate rose for the entire population. This rising tendency to commit suicide in the whole population could be finding expression in terms of rising SMR among farmers as well. If such is the case then it would be incorrect to say that the root of rising farmer SMR lies in any trouble in the agricultural sector particularly. In that case, it would be wrong to infer that the agrarian economy is in distress, as some of us have been doing.

To check if the rising SMR is specific to farmers, one has to compare between the suicide rate of farmers and those who are not farmers. A simple way to do this is to take the SMR of farmer and divide it by the SMR of non-farmers. If this ratio is significantly higher than 1 at any point in time, it would imply that the suicide rate among farmers is much higher than non-farmers at that point in time; on the other hand, if the ratio were rising over time, then it would indicate that over time suicide incidents are affecting the farmers more than the non-farmers.

We have calculated this ratio and the numbers are reported in the column 3 of table 1. A close look at the number confirms that this ratio of SMRs underwent a gradual rise from 1995 till about 2004, after which it fell. Thus, it appears that all the three indicators we have examined – total number of suicides, SMR of farmers, and the ratio of farmers’ SMR to non-farmers’ SMR – have undergone the same pattern of change over the last two decades: a rise in the first ten years (1995 to 2004) and fall in the last ten (2004 to 2014).

Before we discuss the possible reasons responsible for this identifiable pattern a few observation on the data are in order. To interpret the SMR ratio, it is useful to distinguish between levels and trends. If we consider the level of the SMR ratio, we see from column 3 of Table 1 and from (the blue line in) Figure 3, that it was always below unity. At its maximum in 2002 and 2004, the SMR ratio was 0.68. The fact that the SMR ratio has always been below unity at the all-India level suggests that the level of distress (that is a cause for suicides) has been relatively lower among farmers than among non-farmers if we consider the country as a whole. Even in the worst years (2002, 2004), the relative distress among farmers was lower than among non-farmers. This is an important fact that has often been ignored in discussions about farmer suicides.

When we consider the trend, we see the following pattern: the SMR ratio displayed an increasing trend between 1995 and 2004; thereafter, the SMR ratio declined. Thus, for the decade long period between 1995 and 2004, things were going terribly wrong in the agricultural sector at the all-India level so that the relative rate of suicides among farmers was increasing in comparison to non-farmers. The fact that the SMR ratio fell since 2004 is perhaps an indication that those drivers of distress that led to the spurt of farmer suicides between 1995 and 2004 have mitigated to an extent in the subsequent period.

 

Figure 2: Ratio of farmer SMR and non-farmer SMR in Kerala and Maharashtra, 1995-2014. SMR of farmers = suicide mortality rate for farmers (farmer suicides per 1,00,000 farmers) ; SMR of non-farmers = suicide mortality rate for non-farmers (non-farmer suicides per 1,00,000 non-farmers).

­­

farm_suicides_fig2

Second, although the ratio of farmers’ SMR to non-farmers’ SMR was less than 1 this is true at the all-India level only. If we focus on individual states, we see that two states have defied that trend: Maharashtra and Kerala. In figure 2 we have plotted the SMR ratio for Maharashtra and Kerala. In figure 3, we have plotted the all-India SMR ratio with and without these two states, to show the importance of these two states in driving up the all-India SMR ratio.

Figure 3: Ratio of farmer SMR and non-farmer SMR in India with and without Kerala and Maharashtra, 1995-2014. SMR of farmers = suicide mortality rate for farmers (farmer suicides per 1,00,000 farmers); SMR of non-farmers = suicide mortality rate for non-farmers (non-farmer suicides per 1,00,000 non-farmers).

farm_suicides_fig3

Figure 2 shows that in these two states the SMR ratio has been more than 1 for a number of years. In Kerala the ratio has fluctuated around a value that is greater than 2; in Maharashtra, it has stayed above 1 for most years after 2000. It implies that although Maharashtra occupies more media space on the subject of farmers’ plight – which is justifiable given that far more suicides are committed in Maharashtra than anywhere else – the condition of farmers is grimmer in Kerala in a relative sense. Moreover, one observes a gradual rise in the ratio over the years in both these states. Thus, unlike the all-India picture, the suicide indicators have not come down at all in these two states after 2004.

Figure 3 shows that if Maharashtra and Kerala are taken out, the all-India SMR ratio improves a great deal. The red line (all-India data without these two states) is consistently and appreciably below the blue line (all-India data including these two states). Moreover, once Kerala and Maharashtra are removed, the SMR ratio declines continuously from 1996 onwards. This has important policy implications that we discuss below.

 

4. Possible Reasons for Suicide

It is a well-known fact that the agricultural sector has been performing badly since the 1990s. It has been found that the annual growth rate of output in agricultural sector has halved between the period 1980-83 to 1990-93, to the period 1990-93 to 2003-06. It is also evident that a major reason for this slowdown has been the deceleration in the use of inputs. In other words, investment in agriculture has been declining which could be leading to slowing down of output growth. As has been argued by one of us, slowdown of public investment has played an important role in the decline of overall agricultural investment.

As output is rising at a slower rate since 1990, profitability of farmers is getting adversely affected: for they are not being able to produce at as fast a rate as they had been doing before. Note, decline in public investment is a conscious government decision. There are other policy changes which aggravated the decline in profitability. As has been argued in the sanhati article these include, less enthusiasm for disbursing rural credit by the government, slackening of procurement operation of crops from farmers, opening the domestic market to cheap imports, etc. All these factors increased the cost of farming while at the same time reduced the revenue which the farmer obtained by selling her crop. In short, profitability fell. Low profitability is a reason for the depressed state of Indian agriculture since the early 1990s. This depressed state might have set the background for the spate of suicides which intermittently kept occurring across the country.

Without denying the role of agricultural stagnation, low investment and declining profitability, we would also like to highlight a puzzling feature of the issue: the problem of farmer suicides seem to be most acute not in the poorest states (like Bihar and Uttar Pradesh), but in relatively prosperous and faster growing states (like Maharashtra, Kerala, Andhra Pradesh). A few case studies and journalistic reports have highlighted problems specific to the growing of cash crops – like cotton or spices – that seem to be important in causing farmer suicides. For instance, factors like high volatility of cash crop prices and lack of crop insurance mechanisms, requirements of relatively large investments (and hence for loans) in the cultivation of cash crops, lack of social and mental health support systems to deal with distress, might also be important to take into account if a proper understanding of the problem of farmer suicides is to be developed.

5. Cultivators and Agricultural Labourers

Low profitability directly affected the farmer, i.e., the person who cultivates and owns the land, for she is the person who earns profit and suffers losses. But it indirectly affects the agricultural labourers as well. Low profitability implies low investments and therefore low job generation, which adversely impacts the employment prospects of agricultural labourers. It is found that while between 1983 to 1993-94 employment in the primary sector grew at an annual rate of 1.35%, the rate subsequently fell to 0.67% (1993-94 to 2004-05) and then became negative, -0.13% (1999-2000 to 2009-10) [agricultural sector is the largest component of the primary sector]. Thus the number of jobs in farming is actually shrinking, instead of rising. Between 2001 and 2011 the total number of farmers (i.e., owner farmers and laboures taken together) went down from 234.1 million to 225.1 million. It is not surprising to find labourers in the ranks of those who committed suicide.

It is true that indebtedness is often found to be a reason for farmer suicides. These loans are often production loans which the owner-farmer took, and which could not be paid due to crop failure. Such suicides would not impact labourers as much as the owner-farmers. But it must be kept in mind that labourers are the poorest section of agrarian economy. Economic desperation which drives people to take their lives would be far more important to agricultural labourers than to owner-farmers.

We have the data of suicides by agricultural labourers categorised separately from owner-farmers for the year 2014. Thus it is possible to gauge the relative severity of suicide deaths among agricultural labourers, as compared to owner-farmers. In table 2 we have reported the SMR of these two categories in different states. It can be seen that at the all India level, SMR of cultivators (owner farmers) is not much different from that of agricultural labourers: it is marginally higher. But there is lot of variation across states.

In Maharashtra, SMR of cultivators is quite high – it is twice the value of the SMR of labourers. In the land deficient Kerala, the opposite is true. SMR of labourers is more than three times as high as the SMR of cultivators. For other high farmer suicide states, we see the following pattern: In Andhra Pradesh and Karnataka they are almost similar. In Chhattisgarh, like in Maharashtra, cultivators have a higher SMR. In Tamil Nadu labourers have a higher SMR. In Madhya Pradesh cultivators have the higher figure. West Bengal reports zero suicides for cultivators; this makes the data suspect. In general it appears that land deficient states have higher SMR for labourers compared to cultivators. The fact that labourers are also getting killed in large number, in fact more in absolute number than the cultivators, is an important observation which is easy to lose sight of.

 

Table 2: Suicide Mortality Rate (suicides per 1,00,000 population) for Agricultural Labourers and Cultivators across Indian States, 2014
 Agricultural LabourersCultivators
Andhra Pradesh2.792.75
Arunachal Pradesh7.830.00
Assam1.920.51
Bihar0.050.00
Chhattisgarh5.7211.68
Goa Daman and Diu0.000.00
Gujarat8.010.88
Haryana7.190.63
Himachal Pradesh16.011.54
Jammu and Kashmir4.311.13
Jharkhand0.080.00
Karnataka6.405.15
Kerala73.2419.28
Madhya Pradesh2.859.08
Maharashtra10.6820.69
Manipur0.000.00
Meghalaya1.000.00
Mizoram22.820.00
Nagaland0.000.00
Orissa1.360.13
Punjab2.671.34
Rajasthan6.880.00
Sikkim0.0031.21
Tamil Nadu9.541.88
Tripura8.920.00
Uttar Pradesh0.620.36
Uttarakhand0.000.00
West Bengal2.160.00
India4.545.04

 

6. Decline in Suicides

The points discussed above give us a clue regarding the possible reasons for the decline in the farmer suicide rate after 2004. Although the overall direction of government policies did not change since 2004, some important welfare initiatives were put in place in last 10 years. Two of these could have had a salient impact on the rural living conditions, especially households at the very bottom of the income ladder.

The first is the National Rural Employment Guarantee Act, which came into effect since 2006. It started in 200 backward districts with assurance of jobs of 100 days a year for a member per household. The coverage was gradually expanded in three phases. By 2009-10 all 618 districts of the country came under its ambit. NREGA is the biggest employment generation programme of its kind in the world. Since 2009-10 annually 50 million households have been getting the benefit of NREGA jobs, with an average of 42 days of work per year per household. In 2013-14, 74 million individuals in 48 million households in rural India were employed under the programme with each household on average finding work for 46 days. Although its implementation has been often criticised, NREGA gave a modicum of assurance of income to the rural population in the areas where it worked. The impact of NREGA on rural poverty has been acknowledged, and this might have been important in reducing the severity of farmer suicides too.

The second is the reform in the public distribution system which has been undertaken by pro-active states since the early 2000s. Universal PDS was discarded in favour of targeted PDS in 1997. It led to a decline in the reach of subsidised food to the rural poor. A few states started to make changes within the targeted PDS system in order to make it work. Some of the states refused to follow the Central Government policy of restricting access of cheap grain to the so called below poverty line (BPL) segment – they themselves incurred the cost of grain which they supplied to the non-BPL sections. The results started to show from the mid-2000s. Per capita consumption of rice gradually rose from 0.7 kilogram in 2001-02 to 1.18 kilogram in 2007-08. Researchers have estimated the positive impact of food distribution through PDS on poverty reduction. By making subsidized food grains available to the poor, a reformed PDS might have disproportionately affected agricultural labourers – the most destitute and numerous section of rural India – and helped in reducing the number of farmer suicides.

 

7. In Lieu of a Conclusion

There is no scope for complacency from the observation that farmer suicides have declined in the last ten years. We have noted the rise in the number of farmer suicide in the last one year in the beginning of this article. This rise is reflected in the other two indicators as well (farmers’ SMR, SMR ratio).

In this context, policy gestures of the present NDA government are a matter of concern. NREGA works have been derided at the highest level, which does not inspire confidence that the present dispensation is keen to continue funding the programme for long. Indeed allocation for NREGA has stagnated in last few years even before NDA came to power. With PDS also, the long run policy direction seems to be towards folding up distribution of food by the government. In its place a more market-oriented cash transfer scheme could be installed.

It is important to emphasise that the states which have the highest SMR ratio (ratio of farmer SMR and non-farmer SMR) and which contribute a large part of total farmer suicides, Maharashtra and Kerala, have not had a decline in the suicide rate. This is a point we have noted before and is worth reiterating because it has enormous policy implications. These are the only two states where the SMR ratio has been generally higher than unity and where this ratio has not declined over time. All other states have seen a decline in the SMR ratio over the years, especially since 2004. Thus, without denying the problem of general agrarian stagnation in other parts of the country, it is important that activists and policy makers focus their energy and attention on these two states to address the problem of farmer suicides. Because of the enormity of the problem in these states, they might require special intervention.

It is also important to understand the meaning of the declining SMR ratio is all states other than Kerala and Maharashtra. Recall that the SMR ratio is the ratio of farmer SMR and non-farmer SMR. Thus, when the SMR ratio trends downward, it may mean that distress (that is a cause of suicides) among non-farmers rise relative to distress among farmers. The fact that the SMR ratio has been falling in many states other than Kerala and Maharashtra means that the level of distress among the non-farmer population in many states other than Kerala and Maharashtra has been becoming worse compared to farmers.

We can see this more explicitly in Figure 4, where we have plotted the SMR for farmers and SMR for non-farmers separately at the all-India level. From the figure we can see a clear divergence in the trend movements of farmer SMR and non-farmer SMR from 2004 onwards. While the farmer SMR declined from 7.7 in 2004 to less than 4.8 in 2014, the non-farmer SMR increased from about 11.3 in 2004 to 12.7 in 2011, before falling to 11.9 in 2014.

If we juxtapose this fact to the well-known trend of movement of the workforce out of agriculture, we realize that the declining SMR in most of the country implies a growing distress among an increasing part of the workforce. But who are the non-farmers? In India (and much of the Third World), the vast majority of non-farmers are informal sector workers, workers who work in hazardous conditions, for very low wages, with no legal protections or social security, and hardly any rights of collective bargaining. Thus, while the government should re-double its efforts to address the problem of farmer suicides in Kerala and Maharashtra, it needs to step up its work to address the specific problems of non-farmers in the other states of the country.

 

Figure 4: Farmer SMR (farmer suicides per 100000 farmers), and non-farmer SMR (non-farmer suicides per 100000 non-farmers) in India, 1995-2014.

farm_suicides_fig4

Data: Sources and Definitions

We capture the severity of farmer suicides by the ratio of the farmer suicide mortality rate (SMR) and the non-farmer suicide mortality rate. By “farmer” we mean a person who is either a cultivator (someone who works on owned or hired land) or an agricultural labourer (someone who works on others’ land). The SMR for farmers is defined as the number of farmer suicides per 1 lakh farmer population; the SMR for non-farmers is defined in an analogous manner as the number of non-farmer suicides per 1 lakh non-farmer population. We calculate SMRs for farmers and non-farmers at the all-India level and for all major states.

The data on the number of farmer suicides and total suicides have been extracted from various years of Accidental Deaths and Suicides in India, an annual publication of the National Crime Records Bureau. The difference between total and farmer suicides gives us the number of non-farmer suicides.

The data on the population of cultivators and agricultural labourers have been taken from reports of the Censuses for 1991, 2001 and 2011. The number of farmers is the sum of the number of cultivators and agricultural labourers in rural India.

1 Comment »

One Response to “Farmer Suicides across Indian States: What Do We Know? What Does it Mean?”

  1. bhaskararao Says:
    March 3rd, 2018 at 12:22

    NCRB database analysis leads to skewed picture and so this article analysis is unacceptable. for example our (OPDR,a democratic rights organisation) field report of Ananthapur district of Andhra Pradesh for the period June 2014 to December 2017 349 farmers committed suicides but government( police station records) show 150 only. compensation is paid to 19 only is a related horrible story.

Leave a comment