Government employment guarantee and its
effect on bank lending: evidence from the
largest public workfare program in the
world
Harm Helmus (s2760738)
Thesis MSc. Finance University of Groningen Faculty of Economics and Business
Supervisor: A. Rauf, CFA, PhD.
Abstract: This study examines the impact of the Indian employment guarantee scheme on the amount of credit outstanding. The general findings of the study suggest that overall bank lending increased due to the workfare program. This effect is attributed to increased funds to a specific region and its local scale economies next to the improved infrastructure due to the program. Moreover, this study finds that the workfare program leads to increased lending in the agricultural, industrial, and financial sector.
Keywords: Public Workfare, Bank Lending, NREGA, India
1
1. Introduction
Should governments provide unemployment insurance and, if so, what should the provisions be? Among such policies that affect labor and the economy, public workfare has become more popular in low and middle-income countries, as nearly two-thirds of these countries have public workfare programs. There has been an ongoing debate among researchers on the effects of such workfare programs on poverty, consumption, employment and crime. What, to the best of my knowledge, remains unexplored is the link between workfare programs and bank lending. This study attempts to fill the gap by examining the effects of a workfare program, the ‘Mahatma Gandhi National Rural Employment Guarantee Act’ (NREGA)’ on credit supply.
The main contribution of this paper is to document empirical evidence if the workfare program leads to increased lending. Given the general consensus that there is a positive relationship between financial development and economic growth (see Caporale et al. (2014)), it seems beneficial for a government to employ a workfare program if it indeed leads to increased lending. This may be even more so the case for countries with an undeveloped financial system, as this is one of the reasons why developing or transition countries have low rates of growth (see Miskhin (2004)).
Based on workfare data from the official NREGA website, running from 2006 until 2019 and bank lending data from the official website of the Reserve Bank of India, running from 2002 until 2019, this study aims to examine the effect of a workfare program on credit outstanding in an economy. When using a fixed effects model, controlling for district and state by year effects and clustering the standard errors at the district level, I find that the workfare program indeed leads to increased lending in the overall economy. The intuition behind these findings is that of local scale economies. Furthermore, I find that better infrastructure due to the welfare program leads to higher lending in the economy in the next year. More specifically, I find that the agricultural, industrial, and financial sector see an increase in lending a year after the infrastructure has been improved. The intuition behind these findings is that the improved infrastructure leads to higher productivity, lower prices for goods, increased demand and thus higher lending to keep up with demand.
2
Related Literature. Several studies investigated the effect of NREGA on different factors. For
example, Imbert and Papp (2015) find that the act leads to increased wages in the private sector. These findings are supported by Berg et al. (2012) and Azam (2012), who found the same results. Moreover, research by Ravi and Engler (2015) shows that the act leads to higher incomes and consumption, particularly nutrition. This is supported by Sukhtankar (2017) who finds that an increase in income is likely in places that implemented the program well. Furthermore, Gehrke (2019) finds that the act led to farmers choosing a riskier profile of their crop mix due to the insurance effect of employment, leading to higher productivity. All these positive findings, however, do not come without a cost. Research has shown educational enrollment and attainment worsened for older children, due to the opportunity cost of time: working is more profitable, and these children will drop out of school (Shah and Steinberg (2015)). Moreover, when looking at the link between economic growth and increased infrastructure prior research shows that the benefits and importance of transportation infrastructure has been recognized for a long time (see Phang (2013)). Raguram & Babu (2001) argue that, in India in particular, transportation has been noted to be required for economic growth. Pradhan and Bagchi (2013) show that an expansion of transport infrastructure (both road and rail) along with gross capital formation will lead to substantial growth of the Indian economy.
This study contributes several ways to the academic literature and policymakers. First, as to the best of my knowledge, the effects of a workfare program on bank lending has not yet been explored, it adds valuable insights that a workfare program has a positive effect on credit outstanding. Second, it shows evidence that increased infrastructure due to a workfare program leads to increased lending in the overall economy. This increased lending is indicative of increased economic growth, showing additional benefits of the workfare program. Third, when looking at lending to specific sectors, it shows that not only the targeted industry sees increased lending, but also other sectors, implying the workfare program positively influences other sectors. Lastly, this study shows that the workfare program leads to increased lending in the financial sector, such increased lending may indicate an improved financial system, which is fruitful, especially for emerging countries, as such countries are more likely to have an underdeveloped financial system.
3
2. Theoretical Motivation and Hypotheses Development
2.1 Background of NREGA
The National Rural Employment Guarantee Scheme of India is the world’s largest workfare program that comes with a guarantee of 100 days of employment to every citizen of India as guaranteed by the Mahatma Gandhi National Rural Employment Guarantee Act, 2005. The act has been designed to address seasonal unemployment mainly in the agricultural economy and therefore its performance is organically linked to agriculture. Nevertheless, projects undertaken under the act include the construction of productive infrastructure such as roads as well. The program is demand-driven, meaning that if a person demands work, he/she is to be provided work within 15 days. In case work is not provided, the state (as per the act) will pay an unemployment allowance to the beneficiary.
The act was passed by Parliament and notified in 2005, following up on an electoral promise made by the political party ‘UPA’ after it came into power in 2004. The act became operational in the 200 “poorest” districts, determined by the government’s Planning Commission, in the country on February 2, 2006. An additional 130 districts received the program in the financial year 2007-2008, and the remaining districts in India followed in April 2008. In order to incentivize States to generate employment, the Central government pays all labor costs fully, but only 75% of material costs.
2.2 Multiplier effect
4
H1: Increased NREGA activity leads to higher lending in every sector due to the multiplier
channel
2.3
Increased infrastructure
Besides money flowing into rural areas, projects undertaken under the employment schemes include the construction of productive infrastructure such as roads, irrigation, drought-proofing and flood management. A well-oiled transportation infrastructure expands the productive capacity of a nation, both by increasing the mobilization of available resources and by enhancing the productivity of those resources. For example, a well-designed road allows goods to be transported to the market in a shorter time period, and, hence, reduces the transportation costs in the production process. This has been proved by Pradhan and Bagchi (2013) who found that an expansion of transport infrastructure (both road and rail), along with gross capital formation, will lead to substantial growth of the Indian economy. However, this increased infrastructure is mainly built by skilled workers. Given that NREGA workers are unskilled workers, I analyze whether infrastructure improvements by NREGA workers lead to increased lending. This leads to the following hypothesis:
H2: Increased NREGA completed works lead to future increased lending in the overall
economy
The above hypothesis is explored further, as looking to lending in each specific sector may give more detailed insights. This leads to the following hypothesis:
H2a: Increased NREGA completed works lead to future increased lending in every sector
3. Data and Methodology
3.1
Methodology
I estimate the effect of NREGA on lending activity in a fixed effect model. To my knowledge there is no credible paper analyzing the effect of NREGA on bank lending in India. Hence, I had to explore my own ways to measure NREGA activity. I found total works completed in a year to be a suitable measure to look at increased infrastructure. Moreover, I use total expenditure, consisting of labor expense and material expense, households allotted and demanded work to estimate the activity of NREGA in general to test for the multiplier effect. The specification I employ to analyze the effect of NREGA is as follows:
𝐿𝑒𝑛𝑑𝑖𝑛𝑔𝑑 𝑠𝑡 = 𝛼 + 𝜔d + ϒst + β ∗ 𝑋𝑑 𝑠𝑡−𝐿 + ε𝑑𝑠𝑡
5
controlling for aggregate macroeconomic shocks and trends. The standard errors are clustered at the district level.
The coefficient of interest is β which shows the effect of NREGA activity on lending. A positive correlation between NREGA activity and credit outstanding implies that higher NREGA activity leads to more credit outstanding in the Indian economy.
Perhaps the greatest issue with equation above is endogeneity. Basically, this means the equation above would have a missing explanatory variable inside of the error term. One such variable may be a change in bank-specific regulations, like the capital to asset ratio.
Fortunately, this issue can be tackled as the act was implemented in a phase-wise manner, as mentioned in section 2.1. This staggered implementation lends itself perfectly to a generalized version of the difference-in-difference setup for multiple treatment groups and time periods (See Bertrand and Mullainathan (1998), House and Shapiro (2006), and Agarwal et al. (2020)). In this study, a district stays in the control group until the program is implemented in that district and, after, becomes a part of the treatment group. For example, all phase 1 districts belong to the control group during 2002 to 2006, as this is when phase 1 districts were treated, and serve as part of the treatment group from year 2007 until 2010. Therefore, I have information for up to four years before the launch of NREGA and four years thereafter.
Formally, the regression is as follows:
𝐿𝑒𝑛𝑑𝑖𝑛𝑔𝑑𝑠𝑡 = α + 𝜔d+ ϒst + β ∗ 𝑃𝑜𝑠𝑡 𝑁𝑅𝐸𝐺𝐴𝑑𝑡 + ε𝑑𝑠𝑡
Here, once again, the unit of observation is lending. Post NREGA is a dummy variable, which takes the value of one for a treated district in the post NREGA period. 𝜔d refers to district fixed
effects absorbing all time-invariant differences between districts. Finally, ϒst refers to state by year fixed effects, controlling for aggregate macroeconomic shocks and trends. The standard errors are clustered at the district level.
6
3.2
Data
The data used in this study are from two sources. The first isfrom the official website of the Reserve Bank of India1 and contains bank lending data. The unit of observation is a bank-year
and spans the period from 2002 up until 2019 and contains information for lending in each state and district in India. The credit outstanding is noted for the following sectors: 1) Agriculture, 2) Industry, 3) Transport operators, 4) Professional and other services, 5) Personal loans, 6) Loans in the financial sector, 7) Other. Only other is not used in this study, as this does not lead to any useful insights. Next to credit outstanding per sector ‘Total Credit Outstanding’ is used in this study to note overall lending in India.
Second, I collected information regarding 1) the number of completed works, 2) total expenditure, consisting of both labor expenditure and material expenditure, 3) total households demanded work, and 4) households allotted work under the NREGA from the official website of Mahatma Gandhi National Rural Employment Guarantee Act2. The unit of observation for this data is a district-year and spans the period between fiscal years 2006 up until 2019. Unfortunately, the data from the period 2006-2010 and 2011-2019 is different. The main difference is that after 2010 the precision and accuracy of documentation increased. For example, from 2006-2010 total work demanded and allotted was only noted for households. From 2011 onwards, however, work demanded was measured not only in households, but also in persons.
All data is classified not only per state, but also per district and year, lending itself to a panel structure. Moreover, all data has been transformed to natural log scale, giving a more symmetric distribution, making it easier to model. Lastly, the data has been trimmed at the 1st and 99th percentile to remove outliers.
3.2.1
Summary of dataTable 1 reports the summary statistics of my sample. In total there are 698 districts. This is also due to districts reorganizing, as having data from pre-split and post-split, some districts split up, hence are counted as three separate districts: one pre-split and two post-split. As can be seen, lending in the industrial sector is the most volatile, as it has the highest standard deviation and personal loans is the least volatile. Furthermore, the median value for agriculture is the highest, implying the agricultural sector is the largest lender, as so to say and the financial sector the lowest. Moreover, it can be seen works completed is by far the most volatile when looking at NREGA data.
1 https://www.rbi.org.in/
7
Table 1: Summary statistics. Panel A describes the summary statistics of the bank lending data.
All variables are noted in log and the sample period is from 2002 until 2019.
The name of each sector implies lending in that given sector and total credit means the total credit outstanding in the Indian economy.
Panel B describes the summary statistics of the NREGA data.
All variables are noted in log and the sample period is from 2006 until 2019. Work allotted and work demanded is noted per household. Works completed indicates the total amount of works completed in one year. Total expenditure contains both labor and material expense in one year.
8
4
Results and Discussion
4.1
Increased NREGA activity on total credit
Table 2: Impact of households allotted work on total credit outstanding.
This table reports the results of the regression of log total work allotted to households on log total credit outstanding in the Indian economy in a given year. The observation level is bank-firm-year with bank data running from 2002 until 2019 and NREGA data from 2006 until 2019. L1, L2, L3 means one, two, and three lagged years respectively.
Significance of the parameters are indicated as follows: ∗ p < 0.10, ∗∗ p < 0.05, ∗∗∗ p < 0.01 Standard errors are clustered at the district level and are shown in the parentheses.
(1) (2) (3) (4) Total credit outstanding Total credit outstanding Total credit outstanding Total credit outstanding Work Allotted 0.028** (0.013) L1. Work Allotted 0.017 (0.013) L2. Work Allotted 0.011 (0.011) L3. Work Allotted 0.006 (0.012) Constant 16.172*** 16.366*** 16.495*** 16.620*** (0.145) (0.140) (0.117) (0.127) District fixed effects
Yes Yes Yes Yes
State by year fixed effects
Yes Yes Yes Yes
9
Table 3: Impact of households demanded work on total credit outstanding.
This table reports the results of the regression of log total work demanded households on log total credit outstanding in the Indian economy in a given year. The observation level is bank-firm-year with bank data running from 2002 until 2019 and NREGA data from 2006 until 2019. L1, L2, L3 means one, two, and three lagged years respectively.
Significance of the parameters are indicated as follows: ∗ p < 0.10, ∗∗ p < 0.05, ∗∗∗ p < 0.01. Standard errors are clustered at the district level and are shown in the parentheses.
(1) (2) (3) (4) Total credit outstanding Total credit outstanding Total credit outstanding Total credit outstanding Work Demanded 0.026* (0.013) L1. Work Demanded 0.013 (0.013) L2. Work Demanded 0.013 (0.011) L3. Work Demanded 0.005 (0.012) Constant 16.198*** 16.401*** 16.480*** 16.632*** (0.145) (0.140) (0.118) (0.126) District fixed effects
Yes Yes Yes Yes
State by year fixed effects
Yes Yes Yes Yes
Observations R² 7160 0.98 6541 0.98 5911 0.98 5276 0.98
10
The results of both table 2 and 3 provide evidence supporting the first hypothesis that there is a relation between NREGA activity and total bank lending. The positive relation can be interpreted as increased funds flowing into rural areas leading to higher lending in the overall economy. One could argue, however, that these workfare programs do not lead to increased flows in rural areas, as workers previously employed in the private sector now work in the workfare program as a substitute. This argument, however, is refuted by Sukhanktar (2017) who found no such effect. In fact, they found workers in NREGA actually had a decline in self-employment and idle time. Although a decline in self-self-employment may not lead to an increase in flows to rural areas, substituting idle time for employment certainly does.
Table 4: Impact of total expenditure on total credit outstanding
This table reports the results of the regression of log total expenditure on log total credit outstanding in the Indian economy in a given year. The total expenditure consists of labor expense and material expense. The observation level is bank-firm-year with bank data running from 2002 until 2019 and NREGA data from 2006 until 2019. L1, L2, L3 means one, two, and three lagged years respectively.
Significance of the parameters are indicated as follows: ∗ p < 0.10, ∗∗ p < 0.05, ∗∗∗ p < 0.01. Standard errors are clustered at the district level and are shown in the parentheses.
(1) (2) (3) (4) Total credit outstanding Total credit outstanding Total credit outstanding Total credit outstanding Total Expenditure 0.020** (0.010) L1. Total Expenditure 0.017* (0.009) L2. Total Expenditure 0.005 (0.009) L3. Total Expenditure 0.011 (0.010) Constant 16.309*** 16.401*** 15.567*** 16.599*** (0.082) (0.076) (0.074) (0.076) District fixed effects
Yes Yes Yes Yes
State by year fixed effects
Yes Yes Yes Yes
11
Table 4 reports the link between NREGA activity and total credit outstanding again, however, this time, NREGA activity is measured by total expenditure. As can be seen, when total expenditure goes up by 1%, total credit outstanding goes up by 0.020% in the same year and 0.017% in the next year. The extra lagged effect may be explained by the fact that total expenditure involves the actual flow of money and may be stronger evidence of the multiplier channel, hence showing an additional lagged effect. This is further evidence for hypothesis 1 to be true.
Table 5: Impact of total completed works on total credit outstanding
This table reports the results of the regression of log total works completed in a year on log total credit outstanding in the Indian economy in a given year. The observation level is bank-firm-year with bank data running from 2002 until 2019 and NREGA data from 2006 until 2019. L1, L2, L3 means one, two, and three lagged years respectively.
Significance of the parameters are indicated as follows: ∗ p < 0.10, ∗∗ p < 0.05, ∗∗∗ p < 0.01. Standard errors are clustered at the district level and are shown in the parentheses.
(1) (2) (3) (4) Total credit outstanding Total credit outstanding Total credit outstanding Total credit outstanding Total Works 0.009 (0.006) L1. Total Works 0.012* (0.006) L2. Total Works 0.002 (0.006) L3. Total Works -0.001 (0.005) Constant 16.395*** 16.440*** 16.609*** 16.708*** (0.060) (0.063) (0.059) (0.050) District fixed effects
Yes Yes Yes Yes
State by year fixed effects
Yes Yes Yes Yes
12
Table 5 shows the effect of total works completed on total credit outstanding up until the third lagged year. All models have an R² of 0.98, implying all models do well in explaining the variation. As can be seen, completed works only have an effect on total credit in the next year. More specifically, an increase of 1% in total works completed leads to an increase of 0.012% in total credit outstanding in the next year, but has an insignificant effect in other years. This is in line with hypothesis 2, which links an increase in completed works to total credit outstanding due to a better infrastructure. As argued, an improved infrastructure increases the mobilization of available resources and increases the productivity of those resources. Therefore, to keep up with increased productivity and its effect on the overall economy, firms may take out credit to invest.
However, it is possible a lag of one year may be too short of a time period for the better infrastructure to have a significant effect on the enhanced productivity. Especially given the fact prior research found the positive short-term effects of infrastructure spending on GDP increases are mainly because of the increased funds and labor demand and not from an increase in productivity. Thus, it remains debatable whether it is indeed the increased productivity in a period of one year that leads to more lending, even more so considering the same year or later lagged years do not show any effect.
4.2
Effect of increased infrastructure on credit in each sector
Table 6: Impact of total works completed on credit in each sector (part 1).
This table reports the results of the regression of log total works completed in the previous year on log total credit outstanding in a year. The observation level is bank-firm-year with bank data running from 2002 until 2019 and NREGA data from 2006 until 2019. L1 means one lagged year.
Significance of the parameters are indicated as follows: ∗ p < 0.10, ∗∗ p < 0.05, ∗∗∗ p < 0.01. Standard errors are clustered at the district level and are shown in the parentheses.
(1) (2) (3) (4)
Agriculture Industry Transport Professional
services L1. Total Works 0.017*** (0.006) 0.034** (0.015) -0.005 (0.013) -0.008 (0.010) Constant 15.096*** 13.955*** 12.415*** 13.208*** (0.063) (0.145) (0.130) (0.098) District fixed effects
Yes Yes Yes Yes
State by year fixed effects
Yes Yes Yes Yes
13
Table 7: Impact of total works completed on credit in each sector (part 2).
This table reports the results of the regression of log total works completed in the previous year on log total credit outstanding in a year. The observation level is bank-firm-year with bank data running from 2002 until 2019 and NREGA data from 2006 until 2019. L1 means one lagged year.
Significance of the parameters are indicated as follows: ∗ p < 0.10, ∗∗ p < 0.05, ∗∗∗ p < 0.01. Standard errors are clustered at the district level and are shown in the parentheses.
(1) (2) (3)
Personal Lending Trade Finance
L1. Total Works 0.002 (0.005) 0.009 (0.008) 0.053* (0.027) Constant 15.165*** 14.105*** 10.241*** (0.053) (0.074) (0.267) District fixed effects
Yes Yes Yes
State by year fixed effects
Yes Yes Yes
Observations R² 6532 0.98 6532 0.97 6180 0.83
Table 6 and 7 show the effect of the completed works with a lag of one year to credit outstanding in each sector. Here, I have chosen just one lag, as it shows in table 5 total completed works only has an effect with a lag of one year. As can be seen, the effect of total works on lending is insignificant for the trade, personal lending, professional services, and transport sector. Furthermore, R² is high for every model, but slightly lower for the model estimating the effect of works completed on lending in the financial sector, nevertheless, it is indicative of the model being a good fit.
As can be seen in column (1) in table 6, when total completed works go up by 1%, agricultural lending goes up by 0.017%. This implies that after NREGA works have been completed in a given district, agricultural lending increases the next year. The lagged effect of total completed works is also significant for industrial lending, shown in column (2) of table 6. To be more precise, when total completed works went up by 1%, industrial lending went up by 0.034% the next year. This increase is even larger than in the agricultural sector, which is interesting, considering the fact NREGA is focused on the agricultural sector. Lastly, as can be seen in column (3) of table 7, the largest effect of completed NREGA works is seen in the financial sector, as a 1% increase in total works completed leads to an increase of 0.053% increase in lending in the financial sector.
14
focused on the agricultural sector. Next, as the agricultural sector transports goods, the increased infrastructure may also enhance productivity. Moreover, the increase in industrial lending can also be explained by the same fact. Given that the improved infrastructure leads to a better availability of resources and enhances the productivity of it, firms become more profitable and products may become cheaper, increasing demand for its products. To keep up with such increased demand, industrial firms may lend to invest. Lastly, lending in the financial sector is the sector impacted the most by works completed in the previous year. It could be that financial institutions become more interconnected as the economy develops as a whole and that it was lagging behind.
However, one should view these conclusions with caution, because, if the infrastructure does lead to higher productivity in a short time, it seems likely the other sectors would also see an increase in lending, especially the transport sector, as they have to transport more goods. Summarizing the results so far, I find that the workfare program leads to increased lending, either in the same year or the next year, or, when measured as total expenditure, both the same year and next year. More specifically, hypothesis 1 is supported, as increased NREGA activity does lead to increased lending in the whole economy. Moreover, hypothesis 2 is supported by the results, as total completed works lead to increased total credit outstanding in the next year. Lastly, hypothesis 2a is partly supported, as total completed works only leads to increased lending in the agricultural, industrial, and financial sectors in the next year.
One must also look at the significance of the effects. Although, for example, an increase of 1% in households demanded work leads to an increase in 0.026% of total credit outstanding may not seem much, given the small size of the act compared to the overall Indian economy, the influence is noticeable. This is the case for each interpretation.
15
5. Alternative explanations
5.1 Endogeneity problem
Does the act indeed have a significant effect on credit outstanding or are there other factors influencing total credit outstanding? Fortunately, this can be analyzed by employing a difference-in-difference analysis due to the staggered implementation of the program, as explained in section 3.1. In short, in this difference-in-difference test, districts belong to the control group before implementing the act and serve as part of the treatment group after implementing NREGA. This test serves to rule out whether there are missing variables in the error term.
Table 8: Difference-in-difference test for an endogeneity check.
This table reports the results of the difference-in-difference regression to check for endogeneity. The observation level is bank-firm-year with bank data running from 2002 until 2010 and NREGA data from 2006 until 2010. The dummy variable consists of ‘treated’ districts, meaning districts that implemented NREGA. These districts belonged to the control group before the implementation of the act.
Significance of the parameters are indicated as follows: ∗ p < 0.10, ∗∗ p < 0.05, ∗∗∗ p < 0.01. Standard errors are clustered at the district level and are shown in the parentheses.
(1) Total credit outstanding Dummy variable 0.026 (0.022) Constant 15.373*** (0.010) District fixed effects Yes State by year fixed effects Yes Observations R² 5427 0.97
16
5.2 Other effects of the act on lending.
Although the model shows significant results of increased infrastructure leading to increased lending in the agricultural, industrial, and financial sector there may be other effects in place instead of the increased productivity. As noted earlier, infrastructure may have a positive influence on GDP in the short-term due to the extra labor demand for projects. This is more in line with the view of increased funds. However, there could also be alternative explanations and they will be discussed below.
When looking at lending in each industry, agricultural lending may also have increased not due to increased profitability due to infrastructure or an increase in funds, but simply because lenders are more credible due to the assets created on their land (see Sambodhi Research Group (2013)). Increased lending in the industrial sector may be explained due to firms lending to invest in machines as they resort to increased mechanization due to the adverse labor supply of NREGA (see Agarwal et al. (2020)). Lastly, the increased lending in the financial sector may be explained due to the increased demand for financial institutions, implicitly leading to more lending overall in the sector, not per se lending per institution, as many NREGA workers did not have a bank account before the start of the program, but have one now (see Carswell and De Neve (2014)).
6 Conclusion
In this study I analyzed the effect of the largest workfare program on lending by using NREGA data from its official website and bank lending data from the official website of the Reserve Bank of India. I document that NREGA activity has a positive significant effect on overall lending in India, most likely due to local scale economies.
I also found that improved infrastructure by NREGA workers leads to higher lending in the overall economy. Given the fact that increased lending is in general associated with higher economic growth, it is in line with existing literature, that show that especially in India, improved infrastructure seems crucial for economic growth. More specifically, these effects were significant in the agricultural, industrial and financial sectors, and not in the trade, personal lending, transport, or in the professional services sector.
17
improvement of the financial system, which is fruitful, especially for emerging countries, as such countries are more likely to have an underdeveloped financial system.
18
7. References
Agarwal, Sumit, Shashwat Alo, Yakshup Chopr and Prasanna Tantri, 2020, Government employment guarantee, labor supply and firms’ reaction: Evidence from the largest public workfare program in the world, Journal of Financial and Quantitative Analysis, 1-34. Aggarwal, Ankita, Aashish Gupta and Ankit Kumar, 2012, Evaluation of NREGA wells in Jharkhand, Economic and Political Weekly, 24-27.
Azam, Mehtabul, 2012, The Impact of Indian Job Guarantee Scheme on Labor Market Outcomes: Evidence from a Natural Experiment, IZA Discussion Series 6548.
Berg, Erlend, Sambit Bhattacharyya, D. Rajasekha and R. Manjula, 2012, Can Rural Public Works Affect Agricultural Wages? Evidence from India, CSAE Working Paper 05.
Bertrand, Marianne and Sendhil Mullainathan, 1998, Executive Compensation and Incentives: The Impact of Takeover Legislation, NBER Working Paper w6830.
Bivens, Josh, 2014, The short- and Long-Term Impact of Infrastructure Investments on Employmend and Economic Activity in the U.S. Economy, Economic Policy Institute, Briefing Paper #374
Carswell, Grace, and Geert de Neve, 2014, MGNREGA in Tamil Nadu: A Story of Success and Transformation?, Journal of Agrarian Change 14(4), 564-585
Caporale, Guglielmo, Stefano Di Colli, Roberto Di Salvo, Juan Sergio Lopez, 2014, Local banking and local economic growth in Italy: some panel evidence, DIW Berlin, Discussion Paper 1409.
Deininger, Klaus and Yanyan Liu, 2013, Welfare and Poverty Impacts of India's National Rural Employment Guarantee Scheme: Evidence from Andhra Pradesh, International Food Policy Research Institute (IFPRI) Discussion Paper 01289.
Gehrke, Esther, 2019, An Employment Guarantee as Risk Insurance? Assessing the Effects of the NREGS on Agricultural Production Decisions, The World Bank Economic Review 33(2), 413–435.
Imbert, Clement and John Papp, 2015, Labor Market Effects of Social Programs: Evidence from India’s Employment Guarantee, American Economic Journal: Applied Economics 7(2), 233–263.
19
Krugman, Paul, 1991, Increasing Returns and Economic Geography, Journal of Political Economy 99(3), 483–499.
Martynova, Natalya, 2015, Effect of bank capital requirements on economic growth: a survey, DNB Working Paper No. 467
.
Mishkin, Frederic, 2004, The Economics of Money, Banking and Financial Markets, 7th Edition, Columbia University, pp 188
Phang, S. (2003). Strategic development of airport and rail infrastructure: the case of Singapore. Transport Policy 10, 27-33
Ravi, Shamika and Monika Engler, 2015, Workfare as an Effective Way to Fight Poverty: The Case of India's NREGA, World Development 7, 57-71
Rudra P. Pradhan, Tapan P. Bagchi, 2013, Effect of transportation infrastructure on economic growth in India: The VECM approach, Research in Transportation Economics 38(1), 139-148,
Sambodhi Research Group, 2013, Impact Assessment of assets Created on Individual Land under Mahatma Gandhi National Rural Employment Guarantee Act. A Report submitted to UNDP
