# I. INTRODUCTION In the financial system, non-bank financial institutions (NBFIs) play a crucial dual role. In the service of the commercial banks, which have some gaps, they supplement their functions. However, they also put commercial banks in a position of competition, which makes them more responsive to client demands and efficient. Non-Bank financial organizations, finance businesses and consumers, investment banks, and others that deal with pensions and mutual funds exist in relatively mature notions. NBFIs provide advances and loans for a commercial organizations, farming, housing, and real estate. They also underwrite or purchase securities and lease financing, buy and sell securities, invest in shares, stocks, bonds, or debentures, and reinvest in those securities. There are now 35 NBFI's, the first of which was opened in 1981. Out of the total number of NBFIs, 19 were established by private domestic enterprises, 13 through joint ventures, 2 were entirely under government supervision, and one is a branch of the state-owned commercial bank (SOCB). There were 277 branches of the 24 NBFIs in FY 2021, with assets totaling BDT 914.3 billion. Bangladesh has achieved remarkable progress in financial diversification, asset growth, NBFI numbers, area coverage, and account holder numbers. Despite significant progress, NBFI's face several challenges, including rising lending rates, rising NPL levels, higher interest spreads, credit concentration in trading and business, and reduced investment in the social sector. The evaluation of NBFI's performance would enable them to carry out their responsibilities while also contributing to improving of NBFI's performance in Bangladesh. Even though several performance studies of the performance of NBFI's have been conducted in advanced industrial, emerging, and developing countries, Bangladesh has seen only a small number of these studies. A non-parametric data envelopment analysis (DEA) method is used in this work to empirically investigate the performance of Bangladeshi NBFI's in terms of technical, pure technical, and scale efficiency from 2016 to 2021. The article has two goals: first, it examines the efficiency performance of Bangladeshi NBFI's from 2016 to 2021, and second, it offers policy recommendations for improving NBFI performance in Bangladesh. The remainder of the paper has the following structure: The second section covered a literature review. The third section of this paper focuses on methodology, while the fourth section looks at the results of overall technical, pure technical, and scale efficiency findings. The fifth section illustrates the comparative effectiveness of NBFIs, while the sixth section describes policy options. The seventh section focuses on the conclusions and recommendations. # II. LITERATURE REVIEW In Bangladesh, conventional financial institutions like banks have been the subject of numerous efficiency studies, but there is a shortage of information on non-banking businesses. There are still gaps in the efficiency study of non-bank institutions in Bangladesh. Due to the importance of studies, we evaluated the efficiency of non-bank institutions in Bangladesh in this study. The paper "Performance Analysis of non-banking finance companies using two-stage data envelopment analysis" by Dutta, Jain, and Gupta (2020) aims to evaluate the performance of non-banking finance companies (NBFCs) in India using two-stage data envelopment analysis (DEA). The authors have collected financial data on NBFCs from 2011-2019. In the first stage, the study examines the input/output efficiency of the NBFCs using traditional DEA models. In the second stage, the study explores the impact of environmental factors on the efficiency scores of NBFCs using a Tobit regression model. The authors' findings reveal that the average efficiency score of the NBFCs is relatively low, indicating significant inefficiencies in their operations. Further analysis using Tobit regression shows that macroeconomic factors such as inflation, GDP growth, and market concentration significantly affect the efficiency of NBFCs. A data envelopment analysis" by Sharma, Rastogi, and Gupta (2020) examines the financial efficiency of Non-Banking Financial Companies-Microfinance Institutions (NBFC-MFIs) in India using data envelopment analysis (DEA). The authors collected financial data of NBFC-MFIs from the Indian Microfinance Pulse database for 2014-2018. The study aims to evaluate the technical efficiency, pure technical efficiency, and scale efficiency of the NBFC-MFIs. The authors' findings reveal that the average technical efficiency score of NBFC-MFIs is relatively low, indicating significant inefficiencies in their operations. Further analysis shows that the pure technical efficiency score is lower than the technical efficiency score, meaning that there is scope for improvement in managerial and operational practices. The authors also found that the scale efficiency of NBFC-MFIs is high, suggesting that they are operating at an optimal scale. In "The efficiency of non-bank financial institutions: empirical evidence from Malaysia" (2006), Sufian investigates the technical efficiency of non-bank financial institutions (NBFIs) in Malaysia using data envelopment analysis (DEA). The study uses panel data covering 1998-2002 and includes 17 NBFIs. The results reveal that the average technical efficiency of NBFIs in Malaysia is low, indicating significant inefficiencies in their operations. The study also finds that small NBFIs are more efficient than larger ones, and the efficiency of NBFIs is positively related to their profitability. Moreover, the study finds that NBFIs that are more specialized in their operations tend to be more efficient than those that are more diversified. The study concludes that the Malaysian NBFIs' overall efficiency can be improved by increasing their scale of operations, adopting specialized functions, and improving their managerial and operational practices. The profitability of businesses in Bangladesh's non-banking financial institutions (NBFIs) from 2005 to 2014 is examined by Mazumder, M. A. (2015). The findings show that profitability indicators affect net profit, but total assets, total equity, and operating income have a discernible influence on the profitability of Bangladesh's non-banking sector. Total assets are one of the most straightforward metrics for assessing the financial soundness of financial organizations. Almost all independent and dependent variables have strong positive associations, except operating costs. Except for total liabilities, term London Journal of Research in Management and Business 2 deposits, and operating expenses, almost every element has a positive effect. Debnath, G. C., Rahman, S. N., & Akhter, S. (2011) compares and analyze the liquidity positions of a few chosen non-bank financial organizations in Bangladesh from 2011 to 2015. The analysis considered the five NBFIs' places in terms of liquidity over the short and long terms according to maturity. We concluded from the complete investigation that all the chosen financial institutions have a positive and improving overall liquidity situation. However, the rate of liquidity expansion varies. Based on analysis, a significant portion of the company's short term liquidity is negative. # III. DEA Methodology Data Envelopment Analysis was introduced in 1978 by Charnes, Cooper, and Rhodes. (DEA). The DEA maintains that returns to scale are constant. DEA is a non-parametric linear programming model that seeks to improve the effectiveness of each decision-making unit by optimizing its weighted output/input ratio (DMU). The efficacy of various input and output orientations was evaluated. According to Banker et al.'s assumptions, Scale efficiency (SE) and pure technological efficiency (PTE) are the outcomes of two components. The DEA gives various weights to the input and output of companies to maximize efficiency in contrast to other companies. Each unit is given a score, with one being the most effective and ranging from zero to one. The CCR model assumes the production function has constant returns-to-scale (CRS). The CCR model's objective score is technical efficiency (TE). Consider that two DMUs require evaluation. Like DMU r , which requires X ir quantities of input and generates X volumes of output, each requires a different amount of input and yields a distinct volume of output. It is necessary that each DMU's have at least one positive input and output value, and it is anticipated that none of these values will be negative. The CCR model aims to maximize the weighted output to the weighted input ratio for the NBFIs under consideration. The objective function is maximized for NBFIs under the restriction that no other NBFIs in the sample may attain unit efficiency by utilizing the same weights. As a result, the objective function is: Here, j = j th output, j = 1,...,l; i = i th input, i = 1,...., k; r = 1, a r = an objective measure of efficiency for r th ; Y jr is the amount of output, X ir is the amount of input, The input weight is V i , the number of NBFI's is S, the number of outcomes is l, and the number of inputs is k. # The CRS Model By limiting the denominator of the target function to unity, the least issue can be simplified to a linear program. Thus, linear programming takes the following structure: # Data and Variables The production strategy and the intermediate approach are widely used to choose the input and output and compute various efficiency scores in scenarios mentioned in the literature. Technical inefficiency is 21 percent in BD Finance and 31 percent in National Housing, based on technical efficiency of 0.79 and 0.69 for BD Finance and National Housing, respectively. Technical inefficiency is 17% and 9%, respectively, for BD Finance PTE and SE, which are 0.83 and 0.91, respectively. The national housing PTE and SE scores are simultaneously 0.86 and 0.77, indicating that 23% of inefficiency is due to scale inefficiency and 14% is due to pure technical inefficiency. Regarding efficiency scores, United Capital and Lanka Bangla Finance are ranked Tenth and Twelfth, respectively. United Capital's and Lanka Bangla's technical efficiencies of 0.65 and 0.66, respectively, represent technological inefficiencies of 34% and 35%. When the scale efficiency of two companies is compared, Lanka Bangla has an 8% inefficiency and United Capital has a 20% inefficiency. # Returns to Scale (RTS) of Non-bank Financial Institution Among the NBFIs in the sample, ICB, GSP, IDCOL, and Fareast Finance exhibit continuous returns to scale (CRS) across the 2016-2021 study period, indicating that they are operating at maximum efficiency. It is impossible to adjust the production scale without diminishing effectiveness. The fact that IDLC runs at its optimal or most productive scale in 2020 and 2021 suggests that scale reduction might have enhanced efficiency during these periods. The fact that Lankabangla shows DRS from 2016 to 2019 and IRS from 2020 to 2021 suggests that it may have adjusted its production scale to be more productive from 2016 # V. COMPARISON OF THE EFFICIENCY OF NBFIS The graph reveals that in 2016, the overall technical efficiency average was 0.95, which representing 5% inefficiency. Efficiency scores improved at a constant rate of 0.97 from 2017 to 2019, showing 3% inefficiency. In the years 2020 and 2021, the technical efficiency score dropped to 0.57 and 0.54, respectively, reflecting inefficiencies of 43% and 46%. This indicates that the COVID-19 pandemic's effects prevented NBFIs from using their creative management abilities to manage the organization's resources and maintain production. # VI. POLICY OPTIONS The study on the efficiency of NBFIs suggests that those with scale inefficiencies, rather than pure technical inefficiency, should focus on enhancing their management performance to enhance their technical efficiency. According to the NBFIs, to improve technological efficiency, DRS must either reduce output or diversify its product line. IRS, on the other hand, demonstrates a need to increase production capacity. As a result, while overseeing a variety of financial items, the company's management verified 100% technological efficiency. # VII. CONCLUSION The contribution of non-banking financial institutions (NBFIs) in Bangladesh is crucial for economic growth. Over time, NBFIs' influence on Bangladesh's financial industry has increased along with that of the conventional banking sector. The challenge posed by NBFIs to the traditional banking sector is growing. NBFIs have remarkably aided financial inclusion and made progress in closing the credit gap for retail consumers in Bangladesh's underserved and unbanked areas. NBFIs play a significant role in delivering a range of consumer services and bridging the financial services supply and demand gap for those needing loans. Our metrics evaluate the operational effectiveness and scale economies of 17 NBFIs operating in Bangladesh. This analysis reveals that only ICB, GSP, IDCOL, and Fareast Finance received scores of one, indicating excellent efficiency in the three categories of technical efficiency, pure technical efficiency, and scale efficiency. Throughout the period, Lanka-Bangla Finance, United Finance, and National Housing have all consistently been inefficient NBFIs. The findings make it evident that the technical inefficiency in Bangladesh's non-bank financial institution industry is a result of poor input utilization, including managerial and scale inefficiencies, as well as inability to work at their maximum capabilities. Policymakers should take measures to increase the size and effectiveness of the non-bank financial sector to expand the financial sector of Bangladesh. Regulatory agencies should take extra precautions in order to improve efficiency through economies of scale. Future research may assess how information technology (IT) impacts NBFI performance in Bangladesh. 1![2?1, I=1, 2?. K and r= 1, 2?...s. For the NBFI's r, The Linear programming method above yields an efficiency score of (a r ); where, .0 ? ?? ? 1](image-2.png "J = 1 ,") ![to 2019 and 2020 to 2021. Phoenix exhibits DRS in 2016 and 2018, CRS in 2017 and 2019, and IRS in 2020 to 2021, which may indicate that it increased production size in 2020 to 2021 while reducing scale in 2016 and 2018 to increase efficiency. It will need to change the manufacturing size between 2017 and 2019. For CRS in the next five years (2016, 2017, 2018, 2019, 2020) and for IRS in 2021, Prime Finance indicates the ideal or most productive size of production, indicating it could have extended scale by improving efficiency. DBH might have increased its production scale from 2020 to 2021 and decreased scale in 2018 for efficiency; it did not need to modify scale in 2026, 2017, or 2019.](image-3.png "") ![IPDC shows Drs from 2016 to 2019 and Irs from 2020 to 2021, which indicates that production size may have increased from 2020 to 2021 while declining from 2016 to 2019. The fact that United Finance had Irs for all years has revealed that it should have raised production to improve efficiency. The Union Capital may have raised its production size from 2019 to 2021 in order to maximize efficiency after working at its optimum from 2016 to 2018. In addition to Irs, United Capital also displays Crs from 2016 to 2018. National Housing exhibits Irs all year long, suggesting that it may need to increase manufacturing all year long to increase productivity. The manufacturing scale was probably enlarged for all years before to 2016, as Midas Finance reported Drs for 2016 and Irs for all preceding years. To demonstrate that they consistently outperform the market, First Finance and BD Finance provide their respective Irs from 2020 to 2021 and Crs from 2016 to 2019. They had the potential to generate more during these two periods. Except for 2020, all years in Islamic Finance are stated in pennies. It operates at its highest level throughout the year, except 2020. The efficiency might have improved as a result of production expansion.](image-4.png "") 1considered output variables. For each variable,millions of Bangladeshi Taka are used as ameasurement. We used balance panel data fromtheir annual reports for 2016-2021. The DEA wasBerger, A.computed and applied using Stata 14.According to N. & Humphrey, D. B. (1997), bothIV. RESULTS AND FINDINGSstrategies are inefficient since they disregardmultiple roles. Intervention of resources andThe study's results are presented in Table 1,inputs like labor and capital, Many authors, likeshowing the technical efficiency (TE), pureSathye, M. (2001), Neal, P. (2004), and others,technical efficiency (PTE), and scale efficiencyemploy the production approach. Mokhtar, et al.(SE) scores for the 17 non-bank financial(2008) and Bhattacharya, et al. (2013) employ aninstitutions sampled from 2016 to 2021. Notably,intermediation strategy. Most of the empiricalthe results indicate that only four NBFIs, namelystudies follow the intermediation method, whichICB, GSP, IDCOL, and Fareast Finance, achieveduses input and output variables to calculate theperfect scores of 1 in all three categories,numerous efficiency results for different NBFI's.suggesting that they are efficiently managing theirresources and production scale. However, it isLondon Journal of Research in Management and BusinessThe current study has chosen 17 non-bank financial companies from Bangladesh for 2016-2021. Total deposits, fixed assets, and operating expenses are considered input variables. Total loans, investments, and operating revenue are NBFIs Efficiency 2016 2017 2018 worth noting that IDLC showed a scale inefficiency of 23 percent implying that its production scale is suboptimal despite achieving full efficiency in terms of pure technical efficiency. 2019 2020 2021 Mean Inefficiency (%) ICB TE 1 1 1 1 1 1 1 0 PTE 1 1 1 1 1 1 1 0 SE 1 1 1 1 1 1 1 0 IDLC TE 1 1 1 1 0.37 0.25 0.77 0.23 PTE 1 1 1 1 1 1 1 0 SE 1 1 1 1 0.37 0.25 0.77 0.23 GSP TE 1 1 1 1 1 1 1 0 PTE 1 1 1 1 1 1 1 0 SE 1 1 1 1 1 1 1 0 Lanka bangla TE 0.79 0.92 0.91 0.90 0.22 0.19 0.66 0.34 PTE 0.99 1 1 1 0.22 0.19 0.73 0.27 SE 0.81 0.92 0.91 0.90 1 1 0.92 0.08TE0.9810.9410.350.390.780.22PhoenixPTE11110.350.380.790.21SE0.9810.941110.990.01TE111110.750.960.04Prime financePTE111110.930.990.01SE111110.810.970.03TE110.9910.570.410.830.17DBHPTE11110.600.440.840.16SE110.9910.950.940.980.02TE0.810.9610.930.340.280.720.28IPDCPTE0.88110.970.340.280.740.26SE0.920.9610.960.9910.970.03Source: The Author's calculations4| Volume 23 Issue 3 ?"? Compilation 1.0 |© 2023 Great ] Britain Journals Press 2London Journal of Research in Management and BusinessRTS ICB IDLC2016 Crs Crs2017 Crs Crs2018 Crs Crs2019 Crs Crs2020 Crs Drs2021 Crs DrsGSPCrsCrsCrsCrsCrsCrsLanka BanglaDrsDrsDrsDrsIrsIrsPhoenixDrsCrsDrsCrsIrsIrsPrime FinanceCrsCrsCrsCrsCrsIrsDBHDrsCrsDrsCrsIrsIrsIPDCIrsDrsDrsDrsDrsIrsUnited FinanceIrsIrsIrsIrsIrsIrsUnion CapitalCrsCrsCrsIrsIrsIrsIDCOLCrsCrsCrsCrsCrsCrsNational HousingIrsIrsIrsIrsIrsIrsMidas FinanceDrsIrsIrsIrsIrsIrsFirst FinanceCrsCrsCrsCrsIrsIrsBD FinanceCrsCrsCrsCrsIrsIrs6| Volume 23 Issue 3 ?"? Compilation 1.0 |© 2023 Great ] Britain Journals Press Measuring Efficiency of Non-Bank Financial Institutions in Bangladesh: A Non-Parametric Data Envelopment Approach * Efficiency Analysis of Microfinance Institutions in Pakistan. Munich Personal RePEc Archive UAhmad 2011 * Constraints and Challenges of Financial Institutions in Bangladesh: A Possible Way Out TAlif Journal of Management, Economics, and Industrial Organization 5 1 2021 * Efficiency of microfinance institutions in the Mediterranean: an application of DEA BSBassem Transition Studies Review 15 2 2008 * Efficiency of financial institutions: International survey and directions for future research ANBerger DBHumphrey European journal of operational research 98 2 1997 * Does earnings quality affect information asymmetry? Evidence from trading costs NBhattacharya HDesai KVenkataraman Contemporary Accounting Research 30 2 2013 * Measuring the efficiency of decision-making units ACharnes WWCooper ERhodes European journal of operational research 2 6 1978 * Market risk and financial performance of listed non-bank financial institutions in Kenya BPChepkemoi SKanini JKahuthia International Academic Journal of Economics and Finance 3 3 2019 * GCDebnath SNRahman SAkhter An Analysis of Liquidity Position of Non-Bank Financial Institutions 2011 * Performance analysis of non-banking finance PDutta AJain AGupta 2020