nach oben

Journal of Financial Services Research

12.09.2023

What Determines Enterprise Borrowing from Self Help Groups? An Interpretable Supervised Machine Learning Approach

verfasst von: Madhura Dasgupta, Samarth Gupta

Erschienen in: Journal of Financial Services Research

Einloggen

Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.

search-config

KI-gestützte Suche

Aus

Abstract

Despite several advantages associated with borrowing from micro-finance institutions, such as self-help groups (SHGs), many enterprises in developing countries continue to rely on informal lenders. Using machine learning techniques on a novel village-enterprise matched dataset from India, we predict an enterprise’s choice of credit source as a function of three key mechanisms: supply-side factors, infrastructural facilities and socio-demographic characteristics. Proximity to markets and social norms of the village, proxied by high literacy rates and sex ratios, play important roles in credit uptake from SHGs. However, the absence of financial access points, such as commercial or cooperative bank branches, is not prohibitive.

Sie haben noch keine Lizenz? Dann Informieren Sie sich jetzt über unsere Produkte:

Springer Professional "Wirtschaft"

Online-Abonnement

Mit Springer Professional "Wirtschaft" erhalten Sie Zugriff auf:

über 67.000 Bücher
über 340 Zeitschriften

aus folgenden Fachgebieten:

Bauwesen + Immobilien
Business IT + Informatik
Finance + Banking
Management + Führung
Marketing + Vertrieb
Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

Jetzt informieren

Springer Professional "Wirtschaft+Technik"

Online-Abonnement

Mit Springer Professional "Wirtschaft+Technik" erhalten Sie Zugriff auf:

über 102.000 Bücher
über 537 Zeitschriften

aus folgenden Fachgebieten:

Automobil + Motoren
Bauwesen + Immobilien
Business IT + Informatik
Elektrotechnik + Elektronik
Energie + Nachhaltigkeit
Finance + Banking
Management + Führung
Marketing + Vertrieb
Maschinenbau + Werkstoffe
Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

Jetzt informieren

Nur mit Berechtigung zugänglich

For usage of ML methods in financial markets see Tian et al. (2015), Albanesi and Vamossy (2019), Freyberger et al. (2020) and Bjorkegren and Grissen (2020). For India, in particular, see Chernozhukov et al. (2018) for an application of the impact evaluation of interventions to increase immunization.

AUC stands for area under the receiver operating characteristic (ROC) curve and is a commonly used metric of evaluation for ML methods. It quantifies a model’s ability to correctly segregate observations into two classes (SHG and moneylenders). See Fawcett (2006) for more details.

Measuring the average change in predicted probabilities will provide the slope or the first derivative of the variable. However, such a measure would be statistically biased.

Economic Censuses list all economic enterprises with the following exceptions: establishments classified under 011 and 012 of Section A of NIC 2008 (farm enterprises), Section O of NIC 2008 (public administration, defense, compulsory Social Security), Section T of NIC 2008 (territorial organization and bodies) and Section R of NIC 2008 (illegal gambling and betting activities).

A source of financing is defined as major if the enterprise primarily seeks loans from that source. The Economic Census does not record the amount of credit taken. AIDIS 2012-13, which does record the amount of credit by source, indicates little or non-existent overlap between these two sources of financing for enterprises.

In addition to these ML models, we also run a standard logit model.

The predictions are considered correct if the model can correctly distinguishes between enterprises taking loans from SHG and moneylenders.

AUC takes a value from 0 to 1 representing the degree of separability between the two classes.

See Albanesi and Vamossy (2019) for an application of this method on detecting default behavior among borrowers.

In addition to the degenerate counterfactual, we derive a non-degenerate counterfactual dataset, where villages are randomly allotted binary characteristics. For instance, if 80% of villages have roads in the original dataset, then in the counterfactual data, we randomly assign road, a value 1 to 80% of the remaining 20% of villages. The results are qualitatively similar.

We restrict literacy and sex ratio to a maximum of 100% in the counterfactual sets.

See Bjorkegren and Grissen (2020), for a similar demand estimation exercise.

The Appendix provides details on how we computed the index of demand from self-financed firms using our model and actual uptake as per AIDIS 2019. We also discuss some caveats in this comparison.

Agarwal S (2018) Do rural roads create pathways out of poverty? Evidence from India. J Dev Econ 133(February 2017):375–395CrossRef

Agarwal S, Hauswald R (2010) Distance and private information in lending. Rev Financ Stud 23(7):2758–2788CrossRef

Agarwal S, Alok S, Ghosh P, Ghosh S, Piskorski T, Seru A (2017) Banking the unbanked: what do 255 million new bank accounts reveal about financial access? Columbia Business School Research Paper, (17–12)

Albanesi S, Vamossy DF (2019) Predicting consumer default: A deep learning approach. No. w26165. National Bureau of Economic Research

Altmann A, Toloi L, Sander O, Lengauer T (2010) Permutation importance: a corrected feature importance measure. Bioinformatics 26(10):1340–1347CrossRef

Asher S, Lunt T, Matsuura R, Novosad P (2019) The Socioeconomic High- resolution Rural-Urban Geographic Dataset on India (SHRUG)

Asher S, Novosad P (2020) Rural roads and local economic development. Am Econ Rev 110(3):797–823CrossRef

Banerjee A, Duflo E, Glennerster R, Kinnan C (2015) The miracle of micro finance? Evidence from a randomized evaluation. Am Econ J: Appl Econ 7(1):22–53

Banerjee A, et al (2013) The diffusion of microfinance. Science 341.6144: 1236498

Barnes D (2019) Electric power for rural growth: how electricity affects rural life in developing countries. RoutledgeCrossRef

Basu P (2006) Improving access to finance for India’s rural poor. World Bank PublicationsCrossRef

Bell C (1990) Interactions between institutional and informal credit agencies in rural India. The World Bank Economic Review, pages 186–213

Besley T, Coate S (1995) Group lending, repayment incentives and social collateral. J Dev Econ 46(1):1–18CrossRef

Bruhn M, Love I (2011) Gender differences in the impact of banking services: evidence from Mexico. Small Bus Econ 37(4):493–512CrossRef

Bjorkegren D, Grissen D (2020) Behavior revealed in mobile phone usage predicts loan repayment. World Bank Econ Rev 34.3(2020):618–634CrossRef

Bruhn M, Love I (2014) The real impact of improved access to finance: evidence from Mexico. J Financ 69(3):1347–1376CrossRef

Burgess BR, Pande R (2005) Do rural banks matter? Evidence from the Indian social banking experiment. Am Econ Rev 95(3):780–795CrossRef

Chaudhuri K, Sasidharan S, Raj RSN (2020) Gender, small firm ownership, and credit access: some insights from India. Small Bus Econ 54(4):1165–1181CrossRef

Chen T, Guestrin C (2016) XGBoost: A scalable tree boosting system. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 13–17-Augu:785 794

Chernozhukov V, Demirer M, Duflo E, Fernandez-Val I (2018) Generic Machine Learning Inference on Heterogeneous Treatment Effects in Randomized Experiments, with an Application to Immunization in India (No. w24678). National Bureau of Economic Research

Cull R, Demirgüç-Kunt A, Morduch J (2014) Banks and microbanks. J Financ Serv Res 46:1–53CrossRef

Davis J, Heller S (2017) Using causal forests to predict treatment heterogeneity: an application to summer jobs. Am Econ Rev 107(5):546–550CrossRef

Davis J, Heller S (2020) Rethinking the benefits of youth employment programs: the heterogeneous effects of summer jobs. Rev Econ Stat 102(4):664–677CrossRef

de Aghion BA, Morduch J (2006) The economics of microfinance. J Int Econ 70(1):328–333CrossRef

Demiguc-Kunt A, Klapper L, Singer D, Ansar S, Hess J (2017) The Global Findex

Du M, Ninghao L, Hu X (2019) Techniques for interpretable machine learning. Commun ACM 63(1):68–77CrossRef

Fawcett T (2006) An introduction to ROC analysis. Pattern Recogn Lett 27(8):861–874CrossRef

Fletschner D (2009) Rural women’s access to credit: market imperfections and intra- household dynamics. World Dev 37(3):618–631CrossRef

Freyberger J, Neuhierl A, Weber M (2020) Dissecting characteristics nonparametrically. Rev Financ Stud 33(5):2326–2377CrossRef

Friedman J (2001) Greedy function approximation: a gradient boosting machine. Ann Stat 29(5):1189–1232CrossRef

Ghatak M (2000) Screening by the company you keep: joint liability lending and the peer selection effect. Econ J 110(465):601–631CrossRef

Ghatak M, Guinnane T (1999) The economics of lending with joint liability: theory and practice. J Dev Econ 60(May):195–228CrossRef

Ghosh S, Vinod D (2017) What constrains financial inclusion for women? Evidence from Indian micro data. World Dev 92(60):81

Hastie T, Tibshirani R, Friedman JH, Friedman JH (2009) The elements of statistical learning: data mining, inference, and prediction. Springer, New York, vol. 2, pp. 1–758

Hoerl AE, Kennard RW (1970) Ridge regression: Biased estimation for non-orthogonal problems. Technometrics 42(1):80-86.CrossRef

Hoff K (2016) Caste system. World Bank policy research working paper, (7929)

Kaboski JP, Townsend RM (2012) The impact of credit on village economies. Am Econ J Appl Econ 4(2):98–133CrossRef

Karlan DS (2007) Social connections and group banking. Econ J 117(52):84

Kirubi C, Jacobson A, Kammen DM, Mills A (2009) Community-based electric micro-grids can contribute to rural development: evidence from Kenya. World Dev 37.7(2009):1208–1221CrossRef

Kleinberg J, Ludwig J, Mullainathan S, Obermeyer Z (2015) Prediction policy problems. Am Econ Rev 105(5):491–495CrossRef

Kochar A (1997) Does lack of access to formal credit constrain agricultural production? Evidence from the land tenancy market in rural India. Am J Agric Econ 79(3):754–763CrossRef

Krishna A, Bajpai D (2011) Lineal spread and radial dissipation: experiencing growth in rural India, 1993–2005. Econ Pol Wkly 46(38):1993–2005

Menon N, van der Meulen RY (2011) How access to credit affects self-employment: differences by gender during India’s rural banking reform. J Dev Stud 47(1):48–69CrossRef

Mothilal R, Sharma A, Tan C (2020) "Explaining machine learning classifiers through diverse counterfactual explanations." Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency

Mullainathan S, Spiess J (2017) Machine learning: an applied econometric Ap- proach. J Econ Perspect 31(2):87–106CrossRef

Petersen M, Rajan R (1994) The benefits of lending relationships: evidence from small business data. J Financ 49(1):3–37CrossRef

Rudin C (2019) Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead. Nat Mach Intell 1(5):206–215CrossRef

Schapire R (2013) Explaining adaboost. Empirical Inference, 37–52

Swain R (2002) Credit rationing in rural India. J Econ Dev 27(2):1–20

Swain R, Wallentin FY (2009) Does micro finance empower women? Evidence from self-help groups in India. Int Rev Appl Econ 23(5):541–556CrossRef

Tian S, Yu Y, Guo H (2015) Variable selection and corporate bankruptcy forecasts. J Bank Financ 52(1):89–100CrossRef

Tibshirani R (1996) Regression shrinkage and selection via the Lasso. J R Stat Soc Ser B (Methodological) 58(1):267–288

Townsend R (1997) Discrimination in Financial Services: How Should We Proceed? In: Benston GJ, Hunter WC, Kaufman GG (eds) Discrimination in Financial Services. Springer, Boston

Wachter S, Mittelstadt B, Russell C (2017) Counterfactual explanations without opening the black box: automated decisions and the GDPR. Harv J Law Technol 31(841):1–52

Zou H, Hastie T (2005) Regularization and variable selection via the elastic net. J R Stat Soc Ser B: Stat Methodol 67(5):768CrossRef

Titel: What Determines Enterprise Borrowing from Self Help Groups? An Interpretable Supervised Machine Learning Approach
verfasst von: Madhura Dasgupta
Samarth Gupta
Publikationsdatum: 12.09.2023
Verlag: Springer US
Erschienen in: Journal of Financial Services Research
Print ISSN: 0920-8550
Elektronische ISSN: 1573-0735
DOI: https://doi.org/10.1007/s10693-023-00416-4