ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846
-A A +A

Do Social Networks Facilitate the Spread of Ponzi Schemes?

Evidence from a Primary Survey

Souvik Dutta (souvik@iimb.ac.in) is with the Indian Institute of Management Bangalore. Abhirup Sarkar (abhirupsarkar.sarkar@gmail.com) is with the Indian Statistical Institute Kolkata.

Ponzi schemes, which have become widespread in some parts of the country, including eastern India and West Bengal, have inflicted heavy losses on the investors and have claimed several human lives. A primary survey shows how social networks created by self-help groups, though they provide higher incomes and social insurance to the members, also facilitate the spread of misinformation regarding Ponzi schemes and hurt naïve investors. The gullibility of potential investors can be reduced if the same networks are used to foster financial literacy.

The authors gratefully acknowledge the research support from the International Growth Centre, New Delhi through a grant which made the primary survey reported in the article possible. They are also indebted to an anonymous referee for helpful comments on an earlier draft of the article.
 

Social and economic networks, which complement impersonal markets, have been observed to play very important roles in shaping up the fabric of a society or economy. Examples of networks creating positive externalities are numerous. Studies have shown that social and economic contacts are important in obtaining information about job openings as well as securing employment. They are also important in developing supply chains and market outlets. They matter in decisions regarding technology adoption, migration, credit disbursement, co-authorship and dissemination of research. On the other hand, there are examples of networks with negative externalities as well. Spreading of computer viruses, expansion of organised drug or human trafficking are indeed facilitated by networks, which have negative impacts on the society (see Jackson [2008] for a detailed analysis of social and economic networks).

While networks are important in both advanced as well as relatively backward parts of the world, they become especially crucial for societies where formal markets are not well-developed. This means that for less developed countries networks are extremely important. One such important network can be observed among microfinance groups. These groups are formed for the purpose of getting small loans from rural banks or microfinance institutions. The repayment of each group member is often joint liability of the group and sometimes it is an individual liability. Be that as it may, the formation of the group itself has important social and economic consequences, in addition to securing the loan. It gives rise to social networks which facilitate other types of economic and social exchanges between the group members.

The authors conducted a survey in two districts of West Bengal, namely Bardhaman and Birbhum. They found that members of self-help groups (SHGs), who came together with the purpose of securing microcredit from a regional rural bank, not only succeeded to earn higher incomes compared with non-members, but also enjoyed social insurance provided by themselves to one another in times of need. The latter externality, created by the social network of the microfinance group, is particularly important because formal insurance markets are almost non-existent in the villages where the survey was conducted. The survey also indicated that the members of the groups, who were all women, experienced improvement in their household decision-making power after they joined the SHGs. The detailed findings of the survey are found in Dutta et al (2017).

The purpose of the present article is to focus on another aspect of the social network created by the SHGs that became evident from our survey. This aspect is decidedly negative. More specifically, our survey reveals that the social networks of the microfinance groups, in addition to providing higher incomes, social insurance and more decision-making power, have helped to spread Ponzi schemes among group members. This has led to substantial losses.

There is an empirical literature emphasising various negative aspects of social networks. Reiss (1980, 1988) finds that two-thirds of criminals commit crimes with others. Glaeser et al (1996) find that social interaction is important in determining criminal activity, especially with respect to petty crime, youth activity in crime, and neighbourhoods with unstable households. Manning (2018) shows that Ponzi investment frauds are facilitated by social capital. Baker and Faulkner (2004) find, through a case study that while investors conducting due diligence before investing face a 49% probability of capital loss, those failing to do so and relying solely on social networks for financial information face a 79% probability of losing their capital. However, to the best of our knowledge, the negative aspect of social networks created through
microfinance groups has never been emphasised before.

In what follows, we discuss the recent growth of Ponzi schemes in India, including in West Bengal where the survey has been conducted. We describe the structure of our survey and talk about the data. Following this the focus is on the incidence and magnitude of Ponzi scheme investments among the respondents of the survey and demonstrate that members of SHGs tend to be more prone to investing in Ponzi schemes than non-members. Finally we present our regression results and conclude the article.

Ponzi Schemes in West Bengal

Ponzi schemes, popularly known as chit funds1 in some parts of the country, have been floated all over India in different forms and structures for quite some time. But they are by no means an exclusively Indian phenomenon. The scheme had got its name from Charles Ponzi, a Boston-based fraudster, who applied basic Ponzi techniques to swindle money from depositors in the 1920s. Several accounts, however, confirm that the idea is much older and had been put into practice by others well before Charles Ponzi. Unfortunately, the method is still in use all over the world. The basic idea is quite simple. The owner of the Ponzi scheme allures investors to deposit money in his scheme by promising much higher interest than what can be earned from the market. To make the promise initially credible the scheme actually pays the promised interest for some time and the news about high interest payments is carefully spread among other potential investors through some network. As more money starts flowing in, old depositors are paid from the new deposits and, of course, substantial amounts are siphoned off by the Ponzi scheme owner himself. It is a bubble. As long as fresh deposits keep pouring in, the bubble becomes larger and larger and the scheme continues. It bursts when new deposits fall short of required payouts and this is bound to happen at some point in time. The larger the bubble, the harder is the shock to the economy when the bubble eventually bursts. When the bubble bursts, it goes without saying, the depositors lose most of their investments.

Understandably, it is not easy to know how much money has been collected by Ponzi operators in a country or region within some specific interval of time. Estimates appear in the media that are sometimes widely different. An Economic Times article published on 4 April 2016 reports that the Central Bureau of Investigation (CBI) has estimated the total amount of money swindled by Ponzi operators all over the country to be more than ₹ 80,000 crore. The estimated number of victims is well over six crore. While these operations were spread across the country, they were especially concentrated in Punjab and some other North Indian states as well as in the eastern states of West Bengal, Assam, Odisha and Tripura. The CBI source estimates that the Pearls Group alone mobilised around ₹ 51,000 crore from unsuspecting depositors. It may be mentioned that the Pearls Group operates exclusively in Punjab and other northern states of India.

Another report, coming out in the Business Standard on 25 April 2015, refers to data collected by the All-India Small Depositors’ Association from agents of 27 money-pooling companies. The data show that these agents col­lectively raised close to ₹ 40,000 crore in West Bengal alone, excluding the money collected by the two biggest money-pooling companies, Rose Valley and Saradha, whose estimated collections were ₹ 15,000 crore and ₹ 2,500 crore ­respectively. So, putting together all Ponzi companies, the total amount mopped up and swindled in West Bengal comes to around ₹ 57,500 crore. This mopping up reportedly took place mostly over the period 2011–14.

Clearly, the two estimates do not match. According to the first estimate the amount of Ponzi fund raised in the four eastern states is around ₹ 29,000 crore. According to the second estimate, the duped amount in West Bengal alone is twice the amount. Of course, the CBI source has said that its estimates are likely to go up because the investigations are still on. But even if the first estimate is revised upwards, it is unlikely to match the second estimate which is based on data collected from agents. Be that as it may, there can hardly be any doubt that the problem is extremely grave and affects a very large number of people most of whom are poor and illiterate. It is also beyond doubt that the problem is serious in the state of West Bengal where we have conducted our survey. According to a report published in the Moneycontrol on 4 April 2017, the Serious Frauds Investigation Office (SFIO) of the central government is currently investigating 185 companies for floating Ponzi schemes out of which 135 schemes were operated out of West Bengal.2

It is not merely a question of monetary loss to the investors. Newspapers are full of reports of suicides, attempts to suicide, lynching of agents by investors and deaths due to stress, all related to the bursting of Ponzi bubbles. For example, after the unravelling of the Saradha scam in 2013 several Ponzi scheme related deaths were reported in different parts of West Bengal.3

One can think of two forces at work. The first force is the floating of Ponzi companies with the sole aim of duping gullible investors. The second is the gullibility itself, the willingness of the people to make quick money without thinking about the huge risks involved. Of late, the union government has put in some effort to contain the first force. It has introduced a comprehensive piece of legislation, namely the Banning of Unregulated Deposit Schemes Bill, 2018, which has been approved by the Union Cabinet in February 2018.4 The bill imposes complete prohibition of unregulated deposit taking activity and provides for severe punishment and heavy penalty to act as deterrent. The state governments have been entrusted with the actual implementation of the bill. The effort is certainly praiseworthy, but clearly it is not enough. Even with appropriate legislation in place, if gullibility persists among the investors, illicit Ponzi companies will surreptitiously keep cropping up and swindle money from unsuspecting small investors in remote corners of the country. Some of these swindlers may be caught and tried from time to time. But it is unlikely that the investors will get back the money they had invested. It is, therefore, necessary to curb the gullibility of the investors.

What are the possible factors responsible for the gullibility? The gullibility, it appears, could come from illiteracy, especially financial illiteracy. The tendency to invest in Ponzi schemes may also be related to the investor’s own and family income. Finally, the misinformation regarding the attractiveness of Ponzi schemes is likely to spread through social networks, through friends, relatives and neighbours, people whom an investor trusts. Through our survey we have made an attempt to understand which of these factors are actually responsible.

Description of the Survey

Our data is based on a primary survey of a random sample of SHGs registered with the Paschim Banga Grameen Bank (PBGB). The PBGB is a regional rural bank (RRB) and was established by the amalgamation of three erstwhile Gramin Banks, namely Howrah Grameen Bank, Bardhaman Grameen Bank and Mayurakshi Grameen Bank in 2007. They have operations in four districts—Bardhaman, Birbhum, Hooghly and Howrah—with their headquarters located in Tikiapara, Howrah.

A typical SHG consists of 10 female members in the village. There are no members from the same household in one group. Once a group is formed they choose a group leader and a group name is assigned. A group then opens a saving account with the PBGB. The group also decides on its basic rules like the frequency of meetings, its location, amount of savings of the group in each month and the minimum amount of savings by its members. A group can also decide to lend money to its members out of the group’s savings and charge an interest rate on such loans. The PBGB encourages such internal loans before providing a loan to the group. Once a group is stable in its functioning, the group then applies for a loan from the PBGB. The loan given by the bank is against no collateral and is a joint liability. The group distributes the loan to its members according to the needs of its members for their projects. A credit-linked SHG is one that has received credit from PBGB whereas a savings-linked SHG is one that has a savings account with the bank but has not received credit yet. As of 31 March 2015 the number of credit-linked SHGs with the PBGB was 34,220 and the number of savings accounts linked was 44,447.

The repayment of the loan is generally made at the meetings. There is some flexibility on repayment within the group, but peer pressure ensures that loans are effectively repaid, so that the default rate for bank loans is very low. Some groups also charge their members an additional interest rate over and above the interest rate charged by the bank. The income that is generated from this additional interest is deposited in the group’s savings account.

Our survey was conducted in the state of West Bengal in July and August 2016 across two districts―Birbhum and Bardhaman. The survey was carried out across 14 villages from two blocks in Bardhaman, and 11 villages across two blocks in Birbhum. The two blocks covered in Bardhaman district are Burdwan-II and Galsi-II, and the two blocks covered in Birbhum are Murarai-I and Nalhati-II. A random sample of 57 SHGs was chosen across these 25 villages with 28 groups in Bardhaman and 29 groups in Birbhum. Each group member in an SHG was interviewed using a detailed questionnaire. A total of 563 SHG members were interviewed, with 281 members in Bardhaman and 282 members in Birbhum. In order to assess the impact of being a member of a SHG, we also surveyed a random sample of women who are not members of SHGs in the same villages. A total of 235 (with 116 individuals from Bardhaman and 119 from Birbhum) such non-members were interviewed using a detailed questionnaire. Thus, a total of 797 individuals were surveyed.

The PBGB operates in four districts—Bardhaman, Birbhum, Hooghly and Howrah. Two districts were chosen from the four. Of the chosen two, Birbhum is under the NRLM mission, which gives additional subsidy on credit to the SHG members, while Bardhaman is not. Therefore, the subsidy received on interest rate is higher for Birbhum. This has dictated our choice of the districts. A list of all SHGs in each block was given to us and we randomly selected the groups from the list. As for the choice of non-SHGs, the enumerators went around the villages in which the SHGs were chosen and randomly selected the non-SHG households.

Demographics

For each member, we collected information on the occupation, marital status, husband’s occupation (if married), income details and other household characteristics. We did the same for the individuals who are not part of any SHG. In addition to this each group leader was interviewed regarding the overall group demographics, number of loans taken, and amount and frequency of weekly meetings, etc. We also collected data on village-level characteristics like population, number of individuals belonging to different religions, castes, etc. Table 1 provides the demographic characteristics of the surveyed women.

We primarily focus on the age, religion, marital status, education and income. The average age of the SHG members is 39 years while the average age of the non-SHG members is 33 years. The survey was carried out in villages in rural areas and hence education levels are pretty low. Apart from the years of education, individuals were also asked to report whether they can read or write. In Bardhaman, 61% of the SHG members said that they cannot read and write, while 44% of the non-SHG members cannot read and write. On the other hand, in Birbhum, 42% of the SHG group members cannot read and write, while 36% of the non-SHG members in Birbhum cannot read and write. Two hundred and seventy-three SHG members (50.4%) have zero years of education, and 160 out of 259 (62%) have zero years of education among non-SHG members.

As reported in Table 1, 85% of the women were married among SHG members and 83% married among non-SHG members. Among the SHG members, 61% of them are Hindu and rest are Muslim, while among the non-SHG members, 63% are Hindu and rest are Muslim. In Bardhaman, out of 397 people surveyed, 85% are Hindu, while the remaining are Muslim. Similar proportions in terms of religion are present when we compare members and non-members in Bardhaman. However, in Birbhum, of the 401 individuals, 57% are Muslim, while the remaining 43% are Hindu. The proportions are similar when we compare members and non-members in Birbhum.

The caste status we consider here is defined by four main caste categories: Scheduled Castes (SCs), Scheduled Tribes (STs), Other Backward Classes (OBCs) and general. Among SHG members, 48% belong to SC, 12% belong to ST, 15% to OBC and the rest 25% to the general category. Among the non-SHG members, 47% belong to SC, 13% belong to ST, 17% to OBC and the rest 23% to the general category. There is almost no ST representation in Birbhum (both among members and non-members), but it has a higher presence of individuals from general category relative to Bardhaman.

Table 1 reports own annual income and this appears to be higher for the SHG members as compared to non-SHG members. The average annual income of a SHG member is ₹ 25,689, while that of a non-SHG member is ₹ 16,779. The same pattern holds for the husband’s income as well. However, there is substantial variation in own income as well as husband’s income for both SHG and non-SHG. For SHG members, the coefficients of variation are 106% for own income and 284% for husband’s income. For non-SHG members, these numbers are 130% and 144% respectively. The proportion of below poverty line (BPL) card holders is higher among the SHG members as compared to non-SHG members. 70% of the SHG members possess a BPL card while only 61% of the non-SHG members do so.

This is somewhat curious. One may ask why SHG members as a category, in spite of having higher own income as well as higher husband’s income on an average, have more BPL households. It is to be noted that while the difference between mean own income is statistically significant between SHG and non-SHG, the difference in mean husband’s income is statistically insignificant. Even though this opens up the possibility that the difference between family incomes between the two groups is statistically insignificant, it still does not explain the difference in BPL percentages which are indeed statistically significant. One possible explanation is that SHG members are better connected both socially and politically which help them obtain BPL status more easily. Our interactions with PBGB officials confirmed this. We were told that the SHG members were chosen not by the bank but by local bodies like the gram panchayat.

In terms of occupation, women in SHGs in Bardhaman are primarily agricultural labourers (53%), bird/animal farmers (13%) and housewives (10%), while women who are non-members are primarily housewives (37%) and agricultural labourers (35%). In Birbhum, women belonging to SHGs are primarily housewives (31%), tobacco/puffed rice makers (17%) and agricultural labourers (15%) and women who are non-members are primarily housewives (69%) and tobacco/puffed rice makers (7%).

Financial Literacy

Next, we look at some of the important variables that would correspond to the financial literacy of the respondents. Table 2 reports such descriptive statistics where it shows that almost 95% of the SHG members and around 97% of the non-members have never approached a commercial bank for a loan. Most of the respondents did not know the interest rate paid by a bank on fixed deposits and the interest rate charged by the commercial bank on a loan. More important, less than 50% of both SHG members and non-members had an account in a bank or a post office and the figure does not significantly improve if their family members are included. Thus, the level of financial literacy among both the SHG members as well as non-members seems to be rather poor.

The situation is actually worse than that suggested by Table 2. While Table 2 might suggest ignorance of members and non-members about the formal financial sector, Table 3 reveals more. Those respondents who reported that they did not have any account in banks or post offices, were asked the reasons behind this. Of those who answered the question, a large proportion of both SHG members and non-members (66% of SHG members and 76% of non-members) said that they had deliberately chosen not to deposit their savings in banks or post offices because they expected to get higher returns elsewhere. “Elsewhere” undoubtedly included Ponzi schemes and it seems that these respondents were not only ignorant about deposits that are safe, but were actually misinformed about investments that are highly risky.

Exposure to Ponzi Schemes

Our survey was conducted in 2016. We asked the respondents about their investments in Ponzi schemes over a period of five years prior to that date. From Table 4 it is clear that a higher percentage of SHG members were aware of Ponzi schemes compared with non-members and the difference is statistically significant. Again, of those who had heard of Ponzi schemes, a much higher percentage of SHG members actually invested in such schemes compared with their non-member counterpart. As percentage of total SHG members, 28.7% had invested in Ponzi schemes. The corresponding figure for non-members is 13.8% and the difference is statistically significant. Again, the average amount of investment in Ponzi schemes for SHG members is 56% higher than the average investment of non-members and this difference is also statistically significant.

A few more things are apparent from Table 4. Though 94% of SHG members and 100% of non-members did not get any return from their investments in Ponzi schemes, very few of them (1% of SHG members and 10% of non-members) were actually aware of the risk involved. Since a large fraction of members (80%) heard of someone who had actually benefited from some Ponzi scheme, and the fraction is only 37.5% for non-members, it seems that the spread of misinformation was higher among members.

Therefore, the evidence suggests that both the incidence and the magnitude of investment in Ponzi schemes were higher among SHG members compared with non-members. What could be the reasons
behind this? One possible reason is, of course, higher income. SHG members, on an average, had statistically significant higher income than their non-member counterparts which, in all likelihood, was due to their SHG activities. The average income differences between the husbands of the members and non-members, on the other hand, were found to be small and not statistically significant. It may be conjectured that higher own income of group members provided them the necessary funds to take up the investments. However, our regression results, discussed below in detail, indicate that neither own income nor husband’s income has an impact on investment in Ponzi schemes.

The factor that seems to be the most important determinant of investment in Ponzi schemes is the social network of friends and relatives. Table 5 indicates that friends have been by far the most important source of information about Ponzi schemes for both members and non-members. For SHG members, the closest friends are likely to be other members of the group. But curiously enough, not a single SHG member reported that she has learned about Ponzi schemes from another member of her group. On the other hand, 75% of the SHG members have reportedly learned about Ponzi schemes from “friends outside the group.” This is hard to reconcile with Table 6 and Table 7 (p 31). Table 6 clearly indicates that for an SHG group member, monetary relationship is much stronger with members within the group as compared with outsiders. This is true both with respect to the possibility of raising a medium or a large loan and with respect to the amount that can be raised. Table 7 further demonstrates that there is sufficient social interaction between group members. It is, therefore, only natural that a typical SHG member will learn about Ponzi schemes primarily from other group members.

It seems likely that the SHG members were “instructed” not to say anything that will implicate the group in Ponzi scheme-related activities. As a result, there was systematic misreporting by SHG members regarding the source from which they learnt about Ponzi schemes. But whether a group member learned about Ponzi schemes from other group members or not, the statistical association between investing in a Ponzi scheme and being in a group is strong and this is more clearly brought out by the regressions to which we now turn.

 

Regression Results

In this section, we discuss the regression results and show that being a member of SHG has a significant positive impact on investments in Ponzi schemes. We use the probit regression model which is given by

Pr (Y = 1|X) = Φ(×ʹβ)

where Y is binary dependent variable. It takes value 1 if an individual has invested in a Ponzi scheme and 0 otherwise. We also have a vector of regressors X, which are assumed to influence the outcome Y. The primary independent variable is a dummy named ingroup which takes a value 1 if an individual is a member of SHG and 0 otherwise. The other independent variables are own years of education, years of education of husband, own annual income and husband annual income. The regression result of Model (1) given in Table 85 shows that the co-efficient of the dummy variable ingroup is positive and highly significant implying that SHG members have a significantly higher probability of investing in Ponzi schemes. Years of education of husband, own annual income and husband annual income do not have any impact on probability of investment while own years of education has a significantly positive impact on the probability of investment in Ponzi schemes. In Model (2), we add the dummy Muslim which takes value 1 if the individual is from the Muslim community and similarly we add broad caste dummies, ST, OBC and GEN which takes value 1 if the member is from the respective community. The results remain unchanged and hence show that belonging to a particular religion or caste does not have any impact on the probability of investment. In Model (3), we add two more dummy variables. The dummy bplcard takes a value of 1 if the member possesses a BPL card and the dummy do you read and write takes a value 1 if the member can read or write in any language. However, we drop two variables own years of
education and years of education of husband. The results remain unchanged and are very similar to Model (1). This gives us confidence that irrespective of different models, the co-efficient of the dummy ingroup remains positive and highly significant.

Our regression results confirm our earlier surmise that being a member of an SHG is closely associated with the probability of investing in Ponzi schemes. Theoretically, we cannot claim any causality between the two; all we can claim is a strong correlation. In other words, we cannot directly claim that being in a group “causes” an increase in the probability to invest in Ponzi schemes. But a bit of reflection should convince us that indeed the group is likely to have exerted some influence on the member to invest in such bogus plans. Let us see why.

To start with, we may dispose of the reverse causality, namely investment in Ponzi schemes “causing” SHG participation, as unlikely. What is possible, however, is that a member having a trait of being more “social” has a higher probability of being a member of an SHG as well as of becoming more vulnerable to Ponzi schemes through her social contacts. But being social should also imply that she socialises with her group members. Indeed, her socialisation is more with her own group members than with outsiders, as reflected in our survey and reported in Tables 6 and 7. So it is very unlikely that she has not learned anything about Ponzi schemes from her closest associates who are her own group members. The fact that none of the SHG members have reported to have learned about Ponzi schemes from other group members can only be explained as “response directed from above” on which we have already commented. In short, while we admit of the possibility that being more “social” can cause both SHG participation and Ponzi investment and hence can explain the significance of the ingroup dummy in our regressions, it is also highly likely that being in a group has enhanced social interactions and therefore the vulnerability of becoming exposed to a Ponzi scheme.

One result reported in Table 8 seems counter-intuitive though. In both Models (1) and (2), investment in a Ponzi scheme is positively and significantly associated with years of own education. While one would have expected that education is likely to caution the investor against risky and spurious investments, one could think of several possible reasons why it may not be so. First, more education may bestow a kind of leadership and prominence on the person which would also make her a target of Ponzi scheme agents, for if the leader invests in the scheme, others are more likely to follow. Second, an educated person may have higher aspirations which may induce her to go for risky investments. These are, of course, mere conjectures and are not substantiated by any evidence.

Concluding Remarks

In this paper we have tried to identify, through a primary survey, the factors that are responsible for the spread of Ponzi schemes among rural households. Our primary survey was conducted in two districts of West Bengal. Our survey suggests that at one level, financial illiteracy in general and ignorance about the risks associated with Ponzi schemes in parti­cular are primarily responsible for the spread of Ponzi schemes. But at another level, even pervasive financial illiteracy cannot guarantee the smooth working of Ponzi schemes unless there is a social network through which misinformation about the schemes can be transmitted. In our survey we found out that SHGs are likely to have provided such a network, so much so that SHG members had a significantly higher tendency to invest in Ponzi schemes compared with non-members.

There is a strong policy implication of our research. Our survey suggests that SHGs build up very strong social networks of mutual trust and friendship. We suggest that these social networks can be used to spread financial literacy. The rural banks or the local governments which set up these groups can take up the task. The ultimate initiative, of course, will have to come from the central and state governments.

Notes

1 The expression “chit funds” is not quite appropriate to describe Ponzi schemes. While Ponzi schemes are essentially illegal, chit funds are perfectly legal instruments, regulated by the Chit Fund Act 1982 of the central government as well as other state laws. They enable savings and credit facilities within a group.

2 Though our study pertains to West Bengal, Ponzi schemes are by no means confined to this state alone. This should be clear from our preceding discussion and a statement made in 2016 by CBI director Anil Sinha which said that the agency was investigating cases spread over 26 states. Two of the biggest scandals, Saradha and Rose Valley were in West Bengal and that had probably put the state into some kind of a spotlight. See http://www.openthemagazine.com/article/openomics-2018/ponzi-schemes-the-....

3 Such reports appeared in the Telegraph on 1 May, 4 May and 5 May of 2013 and in the Kolkata edition of the Times of India on 20 April, 21 April, 24 April, 1 May, 7 May, 8 May, 14 May and 17 May of 2013.

4 A concise description of the bill may be found in a newspaper report appearing in the Hindu, 8 June 2018.

5 The number of observations in Table 8 is less than the total number of people interviewed. This is due to missing data on some variables for some respondents.

References

Baker, Wayne and Robert Faulkner (2004): “Social Network and Loss of Capital,” Social Networks, Vol 26, No 2, pp 91–111.

Dutta, Souvik, Abhirup Sarkar and Suraj Sekhar (2017): “Self-help Groups: Evidence from India,” Project Report, International Growth Centre, New Delhi.

Glaeser, E, B Sacerdote and J Scheinkman (1996): “Crime and Social Interactions,” Quarterly Journal of Economics, Vol 111, pp 507–48.

Jackson, Matthew (2008): Social and Economic Networks, New Jersey: Princeton.

Manning, Paul (2018): “Madoff’s Ponzi Investment Fraud,” Journal of Financial Crime, Vol 25, ­Issue 2, pp 320–36.

Reiss, A (1980): “Understanding Changes in Crime Rates,” Indicators of Crime and Criminal Justice: Quantitative Studies, Bureau of Justice Statistics, Washington DC.

— (1988): “Co-offending and Criminal Careers” Crime and Justice: A Review of Research, M Tonry (ed), Vol 10, Chicago: University of Chicago Press.

Updated On : 13th Sep, 2019

Comments

(-) Hide

EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

Back to Top