Version 03

October 15, 2020

ACCESS TO FINANCE FOR SME &

IMPACT ON JOB CREATION: EMPIRICAL
EVIDENCE BASED ON IDLC FINANCE LTD

Ahsan Mansur and Bazlul H Khondker

POLICY RESEARCH INSTITUTE OF BANGLADESH
 Table of Contents
List of Tables .................................................................................................................................................. i
List of Figures ................................................................................................................................................ ii
List of Boxes .................................................................................................................................................. ii
1 Introduction and Background ............................................................................................................... 1
2 Structure of the SME Sector in Bangladesh .......................................................................................... 3
3 IDLC Employment Survey: Methodology of Sample Distribution ......................................................... 9
4 Profile of Sample Respondents ........................................................................................................... 13
5 Impact on Employment....................................................................................................................... 20
6 Purpose of Taking IDLC Loan ............................................................................................................... 27
7 Concluding Observations .................................................................................................................... 29
Annex 1: ...................................................................................................................................................... 32
Annex 2: LFS – employment growth by sector and self-employment category ......................................... 35

List of Tables
Table 1: Types of Enterprises in Bangladesh............................................................................................. 4
Table 2: Bangladesh--Total Nonfarm Employment, 2003-2013 ............................................................... 4
Table 3: Distribution of Enterprises by Types and Employment Size, 2013 ............................................. 5
Table 4: Distribution of Enterprises by Major Economic Activities 2013 (numbers in 000) ..................... 5
Table 5: Sample Distribution by Region, Sector and Sex of the Entrepreneurs ..................................... 12
Table 6 : Ownership Structure of Firms .................................................................................................. 14
Table 7: Sources of funding at the time of business start-up by the number of business enterprise ... 15
Table 8: Total and Average Investment by Types of Business ................................................................ 15
Table 9: Average Annual Turnover (Million BDT )................................................................................... 16
Table 10: Total Annual Turnover (Billion BDT) ........................................................................................ 16
Table 11: Firms Responding to Questions About Their Need for Credit ................................................ 16
Table 12: Preferred Source of Credit for the Sample Firms (Q 17) ......................................................... 17
Table 13: Percent of enterprises received IDLC’s loan between 2015 and 2019 ................................... 19
Table 14: Growth of Manufacturing Enterprises between 2012 and 2019 ............................................ 19
Table 15: Employment in Manufacturing Enterprises Between 2012 and 2019 .................................... 20
Table 16: Employment in Manufacturing Enterprises Between 2012 and 2019 .................................... 22

i
 Table 17: No of times IDLC loan received by type of establishments..................................................... 28
Table 18: Sources of funding at the time of business start-up ............................................................... 29

List of Figures
Figure 1: Distribution of Employment (in Percent) ................................................................................... 6
Figure 2: Average Labor Productivity 2013 (000 Taka) ............................................................................. 6
Figure 3: Distribution of IDLC’s Total Borrowing Clients by Major Categories (%) ................................. 10
Figure 4: Revised Shares for Sampling Distribution (in percent) ........................................................... 11
Figure 5: Division of Tasks Between the IDLC and PRI Staff as Per the TOR ........................................... 12
Figure 6: Distribution of sample enterprises by residence ..................................................................... 13
Figure 7: Distribution of Sample Enterprises by Location....................................................................... 14
Figure 8: Distribution of Sample Enterprises by Types of Economic Activity ......................................... 17
Figure 9: Destination of Products and Services of Sample Enterprises .................................................. 18
Figure 10: Starting of Business by the Sample Enterprises..................................................................... 18
Figure 11: Employment Growth in Sample Enterprises (%) By Sex of Entrepreneurs ............................ 20
Figure 12: Employment Growth in Sample Enterprises Over Time ........................................................ 21
Figure 13: Employment Generation Pattern by Types of Enterprises .................................................... 23
Figure 14: Employment Generation Patten by Types of Workers .......................................................... 23
Figure 15: Job creation by male entrepreneurs and female entrepreneurs .......................................... 24
Figure 16: Total Investment and Employment Generation .................................................................... 24
Figure 17: Number of people Employed vs. Total Investment ............................................................... 25
Figure 18: Employment by Credit Sizes (persons per million BDT)......................................................... 25
Figure 19: Purpose of taking loan or credit ............................................................................................ 28
Figure 20: Employment growth across sectors (%) ................................................................................ 35
Figure 21: Growth rate of self-employment (%) ..................................................................................... 35

List of Boxes
Box 1: Sectoral Employment Elasticity.................................................................................................... 26

ii
1 Introduction and Background
Bangladesh has successfully graduated from a low-income to a lower-middle income country in
2015. Despite considerable progress, moderate poverty and extreme poverty according to HIES
2016 were still at the 24% and is about 13%of the total population. The GDP and employment
shares of manufacturing and organized services have increased yet, according to the FY2015-16
Labor Force Survey (LFS FY2016), an estimated 79% of the workforce is engaged in low-income
agriculture and informal service-related activities. The manufacturing sector GDP and exports
have increased substantially, but it is heavily concentrated in one-product group - Readymade
Garments (RMG). Following solid growth in employment during 1995-2010, employment in RMG
has stagnated at around 4 million workers. Diversification of manufacturing production and
exports, and job creation outside agriculture and unorganized services are two of the most
pressing development challenges facing Bangladesh today.

International experience with small and medium enterprises (SMEs) suggests that the sector can
be an important driver of employment, investment, exports, and economic growth. Such
enterprises are especially crucial in developing countries where growth of large-scale enterprises
might be constrained by technology, finance, skill levels of workers and management, external
competition and domestic institutions. A healthy MSME sector can provide the bridge for
transition from an agrarian economy to a modern manufacturing and service-based economy. The
economic development of Taiwan is an important example of the potential of MSMEs. The
MSMEs can also be an important element of the growth of large-scale enterprises by creating an
appropriate eco-system through forward and backward linkages. Thus, co-production
arrangements between large scale manufacturing and MSMEs could be instrumental in spurring
the expansion of small enterprises while also providing the major source of employment. The
industrialization of Japan and Korea arguably provide the best examples of this potential.

The transition of some formerly centrally planned developing economies, like Vietnam and China,
and extremely regulated economies, such as India, towards a more market driven framework
along with deregulation – with an emphasis on facilitating a vibrant private sector oriented
economy – are other examples of how business deregulation and process simplification could
reduce transaction cost and spur the growth of small businesses in many developing countries.
MSMEs can also serve as a learning ground for acquisition of technology and skills by getting
connected to the global manufacturing production vertical value chain. Developing countries like
Bangladesh have an abundance of under-utilized low-cost labor. With minimum training this
factor endowment advantage could be converted to development by linking appropriately with
the global manufacturing value chain. The success of China’s industrialization is an important
example of this.

1
In low income countries, the microenterprises have emerged as the major source of non-farm
employment. In view of this development, the MSME sector is now more broadly defined to
incorporate these microenterprises and the acronym MSME is used. A major debacle in
understanding the contribution and performance of the MSMEs in Bangladesh is the absence of
reliable time series data, partly due to multiple and changing definitions. Better information is
available about the number of enterprises and total employment. The data suggests that the
number of micro and small enterprises and their employment share have grown rapidly in recent
years. They account for 64% of non-farm employment and 35% of total employment. In the
absence of reliable consolidated data for value-added, exports and investment it is very difficult
to say anything conclusive about their total contribution to growth of the economy. However,
fragmented evidence suggests that most enterprises work in an informal environment with low
average labor productivity, low real wages, and low capital per unit of labor. Also, data show that
micro and small enterprises are heavily concentrated in the informal service sector.
Consequently, the contribution to export is very small. In summary, while these enterprises have
played an important safety-net role in creating low-income employment and supported the
reduction of poverty in Bangladesh, the absence of a dynamic MSME sector is a big gap in the
process of transition of Bangladesh from a rural-agrarian economy to a modern manufacturing
and organized services-based economy.

SMEs now occupy an important position in the national economy. They account for about 45
percent of manufacturing value addition, about 80 percent of industrial sector employment,
about 90 percent of total industrial units and employ about 25 percent of the labor force. Their
total contribution to export earnings varies from 75 percent to 80 percent. The industrial sector
makes up 31 percent of the country's gross domestic product (GDP), most of which is coming
from SMEs. The country's SME sector has created 15 lakh jobs between 2009 and June 2014.1

IDLC SME division has been growing around 21% on an average in last one decade. According to
the diagnostic report titled SME 2.0 by International Finance Corporation (IFC), IDLC was the most
efficient SME Banker in Bangladesh because of the low cost-to-income ratio of 41% in 2018. IFC
also mentioned that IDLC’s NPL at 2.77% level in 2017 was one of the lowest in the market2. Being
one of the leading brands in this sector IDLC wants to map out the contribution it left in terms of
passive employment creation through financing of SMEs across the country through its 40
branches. IDLC also played a key role in providing non-financial services (NFS) to its clients and
non-client SMEs by establishing platforms like Purnota (Women Entrepreneur Financial Solution)
and arranged multiple trainings and workshops in collaboration with SME foundation. IDLC also

1
https://www.thedailystar.net/supplements/24th-anniversary-the-daily-star-part-2/potential-small-and-medium-
enterprises-70927.
2
Quoted by Ahmed Rashid Joy, General Manager and HoSME at IDLC Monthly Review (MBR), Vol 15, issue 4, April
2019.

2
propelled SME financing awareness through its unique social media campaigns like “Lets Shop
DESHI (Lets shop Local) that encourages sale of local products with local cultural touches during
festival seasons. IDLC has also promoted SME customer’s success stories to inspire others to be
on board. In 2019, IDLC won ‘Best Bank for SMEs: Bangladesh’ by Asia Money Awards, recognizing
the aforementioned activities by the organization.3

The objective of the study is to assess employment impacts of IDLC loans to the SMEs considering
the following aspects by
(i) gender of (female and male) entrepreneurs);
(ii) types of enterprises (manufacturing, services, trading and agro business);
(iii) major job categories (labor/worker, skilled workforce, salesperson, and executives);
(iv) loan sizes (below 10 lacs, 10 -25 lacs, 25-50 lacs, 50-75 lacs) and
(v) locations or regions of their operations.

2 Structure of the SME Sector in Bangladesh

In the absence of an alternative database, any meaningful analysis of the status of the MSME
sector must rely on the Economic Census carried out every 10 years by the Bangladesh Bureau of
Statistics (BBS). The latest round of the ‘Economic Census’ was conducted in 2013. Table 6
provides the distribution of enterprises by types for the two census years – 2003 and 2013. Since
the classification of enterprises used in 2003 Economic Census was not the same as it was in 2013
Economic Census, strict comparison is not possible.
The Economic Census counts a total of 7.8 million enterprises, which is a huge number. Some
89% belong to the cottage and microenterprise category; 11% are small enterprises and only
0.2% are medium and large. Importantly, the total number of enterprises more than doubled
during 2003 – 2013, suggesting an annualized growth rate of around 11.3%. Although the
numbers by enterprise type are not comparable because of changing definition, it would appear
reasonable to conclude that most growth came from the cottage and microenterprise category.
Independent data from the number of microcredit borrowers would seem to corroborate this.
On the surface, the data may seem to suggest that there has also been an astounding growth in
the number of small enterprises but because of changed definition a more robust finding is that
the number of small and medium enterprises together have grown at an unbelievably rapid pace
of 60% per year. Unfortunately, no meaningful statement can be made about large enterprises
because many of the enterprises defined as large in the 2003 definition (100-250 employment)
were classified as medium in the 2013 definition.

3
https://www.euromoney.com/article/b1ddfx091dvhx3/asiamoney-best-bank-awards-2019-bangladesh

3
Table 1: Types of Enterprises in Bangladesh
2003 2013
Enterprise Type Number of % of total Number of enterprises % of total
enterprises (000) (000)
Cottage and Micro 3,589,544 97.7 6,942,891 88.8
Small 74,629 2.0 859,318 11.0
Medium 5,125 0.1 7,106 0.1
Large 5,673 0.2 5,250 0.1
Total 3,674,971 100.0 7,818,565 100

Source: BBS, Economic Census 2013

Table below shows the employment growth between 2003 and 2013. In conformity with the
growth of enterprises, employment also doubled over the two periods, registering an annual
growth rate of 7.1% per year. Although the share of employment by micro enterprises dropped
from 68% in 2003 to 56% in 2013, they still are the main source of employment in the MSME
sector. The contribution of small and medium enterprises in employment doubled from 11.2 %
in 2003 to over 26.9% in 2013. Together, the MSME sector provides 86% of total employment
outside agriculture and the public sector. On the other hand, the share of large enterprises in
employment fell from 17.8 % in 2003 to over 14.1% in 2013. These results are broadly in line
with the findings of the LFS 2013 that shows that informal employment in services and
manufacturing accounted for almost 87% of total employment.
Table 2: Bangladesh--Total Nonfarm Employment, 2003-2013
2003 2013
Enterprise Type Number of % of Average Number of % of Average
persons total employment persons total employment
employed employed
Cottage and Micro 8,375,542 68.2 2.3 13,727,197 56.0 2.0
Small 1,375,223 11.2 18.4 6,600,685 26.9 7.7
Medium 342,737 2.8 66.9 706,112 2.9 99.4
Large 2,191,842 17.8 386.4 3,466,856 14.1 660.4
Total 12,285,344 100.0 3.3 24,500,850 100.0 3.1

Source: BBS, Economic Census 2013

Table 3 provides the distribution of enterprises for 2013 according to four types of enterprises –
Micro including cottage; Small; Medium; and Large. The table shows the predominance of micro
and small enterprises. According to the Economic Census 2013, 99% of the enterprises in
Bangladesh are either micro or small. Together, they accounted for 83% of the total persons
engaged in the nonfarm sector.

4
Table 3: Distribution of Enterprises by Types and Employment Size, 2013
Types Enterprises Total employment Employment per enterprise
Number (000) % Number (000) %
Micro 6,947 88.9 13,727 56.0 2.0
Small 859 11.0 6,601 26.9 7.7
Medium 7 0.1 706 2.9 99.5
Large 5 0.1 3,467 14.2 654.1
Total 7,819 100.0 24,501 100.0 3.1

Source: BBS, Economic Census 2013

Distribution of enterprises in 2013 by broad economic activities and size are shown in Table 4.
Following patterns are observed. Trading and services activities dominate in the case of micro
and small enterprises. Around 44% of the micro enterprises are engaged in different trading
activities while their share in different types of services is 43.6 per cent. The share of trading is
even higher at around 62% for small enterprises. On the other hand, industrial activities
dominate in the case of medium and large enterprises. Almost 50% of the medium enterprises
are involved in manufacturing activities. The share is almost 60% in the case of large enterprises.
Table 4: Distribution of Enterprises by Major Economic Activities 2013 (numbers in 000)
Micro Small Medium Large All enterprises

Number % Number % Number % Number % Number %

Manufacturing 831.2 12.0 30.9 3.59 3.0 42.1 3.12 59.4 868.2 11.1
Other Industrial 26.1 0.4 7.8 0.91 0.3 4.6 0.17 3.2 34.3 0.4
activities
Trading and 3,057.7 44.0 531.0 61.80 0.6 8.9 0.08 1.5 3,589.5 45.9
related activities
Transport and 1,265.9 18.2 37.7 4.39 0.2 2.3 0.04 0.8 1,303.8 16.7
storage
Other services 1,766.0 25.4 251.9 29.32 3.0 42.2 1.84 35.0 2,022.7 25.9
Total 6,946.9 100.0 859.3 100.0 7.1 100.0 5.25 100.0 7,818.6 100.0

Source: BBS, Economic Census 2013

Although the full information on the role of MSMEs in Bangladesh is not available, some useful
conclusions can be drawn about the potential of MSMEs in the future development of
Bangladesh.

First and foremost, MSMEs are the backbone for non-farm job creation in Bangladesh. This is
broadly similar to evidence from other countries. Some 99% of all non-farm enterprises fall in the
micro and small categories and they provided employment to 20.3 million people in 2013. This
makes them the largest source of employment generation outside agriculture (Figure 1). Their
most important development role is in terms of poverty reduction.

5
Second, Economic Census 2013 shows that most of these enterprises are informal in nature,
where formality is defined as enterprises that are registered with the government. Thus, only
35% of the total enterprises are registered with the government. Since 11% of total enterprises
belong to medium and large scale, all of whom are likely to be registered, this implies that some
76% of micro and small enterprises are not registered.
Third, although value-added data are not available, evidence from the manufacturing sector
suggests that the micro and small enterprises tend to have low value-added per worker and low
average wages. Figure 2 shows that average labor productivity in micro and small manufacturing
is very low and barely above the low productivity agriculture. The productivity challenge for the
Bangladeshi micro and small enterprises is indeed substantial
Finally, some 90% of microenterprises and 95.5% of small enterprises are engaged in informal
services for the domestic market. Therefore, there is no vertical or horizontal integration on the
value chain and only negligible contribution to the export market. These aspects are important
reasons for the lack of their dynamism.
Figure 1: Distribution of Employment (in Percent)

Source: BBS, LFS 2013 and Economic Census 2013

Figure 2: Average Labor Productivity 2013 (000 Taka)

Source: Calculated from BBS employment and national accounts data

6
3 Contribution of SMEs in GDP and Employment at the Global
Level
According to IFC (2010)4, there were 420 to 510 million SMEs worldwide, of which 9 per cent
were formal SMEs (excluding micro-enterprises). Between 80–95 per cent of them were in low-
and middle-income countries. Most of them were informal. Despite a lack of detailed data on
informal enterprises and variations in the definitions of informality, it was found that the number
of informal enterprises was high and that they accounted for a large share of all SMEs (i.e. more
than 90 per cent). In some developing countries, informal SMEs far outnumber formal enterprises
of the same size. In India, the ratio between unregistered and registered SMEs was 17:15.

In OCED countries female ownership of SMEs varies between 20 and 40 per cent and was on
average 25 per cent. For developing countries, the IFC estimates that there were 8–10 million
formal SMEs owned by women, which represented roughly a third of all formal SMEs. Female
entrepreneurships were skewed towards smaller enterprises. A third of exceedingly small
enterprises were owned by women and women ownership was limited to only 20 per cent of
medium-sized enterprises6.

Table 5: Estimated Number of SMEs Worldwide

Region No. of SMEs in region Percent of total SMEs Total formal

(million) worldwide SMEs
East Asia 170-205 44-46 11-14
Latin America 47-57 10-12 3-4
Sub-Saharan Africa 36-44 8-10 3.5
Central Asia and Eastern 18-22 3-5 2-4
Europe
South Asia 75-90 16-220 2-3
Middle East and North 19-23 4-6 1-3
Africa
High-income OECD 56-67 12-14 11-14
countries
Total 420-510 100 36-44
Source: IFC 2010

SMEs’ contribution to global gross domestic product (GDP) suggests that this sector, including
both formal and informal enterprises, accounted for 60–70 per cent of global GDP7. It is noted

4
IFC (2010), “Scaling-up SME access to financial services in the developing world (Washington, DC, 2010)”
5
K. Kushnir, M.L. Mirmulstein and R. Ramalho (2010), “Micro, small, and medium enterprises around the world: How
many are there, and what affects the count? World Bank/IFC MSME Country Indicators (Washington, DC, 2010)”
6
IFC, G20 Global Partnership for Financial Inclusion: Strengthening access to finance for women-owned small and
medium-sized enterprises (SMEs) in developing countries (Washington, DC, 2011).
,7 M. Ayyagari, T. Beck and A. Demirgüç-Kunt (2003), “Small & medium enterprises across the globe: A new database,
World Bank Policy Research Working Paper 3127 (Washington, DC, 2003)”

7
that the contribution of the informal enterprises (mostly micro-enterprises) decreases
significantly at higher income levels and there is a marked shift to a larger number of formal SMEs
that make a crucial contribution to national income.

The phenomenon that developing countries have a large number of micro-enterprises and some
large enterprises, but far fewer formal SMEs, is often referred to as the “missing middle” 8. This
missing middle is seen as the root problem of sluggish productivity increases and low growth
rates in developing countries.

Figure 3: Global GDP Shares of SMEs by Types of SMEs and Income Levels of Countries

100.0
90.0
31.5 37.0
80.0 38.5
70.0
GDP share (%)

60.0 11.0
30.0
50.0
40.0 46.5
30.0
52.0
20.0 38.5
10.0 15.0
0.0
Low income countries Lower middle income High income countries
countries

Formal SME Informal Residual

Source: Ayyagari et al., 2003.

In most countries, more than 50 per cent of total net employment have been created by the
smallest size enterprises of between five and 99 employees.9 This finding holds for every country
income group. They also found that “as for enterprises with up to 250 employees, there is no
unbiased global data. Nevertheless, it can be assumed that the net employment creation of the
entire SME segment is significantly higher than the above figure. Data from the EU show that 85
per cent of net employment creation is attributable to SMEs with between one and 250
employees, including enterprise entries and exits”.10 However, SMEs are less resilient than the
large enterprises during the time of crisis. J. de Kok et al. (2011), argued that “during times of
economic downturn, the picture changes. Employment in SMEs seems to be less resilient to
economic crises. Figures from the EU show that during the Great Recession of 2007–09, the

8
Harvard Kennedy School, Entrepreneurial Finance Lab Research Initiative: The missing middle, 2014.
9
M. Ayyagari, A. Demirgüç-Kunt and V. Maskimovic: “Who creates jobs in developing countries?” in Small Business
Economics (2014, Vol. 43).
10
J. de Kok et al.: Do SMEs create more and better jobs? (Brussels, 2011).

8
number of jobs in SMEs fell by an average of 2.4 per cent annually, as opposed to 1 per cent in
large enterprises”.

Figure 4: Share of Total Net Job Creation by Enterprise Size and Country Income Group

50
45
40
35
30
percent

25
20
15
10
5
0
Low income Lower middle Upper middle High income
countries income countries income countries countries

5-19 employess 20-99 employess 100+ employess

Source: Based on De Kok et al. (2013) and Ayyagari et al. (2013)

4 IDLC Employment Survey: Methodology of Sample

Distribution
As mentioned above, the objective of the study is to assess employment impacts considering
following aspects by (i) gender (female and male) of entrepreneur; (ii) types of enterprises
(manufacturing, services, trading and Agro-business); (ii) major job categories (labor/worker, skilled
workforce, salesperson, and executives); (iv) and loan size (below 10 lacs, 10 -25 lacs, 25-50 lacs, 50-75
lacs and above) and (v) location or region.

IDLC provided population data of their Micro and Small enterprise clients for the latest available
period (i.e. 2019). The data has the following features:

1. Total client/sample size comprising all Micro and Small enterprises receiving financial support
in 2019 from IDLC was 13,931.
2. The share of Micro enterprises was 30 per cent and the share of Small enterprise was thus 70
per cent.
3. The Female entrepreneurs constitute only 8.4 per cent of the IDLC clientele and the share of
male entrepreneurs was 91.6 per cent.

9
4. The share manufacturing enterprises was 12.3 per cent; services 20.4 per cent and trading
67.2 per cent.
5. Enterprises located in 8 administrative divisions have been covered to capture the location
or regional variations in the performance of the selected SME enterprises.
Figure 5: Distribution of IDLC’s Total Borrowing Clients by Major Categories (%)

Division share (%) Gender share (%) Enterprise share (%)

100% 3.9
2.9 100% 100%
6.1
90%
80% 18.6
80% 80%
8.9 70% 67.3
60% 60%
60%
91.6
40% 50%
46.2 40%
40%
20% 30% 20% 20.3
2.2 20%
11.3 12.4
0% 10% 0%
Chottagram Barisal 8.4 Trading
Dhaka Rajshahi 0%
Services
Khulna Sylhet
Mymensingh Rangpur Female Male Manufacturing

Source: IDLC (2019)

To assess the employment impact, data from a sample population of 13,931 enterprises were
collected through a random sample survey. It was proposed that out of this large sample,
information/data would be collected from 1,000 SMEs, randomly selected so that the sub-
samples were also of reasonable size for each of the above-mentioned classification – gender;
types of enterprises; loan sizes; regions etc.

One approach that could have been adopted was to determine the number of samples for each
classification based on their respective shares in the total sample population (i.e. 13,931). The
approach would however pose the following problems:

● Since the share of female entrepreneurs was only 8.4 per cent in total population, following
the sample share approach, the number of female SMEs was estimated to be 84 (=> 0.084 *
1,000). The derived 84 samples for female entrepreneurs would seem inadequate considering
the fact that this sample size would need to be divided among all other classifications,
including 8 regions.
● Similarly, since the share of manufacturing enterprises was only 12.3 per cent, total
manufacturing sample following the use of population share was been estimated at 123 (=>
0.123 * 1,000). Again, the derived sample size for 123 manufacturing SMEs seemed
inadequate considering all other classifications.

10
● Analogously, as the share of services enterprises was only 20.4 per cent, the total sample for
the services sector following the use of population share was estimated at 204 (=> 0.204 *
1,000). The derived sample size was also considered inadequate considering the need for
dividing these SMEs into all other categories. Enterprises by the 8 administrative division has
been used to capture the location or regional aspects of the enterprises. The use of all 8
regions (divisions) to represent the regional dimension appeared inappropriate given the
sample size of 1,000.

Thus, to arrive at a sample distribution considering all the above classifications, the following
adjustments were made in sample selection:

1. Increased the number of female entrepreneurs to 250 or 25% of total sample size of 1,000
(in place of 84)
2. Set the distribution of samples among the types of enterprises as Manufacturing – 400
(instead of 123); Services – 325 (in place of 204); and trading – 275 (in place of 538)
3. Instead of 8 divisions, four regions—Dhaka, Chottogram, Khulna and Rajshahi-- were
selected representing largest clusters as well as geographic distribution into account. The
sample sizes for the 4 divisions were determined taking into account the geographical
distribution of the sample population and consideration for a respectable sample size for
each division. Accordingly, the distribution of the samples was Dhaka (300); Chottogram
(225); Khulna (250); and Rajshahi (225).
Figure 6: Revised Shares for Sampling Distribution (in percent)

Region share (%) Gender share (%) Enterprise share (%)

100% 100% 100%
90% 22.5 90% 90%
27.5
80% 80% 80%
70% 70% 70%
25.0 60%
60% 60% 75.0 32.5
50%
50% 50%
40%
40% 30.0 40%
30%
30% 30%
20% 40.0
20% 20% 10%
10% 22.5 10% 25.0
0%
0% 0%
Chottogram Dhaka Manufacturing Services
Khulna Rajshahi Female Male Trading

Based on the above mentioned 3 distribution criteria, the following sample distribution was
derived for the purpose of the sample survey. The sample distribution is shown below.

11
Table 6: Sample Distribution by Region, Sector and Sex of the Entrepreneurs
Manufacturing Services Trading
Region/Types/Gender Total
Female Male Female Male Female Male
Chottogram 23 68 18 55 15 46 225
Dhaka 30 90 24 73 21 62 300
Khulna 25 75 20 61 17 52 250
Rangpur 23 68 18 55 15 46 225
Total Sample 100 300 81 244 69 206 1000
Total Enterprise 400 325 275 1,000
Total female 100 81 69 250

In accordance with the TOR, PRI prepared the questionnaire and provided training to
enumerators selected by IDLC (Figure 6). IDLC conducted the pre-testing of the questionnaire to
assess the ease as well as effectiveness of data collection at the field level. As expected, pre-
testing provided important observations on some specific questions and accordingly the original
questionnaire was revised to address the comments/observations received from field
enumerators and IDLC staff. Data collection was the responsibility of IDLC. Field level primary
data collection were conducted during October and November 2019. Although, the aim was to
collect data for 1000 SMEs, in reality due to various field level constraints, data for 830
enterprises were collected.

To validate the data quality, following a standard practice, PRI conducted a validation exercise by
collecting same information (as contained in the questionnaire) from 100 randomly selected
samples out of the 830 samples.
Figure 6: Division of Tasks Between the IDLC and PRI Staff as Per the TOR

12
5 Profile of Sample Respondents
Even though, the objective was to collect data from 1,000 enterprises randomly selected, in
reality, complete information (as contained in the questionnaire) could only be gathered from
782 enterprises – drop of about 22 per cent between planned and actual samples. The divergence
between the planned and actual samples are attributable to a number of factors such as: (i) errors
in responses; (ii) incomplete information; and (iii) inability to reconcile responses to some of the
key questions even with repeated attempts. As a result, planned distributions related to region,
gender and types of enterprises have diverged in actual distribution discussed below.

More than 83 per cent of the sample enterprises were located in Urban or Semi-Urban areas.
Rural areas accounted for only 17 per cent of the sample enterprises. Choice of Urban or Semi-
Urban areas for establishing enterprises perhaps relate to various factors – better or supportive
infrastructure (including utilities and road, rail and river connectivity); and local demand for their
services and products.
Figure 7: Distribution of sample enterprises by residence

13
Figure 8: Distribution of Sample Enterprises by Location

Industrial (including BSIC) areas have been developed to attract enterprises. Similarly, EPZ areas
also provide all types supportive services to the enterprises. However, only 8 per cent of the
sample enterprises are located in these two areas. Most of the enterprises are located outside
these industrial zones (about 92 percent). This is not surprising given the limited resources that
most of these enterprises operate with. Most enterprises are very small based on any
conventional indicators.
Distribution of Ownership Structure

A review of ownership structure indicates that most of the firms in the sample (almost 97 percent) were
single owner operation. Joint ownership was limited to only 2.4 percent and the proportion of private
limited companies were less than one percent (0.9 percent). This kind of single person ownership had
many limitations in terms of limited capita, access to capital, and consequently the size of their business
operations in terms of turnover and employment.
Table 7: Ownership Structure of Firms

Ownership Status Number Percent

Single Owner 851 96.7
Joint Owner 21 2.4
Private Limited Company 8 0.9

Since most of the firms were single owners and small in size, they tended to start their operations with
very limited access to formal sources of financing. Almost 92% of the firms started their operations solely
with own funding, and the second important source of funding (3.1 percent) was loans from informal
sources like friends and relatives (individual and informal loans). Firms starting operations solely with
loans from institutional sources were only 2.8 percent, of which IDLC loans accounted for 1%. It is
surprising to note that firms starting their business with own resources and access to other formal loans
were only 2.6% of the sample. This shows that even until now, MSMEs are primarily depending on their

14
own resources and loans from individuals or informal sources to start their business. The formal financial
sector was playing a very limited role in in supporting fresh/new business enterprises.
Table 8: Sources of funding at the time of business start-up by the number of business enterprise

No. of business
enterprise Share (%)
Own Funding 716 91.6
Individual/Informal Loan 24 3.1
IDLC Loan 8 1.0
NGO/MFI 9 1.2
Other bank or institutional loan 5 0.6
Own finance plus other loans 20 2.6

Total 782 100

The average size of investment for the sample enterprises depend on the ownership structure of the
business entities. It is in line with expectation that single owner forms had smaller size of investment on
average (Tk. 27.9 million), constrained by access to financing and their own capacity constraints. Joint
owners had somewhat larger average investment in their enterprises (Tk. 43.5 million), as multiple owners
could pull together larger amount of investment and share the profits. It is not surprising to find that
private limited companies had a much higher average investment in the company (Tk. 142.9 million).
Table 9: Total and Average Investment by Types of Business

Total Capital Investments Mean Capital Investments

(Million BDT) (Million BDT)
Single Owner 23750 27.9
Joint Owner 913.9 43.5
Private Limited Company 1143 142.9

It is not surprising that the average annual turnover of the single owner firms was smallest of the three at
Tk 57 million per annum. The corresponding figures for Joint owners and private limited companies were
Tk. 133.7 million and Tk. 172.8 million, respectively. If we look at the average capital investment in relation
to the average annual turnover, the interesting thing that appears is that joint owner businesses were the
most efficient among the three. The ratio of average turnover to average investment was the highest for
joint owners at 3.07. In contrast, the private limited companies with the highest average investment were
the least efficient with the average turnover to investment ratio of 1.21. The corresponding ratio for single
owners was 2.04 indicating that single owner operations were reasonably efficient in managing their
operations as measured in terms of turnover to investment ratio.

Performance of different types of enterprises can also be determined from the growth in turnover for the
various ownership structures. In terms of growth in average annual turnover, joint owners once again
performed the best with annual average turnover growing more than five-fold to Tk. 207.5 million during
FY15-19. This performance in terms of growth in turnover was very impressive. For single owners, the
average turnover more than doubled to Tk. 80.1 million, still quite impressive with more than 20% average
annual growth in turnover. This group, accounting for more than 96% of the sample population,

15
essentially represents the IDLC borrower base and thus reflects very positively on the performance of IDLC
borrowers and the institution itself.
Table 11: Average Annual Turnover (Million BDT)
2014-15 2015-16 2016-17 2017-18 2018-19 Average for the last 5 years
Single Owner 39.5 46.4 54.2 64.8 80.1 57
(17.5) (16.8) (19.6) (23.6)
Joint Owner 39.2 87.6 180 154.2 207.5 133.7
(123.5) (105.5) (-14.3) (34.6)
Private Limited Company 154 164.3 143.9 186.4 215.6 172.8
(6.7) (-12.4) (29.5) (15.7)
***The number in parenthesis are the growth rates for the subsequent years

Given the fact that single owner small firms dominated that sample and also the whole population being
served by IDLC, in terms of total annual turnover, the small single owner firms accounted for much higher
total annual average turnover compared with joint owner and private limited companies. Total annual
turnover of the single owner enterprises was Tk 66.2 billion in FY19, which was quite respectable (Table
12).
Table 12: Total Annual Turnover (Billion BDT)
2014-15 2015-16 2016-17 2017-18 2018-19 Average for the last 5 years
Single Owner 32.3 38.1 44.6 53.4 66.2 46.92
(18) (17.1) (19.7) (24)
Joint Owner 0.74 1.66 3.42 2.93 3.94 2.54
(124.3) (106) (-14.3) (34.5)
Private Limited 0.92 0.98 0.115 1.49 1.73 1.05
Company
(6.5) (-88.3) (1195.7) (16.1)
N/A 0.144 0.17 0.244 0.31 0.337 0.24
(18.1) (43.5) (27) (8.7)
***The number in parenthesis are the growth rates for the subsequent years

An overwhelming majority of enterprises expressed their need for more access to credit (almost 92%;
Table 13). Only about 8% of enterprises noted that they would not need credit in the near term. When
further inquired about preferred sources of credit for them who wanted to more credit, 73% of them
indicated their preference for IDLC credit. This may partly be a reflection of their ongoing business
relationships, but also indirectly indicates a degree of client satisfaction with IDLC. Their preference for
going to private banks were less than 11% and the same for going to government banks was 5.6%.
Table 13: Firms Responding to Questions About Their Need for Credit

Response Freq. Percent Cum.

Yes 803 91.67 91.67

No 71 8.11 99.77
N/A 2 0.23 100

Total 876 100

16
Table 14: Preferred Source of Credit for the Sample Firms (Q 17)

Response Freq. Percent

IDLC 608 73.0

Government Bank 47 5.6
Private bank 90 10.8
Leasing Company 1 0.1
MFI 1 0.1
Others 6 0.7
N/A 80 9.6

Total 833 100

Figure 9: Distribution of Sample Enterprises by Types of Economic Activity

Most dominant type of enterprise was trading (69 percent). The share of enterprises involved in
manufacturing activities was around 18 per cent. More than 9 per cent of the enterprises were
running service-related activities. Agro-processing accounted for only 1.9 per cent. Some of the
sample enterprises were involved in more than one types of activities – their share was relatively
modest at 1.75 per cent11.

11
It is important to note some mismatch between planned sample distribution (figure 4) and actual responses (figure
8). This is due to dropping of about 22 per cent samples due to incomplete responses,

17
Figure 10: Destination of Products and Services of Sample Enterprises

The majority of the sample enterprises cater to the local, regional and national markets. Sales of
their products and services were mostly at the local markets (about 95%). Only 1.3% of the
sample enterprises were catering to the export market. The share of sample enterprises catering
to both local and export markets were even smaller at around 0.8%.
Figure 11: Starting of Business by the Sample Enterprises

Different enterprises came into operation in different years. Some of the enterprises came into
operation as early as 1989—almost 30 years ago. To get a better understanding of the
distribution of the enterprises’ coming into operation, the total sample was distributed among
five selected timeframes. It is interesting to note that 15.4% of the sample enterprises have been
in operation before 2000. Around 25% of the sample enterprises came into operation between
2000 and 2005 and another 25% between 2011 and 2015. The highest incorporation took place
between 2006 and 2010. Lowest incorporation (i.e. 4.9%) has been reported for timeframe 5
ranging from 2016 to 2019. The above trends, however, raise some related questions:

• Is there any link between the number of incorporations to the number/amounts of loans
given by IDLC to new borrowers?

18
 • Did the IDLC’s overall lending size and new lenders declined at the whole population
level? If yes, what were the reasons?
We conducted further review of loan history from 2015 to 2019. According to the loan
disbursement trends between 2015 and 2019, it supports the findings from the field survey that
the proportion of newly incorporated businesses in the total number of IDLC supported
enterprises has declined. This is nothing unusual since as the pool of enterprises with long track
record of engagement increases, the proportion of new enterprises out of the total pool of
borrowers may appear to be lower as a percent of the total, but the absolute number may still
be increasing. Moreover, IDLC’s total lending did not decease over these years but their lending
to newly incorporated clients as a percentage of the total population has declined. Although, no
particular factors are determined, the emerging trends may either be due to the (i) graduation of
existing /older firms to relatively larger loan sizes as their business operations and associated
credit demand increased; and (ii) newer firms start with low access to crdit and the limits
generally increase over time;
Table 15: Percent of Enterprises on the Basis of Length of Financial Relationship with IDLC, During 2015-19
Years Length of IDLC Relationship with Enterprises
5 Years or Less 6 to 10 years 11 to 15 years 16 to 20 years Above 20 years All
2015 25.71% 31.43% 22.86% 14.29% 5.71% 100%
2016 22.19% 32.71% 20.57% 13.45% 11.08% 100%
2017 22.83% 31.60% 21.49% 13.78% 10.31% 100%
2018 17.11% 37.05% 20.43% 15.46% 9.94% 100%
2019 18.20% 36.09% 22.02% 13.97% 9.71% 100%
Source: IDLC (June, 2020)

An analysis of growth of manufacturing enterprises between 2012 and 2019 using the data of the
“Survey of Manufacturing Industries” provide some further insights to growth of micro, small,
medium and large enterprises in Bangladesh. Except for the small enterprises having 25 to 99
employees, all other three categories reported negative growth over the seven-year period.
Small enterprises recorded 50% growth. The BBS survey of manufacturing industries projects a
sobering picture—due to market consolidation the numbers of medium and large enterprises
have declined over the last 7-year period. The expected progression from small to medium and
from medium to large have not been happening.
Table 16: Growth of Manufacturing Enterprises between 2012 and 2019
Types of Establishment 2012 2019* Growth
Micro (TPE 10-24) 17,384 16,689 -4.00
Small (TPE 25-99) 15,666 23,557 50.37
Medium (TPE 100-250) 6,103 3,014 -50.61
Large (TPE 250+) 3,639 3,031 -16.71
Total 42,792 46,291 8.18
Note: * Preliminary Findings, May 2019; and TPE=Total Person Engaged
Source: Survey of Manufacturing Industries, BBS

19
6 Impact on Employment
Employment effect of all enterprises together as well as by male and female ownership has been
shown in Figure below. Employment growth in sample enterprises from inception to 2019 has
been large. Employment growth in all sample enterprises from inception to 2019 was 105.7 per
cent. With regards to employment, female enterprise owners have been found to perform better
than their male counterparts. More specifically, employment growth in female-owned sample
enterprises from inception to 2019 has been found at 146.2 per cent significantly above the job
creation rate at the male owned enterprises (slightly over 100 percent). The significantly better
performance of women enterprises in terms of job creation requires better understanding of
their operations and may be attributable to other factors like larger growth in profitability,
average size of initial capital or preferential access to credit for the female entrepreneurs.
Figure 12: Employment Growth in Sample Enterprises (%) By Sex of Entrepreneurs

Total 105.7%

Female 146.2%

Male 100.5%

0.0 20.0 40.0 60.0 80.0 100.0 120.0 140.0 160.0

Since the inception of years of the sample enterprises were not same, it was not straight forward
to calculate the annual employment growth. However, the employment growth rates by five
intervals or periods have been calculated.

Growth of employment in enterprises in the rural locations (i.e. 114.9 %) was slightly higher than
the employment growth in enterprises in the urban locations (i.e. 107.6 %). However, most of
the employment was generated in the urban locations (i.e. around 81 %) compared to only 15
per cent in the rural locations.
Table 17: Employment in Manufacturing Enterprises Between 2012 and 2019
Person employed at the start of business Person currently employed Growth (%)
Rural location 879 1,889 114.9
Urban location 5,083 10,550 107.6
N/A 435 717 64.8
Total 6,277 13,029 107.6

20
Figure 13: Employment Growth in Sample Enterprises Over Time

Some interesting observations include:

• Sample enterprises those were incorporated before the Year 2000, their overall
employment growth was 137.7 per cent (i.e. 2,952/1,242). Considering at least a 19-year
period when these enterprises were established, as it was found that the simple annual
employment growth was 7.26 per cent. If we assume that some of the firms were
established even 10 years before 2000, the overall time period will be much longer, and
the average simple annual growth rate of employment was 4.75 percent.
• Employment growth in sample enterprises, which were incorporated between 2000 and
2005 was 94.8 per cent (i.e. 3,835/1,969). Thus, the simple annual employment growth
since their establishment (about 17.5 years) was 5.4% per cent.
• Employment growth in sample enterprises those were incorporated between 2006 and
2010 was 102.5 per cent (i.e. 2,381/1,176). Thus, the annualized employment growth over
the 11.5- year period was 8.9 per cent.
• Similarly, employment growth in sample enterprises those were incorporated between
2011 and 2015 was 91.8 per cent (i.e. 2,344/1,222). Thus, the simple annual average
employment growth over the 6.5-year period was 14.1 per cent.
• Finally, the increase in employment in the most recently established sample enterprises
(established between 2016 and 2019) was 75.2 per cent (i.e. 487/278). Thus, the
annualized employment growth over the 1.5-year average period was significantly higher
at about 50 per cent.
● The findings indicate that more recent firms have been expanding more in terms of
employment generation. The older firms were much slower in terms of employment
generation.
● Another important observation is the average employment size of around 16 person per
enterprise (i.e. 16 =>13,156/830). This is close to the average employment size for micro

21
 enterprises as reported by BBS in 2003 (e.g. according to BBS, average employment size
by small enterprise was18.4 in 2003).

We again, used the data of manufacturing enterprises of 2012 and 2019 to generate further
insights to the employment patterns and growth by micro, small, medium, and large enterprises
in Bangladesh.
Table 18: Employment in Manufacturing Enterprises Between 2012 and 2019
Types of Establishment 2012 2019* Growth
Micro (TPE 10-24) 271,644 263,720 -2.92
Small (TPE 25-99) 738,801 1,127,841 52.66
Medium (TPE 100-250) 1,041,220 461,142 -55.71
Large (TPE 250+) 2,964,272 4,027,141 35.86
Total 5,015,937 5,879,844 17.22
Memorandum item
Employment size
Micro-small 30.6 34.6
Medium 170.6 153.0
Large 814.6 1328.7
Note: * Preliminary Findings, May 2019; and TPE=Total Person Engaged
Source: Survey of Manufacturing Industries, BBS

Key observations are:

• Overall employment growth between 2012 and 2019 was 17.2%, growing at a simple average
rate of 2.4% per annum.
• Employment growth was significantly positive only in small and large enterprises – with about
53% and 36% employment growth, respectively, over the 7-year period.
• Employment size in micro-small enterprises increased from 30 in 2012 to 34 in 2019. The
employment size for medium enterprises decreased somewhat, but for the large enterprises
the average employment size increased by more than 63% to about 1329.
Employment generation by types of activities has also revealed some interesting patterns. The
employment generation by types of activities is captured below. The largest employment growth
was in the service sector enterprises (174.2 per cent) over the entire period (i.e. from inception
to 2019). The employment growth among the manufacturing enterprises was also large (131.0
per cent) over the entire period (i.e. from inception to 2019). Employment growth among
enterprises engaged in trading activity was moderate at 74.6 per cent. On the other hand,
employment growth among enterprises engaged in agro-processing activity was modest
compared to the other three types at 30.3 per cent.

22
Figure 14: Employment Generation Pattern by Types of Enterprises
Panel A: Employment generation (persons) Panel B: Employment share (%)

Job creation by types of jobs (namely family worker, salaried jobs, and day labourer) has also
been attempted. The job creation by types of jobs is captured in Panel A in Figure 14. The largest
growth rate has been found for salaried jobs. The growth rate was 134 per cent (8,170/3,491)
over the entire period (i.e. from inception to 2019). The second largest growth rate has been
found for day labourer category (94.3 per cent) over the corresponding period. The lowest
growth was reported for the family labourers with a growth rate of 48.2 per cent.
The distribution of jobs by types of jobs has been captured in Panel B. The largest share was for
the salaried jobs with 62 per cent followed by day labourers (22 %). The share of family workers
was 16 per cent. The distribution is quite good and in line with the general development in the
economy. The higher proportion of salaried employees is generally favorable for the workers of
Bangladesh.
Figure 15: Employment Generation Patten by Types of Workers
Panel A: Employment Generation by Worker Panel B: Shares of Different Types of Workers
Type(persons) (%)

23
Since majority of the entrepreneurs are male, they account for the bulk of the employment
creation in the sample enterprises. This feature is captured below for three years for which data
has been collected. The share of employment generated by male entrepreneurs in total
employment has been around 71% during 2016-18. However, the share dropped by about 2.7
percentage points in 2018-19 as the share of employment generated by female entrepreneurs
increased.
Figure 16: Job Creation by Male and Female Entrepreneurs (%)

Although single owner small firms generally employed lower number of workers per unit of
operation, because of their sheer number, most of the employment was created in firms with
smaller investment ranging up to Tk. 100 million (Figures 17 and 18). Almost 74% of employment
was generated in this low investment—generally single owner owned—enterprises. The second
highest proportion/number of employment (7.4%) was created at the enterprises with highest
level of investment ranging between Tk. 501-800 million.
Figure 17: Total Investment and Employment Generation

501-800 7.4
Invt (in M. BDT)

301-500 3.2

201-300 2.4

101-200 5.8

0 - 100 73.7

0 20 40 60 80
Empl Gen. (%)

24
Figure 18: Number of People Employed vs. Total Investment

12000
Number of people Employed

9599
10000

8000

6000

4000

2000 752 969 727

318 412
0
0 - 100 101-200 201-300 301-500 501-800 N/A
Total Investment (in M. BDT)

Employment generation in sample enterprises has been compared against the credit sizes to
assess credit to employment generation trends. It took about 1 million BDT to create 37 person-
jobs in 2015. This association dropped to about 8 persons in 2019. This is a concerning trend –
may reflect a host of issues encompassing installation of labour replacing technology; increasing
labour productivity; or disbursement of credit to sectors which have low employment intensities.
To get informed answer, more information is needed.
Figure 19: Employment by Credit Sizes (persons per million BDT)

This declining trend is also consistent with national level employment data, which shows a steady
decline in employment elasticity of GDP in recent years (Box 1).

25
 Box 1: Sectoral Employment Elasticity
Employment elasticity is a measure of the percentage change in employment associated with a 1 percentage point change in economic growth. The employment elas ticity
indicates the ability of an economy to generate employment opportunities for its population as percent of its growth (develop ment) process. An employment elasticity of 1
implies that with every 1 percentage point growth in real GDP, employment increases by 1%. An elasticity value greater than 1, indicates that the percentage increase in
employment is greater than the percentage increase in real GDP or output. A value that is less than 1 suggests that the employment generation is relatively insensitive to
changes in growth.
Table 1 depicts the employment elasticity estimates for Bangladesh during FY95 through FY18, by broad economic sectors. It in dicates a number of stylized facts: (i)
Employment elasticity have generally been inelastic for all sectors for the most part; and (ii) there has been a secular decline in elasticity since FY06; (iii) employment elasticity
of agriculture which used to be the highest among all four major sectors, collapsed in during FY 10-18 and became negative. shows that employment elasticity has declined
over the period during 2005-06 and 2017-18.
The fact that in all sectors the elasticity was significantly below 1, indicated that the percentage increase in jobs were about half of the percentage increase in real GDP. It is
observed that the job-creating ability of the agriculture sector—which still employs the highest proportion of workers among the four sectors--has declined overtime and has
become negative in recent years. In case of the agriculture sector, this outcome was attributable to a positive structural transformation as workers moved from agriculture to
industry and services resulting in a higher worker’s productivity and wage/income growth. This migration of labor from agriculture also helped mechanization of the agriculture
sector, overall output increase and the consequent productivity growth.

Table 1: Employment elasticity for Bangladesh, by broad sectors

Employment Elasticity
Sector
1995-96 to 1999-00 1999-00 to 2005-06 2005-06 to 2009-10 2009-10 to 2017-18
Agriculture 0.73 0.82 0.71 -0.09
Manufacturing 0.26 0.78 0.87 0.65
Construction 0.27 0.63 2.22 0.55
Services 0.21 0.69 0.27 0.40
Total 0.54 0.59 0.55 0.25
Source: Sample households survey, GED, Planning Commission and ADB-ILO report (2016)

The RMG sector has helped fuel the growth of the manufacturing sector. However, restructuring has taken place in the manufact uring sector. This sector has become more
capital intensive resulting in a cut in employment. The manufacturing sector is dominated by Ready-Made Garments (RMG) and textiles. RMG and textiles have been subject
to restructuring and has become more capital intensive, as a result of which employment has decreased. The non-RMG part of the manufacturing sector has remained static.
Due to major restructuring in the RMG and textile part, the overall labor absorption has decreased in the manufacturing secto r. The construction sector has experienced sharp
fluctuations in employment elasticity over time. The sharp rise in elasticity during 2005-10 was probably attributable to the construction boom in the real estate sector during
that time. Increased Government spending also contributed to labor elasticity in the construction sector. However, as increasing number of mega projects were being
implemented and the degree of mechanization increased in the sector, the elasticity estimates declined. The employment elasticity of growth in the service sector has also
experienced some fluctuations. For FY10-18, the service sector was the only sector where there was an increase in employment elasticity.
Employment elasticity also acts as an indicator of the direction of movement of labor productivity. Table 2 illustrates the employment elasticity for Bangladesh, by sectors.
The employment elasticity from (2009-10 to 2017-18) in the Agriculture, Forestry & Fishing sector is negative and one of the lowest among the sectors. The job creating ability
of the Paper, printing & publishing sector has been the highest followed by leather & footwear, Pharmaceuticals, Manufacturin g of other products and electronics. Although
the employment elasticity is not very responsive to growth in case of the manufacturing sector, plastic products and Garments and textiles have high employment elasticities.

Table 2: Employment elasticity for Bangladesh, by sectors

Sectors Employment Elasticity (2009-10 to 2017-18)
Agriculture, Forestry & Fishing -0.09
Mining & Quarrying 1.31
Manufacturing 0.65
Garments and Textiles 1.78
Leather & Footwear 3.30
Food, Beverage & Tobacco -0.55
Pharmaceuticals 2.76
Chemicals 0.35
Paper, printing & publishing 5.75
Iron, steel & metal 0.36
Electronics 2.23
Wooden Products 0.14
Plastic products 1.78
Motor vehicles & transport 1.20
Non-metallic mineral products 1.48
Manufacturing of other products 2.72
Electricity, Gas and Water 0.49
Construction 0.55
Wholesale and Retail Trade 0.37
Hotel & Restaurants 0.77
Transport, Storage, and Communication 0.75
Financial Intermediations 0.25
Real Estate, Renting and Business activity 0.25
Public Administration and Defense 0.00
Education 0.59
Health and Social Workers 1.01
Community, Social, and Personal Services -0.78
Total 0.25
Source: Sample households survey, GED, Planning Commission

26
7 Regional Variations
We also assessed the regional variations with respect to employment generation and turnover
by four regions – Chottogram, Dhaka, Khulna, and Rangpur. According to the sample statistics
Chottogram turned out to be the leading region in terms of employment generation and
turnover. The turnover in Chottogram is almost double the other three regions. However, due
to limitations of questions and samples, it was not possible to find out the factors behind the
better performance of the MSM enterprises in the Chottogram region.
Figure 20: Regional variations in employment and turnover
Panel A: Per enterprise employment (Persons) Panel B: Average turnover (Million BDT)
18.3

20.0 80.0 75.0

16.1
15.8

18.0 70.0
15.0
14.2

16.0
60.0
12.6

12.3 50.8
14.0 47.0

Million BDT
50.0
10.5

12.0
40.0
10.0
30.0 26.7
8.0
6.0 20.0
2.5

2.4

4.0 10.0
1.9

1.7

2.0 0.0

Dhaka

Rangpur
Chottogram

Khulna
0.0
Chottogram Dhaka Khulna Rangpur

Male Female Total

8 Purpose of Taking IDLC Loan

The respondents were also asked about the reasons as to why they had taken loans from IDLC.
The main reasons for taking loans from IDLC were of two types:
(i) expansion of business – ranged between 39% for trading and 80% for services; and
(ii) working capital - with large variations across sectors ranging between 55% in trading and 12%
in services. For manufacturing, the use of credit for expansion and for working capital was in the
proportion of 60:40, respectively.
IDLC loans used for re-structuring their business or for other purposes were found to be quite
modest, together ranging between 3%-8%, depending on the sector.

27
Figure 21: Purpose of taking loan or credit

One positive sign was that majority of IDLC borrowers (about 56%) have taken loans from IDLC
more than once. The propensity of borrowing more than once was the highest among enterprises
which are engaged in more than one type of activity (more than 67%). However, that segment
is a very small part of the sample, and the larger services sector enterprises also had a high
propensity to borrow more than once from IDLC (more than 61%).
Table 21: No of Times IDLC Loans Received by Types of establishments

Once Twice More than 2 Total

Type of Enterprise No. % No. % No. % No. %

Trading 240 44.12 174 31.99 130 23.9 544 100

Service 30 38.46 22 28.21 26 33.33 78 100
Manufacturing 69 46.94 45 30.61 33 22.45 147 100
Agro-business 10 62.5 5 31.25 1 6.25 16 100
More than One Type 4 33.33 4 33.33 4 33.33 12 100

Total 353 44.29 250 31.37 194 24.34 797 100

It is quite informative to know that almost all (about 92%) sample enterprises started their
business with own funding. Borrowing from individuals or from informal lending sources was the
distant second representing 3.1% of the sampled enterprises. The third important source at the
time of starting the business was combination of own financing and other loans (2.6% of the
sample). The surprising finding is that only 2.8% of IDLC borrowers started their business with
funding from formal institutional sources including IDLC, MFIs and other financial institutions.

28
IDLC loans were used at the starting of the enterprises by only 1% of the sample enterprises. The
fact that such a small number of enterprises could start their enterprises using IDLC or any other
formal sector loans, raises serious questions about access to finance for new entrepreneurs in
Bangladesh.
Table 23: Sources of funding at the time of business start-up

No. of business Share (%)

enterprise

Own Funding 716 91.6

Individual/Informal Loan 24 3.1
IDLC Loan 8 1.0
NGO/MFI 9 1.2
Other bank or institutional loan 5 0.6
Own finance plus other loans 20 2.6

Total 782 100

9 Concluding Observations
A survey of around 782 SMEs who are client of IDLC has been conducted to assess employment
generation by IDLC assisted SMEs. Some key findings are:
1. Growth in employment generation in IDLC assisted SMEs were large – over 100 per cent
when the time frame of inception to 2019 is considered. Employment generation in
female-owned sample enterprises was significantly higher--40.5 percentage points above
the overall average increase in employment for the whole sample. When compared
between male and female owned enterprises, the employment growth in female-owned
sample enterprises was 45.7 percentage points above the growth rate recorded in male-
owned sample enterprises.
2. The analysis provides strong support for more proactive engagements with female-
owned enterprises. Since female owned enterprises account for a small fraction of the
IDLC client base (8.4%), this is an area where more focus may be given by IDLC in
developing future loan portfolio if employment generation is an important focus of the
organization.
3. Notwithstanding the growth in employment, it is important to note that the average
employment size for IDLC enterprises stood at around 16 person per enterprise (i.e. 16
=>13,156/830). This is close to the average employment size by small enterprise of 18.4
reported in 2003 and about half of the average for micro and small enterprises based on
BBS survey at the national level. This finding suggests that average level of employment
in IDLC assisted enterprises is approaching a saturation point. Given the level of

29
 entrepreneurship and associated management skill and capacity, the level of investment
these firms are not going to grow very much in terms of employment generation.
4. Given the fact that the largest employment growth were recorded in the service (174.2%)
and manufacturing (131%) enterprises over the entire period (i.e. from inception to
2019), interventions in these sectors would probably be more rewarding in terms of
employment generation. This is in line with the general understanding at the national and
global levels that service, and manufacturing sectors create most jobs.
5. Job creation by types of jobs (namely family worker, salaried jobs, and day labourer)
depicts a positive picture. The largest growth rate has been found for salaried jobs with
134 per cent over the entire period (i.e. from inception to 2019). This form of employment
is more stable, creates predictable income stream for the workers. Dependency on day
laborers has been lower but still remained quite high, growing by 94.3% since the
inception to 2019. On a positive note, it is also noteworthy that dependence on family
workers have grown but at less than half the growth of employment in the sample group.
The second largest growth rate has been found for day labourer. The growth rate
was94.3%.
6. The capital intensity of business enterprises has increased significantly across Bangladesh
and IDLC borrowers were no exception. For enterprises supported by the IDLC borrowers
it used to cost about Tk 1 million to create 37 person-jobs in 2015. That was also the time
period when Bangladesh used to enjoy relatively higher elasticity of employment with
respect to output. However, as the elasticity declined and capital intensity or
capital/output ratio increased across all sectors, it started to cost much higher amounts
to create jobs. As a result, among the IDLC borrowers we found that only 8 jobs were
created with investment of Tk. 1 million in 2019. This is a matter of concern and may
reflect a host of issues including installation of labour replacing technology; increasing
labour productivity; or disbursement of credit to sectors which have low employment
intensities. To get informed answer, more information is needed.
7. Analyses on regional variations suggest the Chottogram region turned out to be the
leading region in terms of employment generation and turnover among the four regions
considered in the study. The turnover in Chottogram is almost double that of other three
regions. Due to limitations of questions and samples, it was not possible to find out the
factors behind the better performance of the MSM enterprises in the Chottogram region.
Despite these limitations, from a lender perspective, this finding may help IDLC to review
their regional credit portfolios and therefore act accordingly as how to improve
employment generation as well as profitability. However, due to limitations of questions
and samples, it was not possible to find out the factors behind the better performance of
the MSM enterprises in the Chottogram region.
8. One positive noteworthy development was that IDLC borrowers were able to expand their
sales/business turnover quite rapidly in recent years. More than doubling of turnover in
the 4-year period through 2019 was a very healthy development for the financial position

30
 of the enterprises, indicated enhanced capacity to service their debt and also to
undertake new borrowing for business expansion. These were signs for maintaining a
healthy portfolio with limited loan loss and pointed to further expansion of IDLC MSME
portfolio with limited risk.
9. The positive findings noted above are based on Pre-covid-19 data. Since March 2020, with
the spread of corona virus and the resulting lockdown, the near-term and medium-term
outlook may have changed significantly. Many of the IDLC supported enterprises are in all
likelihood going through a very difficult time and struggling to keep their businesses
afloat. A significant part of the job growth might have been reversed as most MSMEs are
yet to come back to their normal levels of operation.
10. This is an extremely trying time for both the MSMEs and IDLC itself, like all other lending
institutions in Bangladesh. Global experience indicates that many of the MSMEs will not
survive this pandemic and may have to go out of business. This will also impact the
portfolio quality of IDLC, like all other financial institutions in Bangladesh. At the same
time financial institutions including IDLC would need to help the MSME clients to help
them get out of the difficult situation. Most MSMEs will need rapid disbursing financial
support from the lending institutions, including from IDLC. Bangladesh Bank has allocated
certain amount of funds for disbursement to the affected MSMEs and is also providing
liquidity support to the lending institutions. It would be important that IDLC avails this
opportunity to extend financial support to the MSMEs and in the process also help itself.
In this extremely difficult time, all stakeholders must work together in their mutual
interest and in the interest of the country.

31
Annex 1:
ছ োট ও মোঝোরী প্রতিষ্ঠোনের ঋে প্রোতিনি কমমসংস্থোে সৃ তির প্রভোব
ছ োট ও মোঝোরী প্রতিষ্ঠোে জতরপ প্রশ্নপত্র
অংশ-১: ছ োট ও মোঝোরী প্রতিষ্ঠোনের বৈতশষ্ট

১। প্রতিষ্ঠোনের পতরতিতি
ক. প্রতিষ্ঠোনের েোমসহ উন্যোগিোর েোম ছকোড
খ. ছজলো
গ. উপনজলো/থোেো
ঘ. পল্লী/শহর পল্লী–১, শহর–২
ঙ. প্রতিষ্ঠোনের ধরে বযবসো/তবক্রয় ছসবো–১, ছসবো–২, প্রস্তুিকোরী প্রতিষ্ঠোে–৩,
কৃতিকোজ/উৎপো্ে–৪
ি. প্রতিষ্ঠোনের েমু েো েম্বর

২। প্রতিষ্ঠোনের ঠিকোেোাঃ
ছ োট ও মোঝোরী প্রতিষ্ঠোনের ঠিকোেো:……………………………………………………………………………………………
ছ োে েম্বর:……………………………………, ই-ছমইল েং:………………………………………………………………
৩। প্রতিষ্ঠোনের অবস্থোোঃ
১. ইতপনজড এলোকো, ২. তশল্প এলোকো/তবতসক এলোকোর মনধয, ৩. উভয় এলোকোর বোতহনর

৪। প্রতিষ্ঠোনের কমমকোন্ড অথমোৎ পণ্য উৎপো্ে বো ছসবোর ধরে (তবস্তোতরি তলখু ে):
.....................................................................................................................................................................................................
.....................................................................................................................................................................................................
ছকোড

প্রশ্নপত্র পূ রণ পদ্ধতি ও উনেখন োগ্য তৈষনের সংঙ্গো/ধোরেো

* প্রতিটি প্রনশ্নর সঠিক উত্তনর টিক তিহ্ন ( ) ত্ে।

* সব সংখযো ইনরজীনি তলখু ে, ছেমে বয়স 53
উনেখন োগ্য তৈষনের সংঙ্গো/ধোরেোাঃ
1. ছসৈো খোিাঃ ছহোনটল-ছরিুনরন্ট, েোেবোহে, েোত্রী পতরবহে, হোউতজং, তশক্ষো, স্বোস্থয, সব বযতিগি পতরিিমো ও সোতভম নসস, তবিোর আইে-আ্োলি এবং
অেযোেয েোবিীয় বযতিগি সোতভম নসস ছসবো খোনির অন্তভুমি হনব। উনল্লখয এই জতরনপ তবক্রয়/ছেতডং একটি আলো্ো ছসবো খোি তহসোনব ছেণ্ীভুি
করো হনয়ন ।
2. স্থোেী সম্পদাঃ স্থোয়ী সম্প্ বলনি জতম, তবতডং ও অেযোেয স্থোপেো, ছমতসেোতরজ ও েন্ত্রপোতি, েোেবোহে, আসবোবপত্র, ইতিযোত্র মু লযনক বু ঝোয় এবং
ছসবোর ছে সমস্ত তজতেিপত্র ১ (এক) ব নরর অতধক কোল পণ্য ও উৎপো্ে ছসবো প্র্োনে বযবহৃি হনব িোহোই স্থোয়ী সম্প্। ছ োট খোট টুলস ছক স্থোয়ী
সম্প্ তহসোনব ধরনি হনব েো।
3. তৈতেনেোগ্াঃ প্রতিষ্ঠোনের স্থোয়ী সম্প্ এর ওয়োতকম ং মু লধে তহসোনব ছমোট েি টোকো প্রতিষ্ঠোে/বযবসো পতরিোলেোয় লোগোনেো হনয়ন িোহোই তবতেনয়োগ
তহসোনব বযবহৃি হনব।
4. ৈু কতকতপং ও ৈযোনেন্স তসটাঃ প্রতিষ্ঠোনের েোবিীয় তহসোব ত্ে, মোস ও ব র ওয়োরী স্টোন্ডমোড তেয়মোেু েোয়ী তলতপবব্ধ করোনক বু কতকতপং বনল।
প্রতিষ্ঠোনের সোরো ব নরর আয়-বযোয় ও লোভ-ক্ষতির েোবিীয় তহসোব তিরী করোনক বযোনলন্স তসট বনল।

32
 5. ৈোৎসতরক টোেন ওভোরাঃ ছকোে প্রতিষ্ঠোনের সোরো ব নরর উৎপোত্ি পণ্য বো ছসবোর তবক্রনয়র মোধযনম অতজমি আয় ছক বোৎসতরক টোেমওভোর বনল।
ইহো পনণ্যর/ছসবোর তবতক্রি মূ লয হইনি সকল ছসবো বো্ ত্নয় েো থোনক িো েয়।
6. ইেপু ট ছশেোরাঃ পণ্য উৎপো্নের বযবহৃি তবতভন্ন কোাঁিোমোল বো তজতেিপনত্রর খরনির পতরমোে ছমোট উৎপোত্ি পণ্য/ছসবোর মূ নলযর শিকরো কি
ভোগ িোনক বু ঝোয়। এখোনে মনে রোখনি হনব, কমমরি ছলোনকর ছবিে-ভোিোর খরি ইেপু ট ছশয়োনরর মনধয হনব েো। িনব তবদ্যযৎ, পোতে, অত স খরি,
বযোংনকর সু ্, বোড়ীভোড়ো ইিযোত্ ইেপু ট ছশয়োনরর মনধয হনব।
7. ছপইড কমমাঃ মোতসক/সোিোতহক বো ত্তেক কোনজর জেয ছবিে ও মজু রীর তবতেমনয় তেনয়োতজি ছলোক ছক ছপইড কমী বনল।
8. দক্ষ কমমাঃ প্রতিষ্ঠোে/বযবসোর ছকোে কোনজ বো ছপশোয় ছবশ অতভজ্ঞ এবং তক ু ব নরর অতভজ্ঞিো থোকনল এসব তেনয়োতজি ছলোক ্ক্ষ কমী তহসোনব
বযবহৃি হনব। এ োড়োও মোতলক ভোল কনর জোনেে ্ক্ষ কমী কোরো।
9. পোতরবোতরক কমী বলনি তবেো ছবিে কমমরি ছলোক। িোাঁরো সোবমক্ষতেক ও খন্ডকোতলে কোনজ তেনয়োতজি থোকনি পোনর।
10. উন্যোগিো/মোতলক পোতরবোতরক কমী। তিতে বযবস্থোপক তহসোনব প্রতি মোনস তক ু সম্মোেী তহসোনব তেনলও পোতরবোতরক কমী তহসোনব গণ্য হনবে।

গ্ণেোকোরীর েোম ও উত্তরদোিোর/মোতেনকর েোম ও

স্বোক্ষর:...................................................... স্বোক্ষর/স্বীে:....................................................
িোতরখ:...................................................... িোতরখ:....................................................
৫। প্রতিষ্ঠোনের মোতলনকর ধরে
১. একক বযতি মোতলকোেো ২. অংশী্োতরত্ব ৩. প্রোইনভট তল: ছকো: ৪. সমবোয় ৫. অেযোেয (তলখু ে)
................................

৬। প্রতিষ্ঠোনের উৎপোত্ি পণ্য/ছসবোর বোজোরজোি করনে ধরে

১. অভযন্তরীে বোজোর ২. রিোতে ৩. রিোতে ও অভযন্তরীে ৪. অেযোেয (তলখু ে) ................................

বয়সাঃ তলঙ্গাঃ পু রুি–১, মতহলো–২

৭। উন্যোগিোর বয়স, তলঙ্গ এবং তশক্ষোগি ছেোগযিো

৮। তশক্ষোগি ছেোগযিোাঃ
১. প্রোথতমক ২. মোধযতমক পেমন্ত ৩. উচ্চ মোধযতমক অথবো সমমোে ৪. তডগ্রী ও িদ্যদ্ধ ৫. কোতরগতর তশক্ষো

৯। বযবসো আরনের সে
সে

১০। বযবস্থো আরনের ধরে তক?

১. পোতরবোতরক বযবসো ত ল ২. তেজ উন্যোনগ বযবসো ৩. িলমোে বযবসো ৪. অেযোেয (তলখু ে) ................................

১১। প্রতিষ্ঠোনের স্থোয়ী সম্প্ ও ছমোট তবতেনয়োনগর পতরমোে কি?

১. স্থোয়ী সম্প্.................................................. ২. ছমোট তবতেনয়োনগর পতরমোে..................................................

33
১২। বযবসো আরনের সময় অথমোয়নের উৎসাঃ
১. তেজস্ব অথমোয়ে ২. বযতি ছলোে/ধোর ৩. আইতডএলতস ঋে ৪. অেযোেয বযোংক ঋে

১৩। বযবসো বোব্ তবতভন্ন উৎস হনি ছেয়ো সবমনমোট ঋনের পতরমোে কি?
টোকোাঃ..........................................................

১৪। আইতডএলতস হনি ঋে ছেয়োর পতরমোে কি?

১. ১০ লক্ষ টোকোর তেনি ২. ১০-২৫ লক্ষ ৩. ২৬-৫০ লক্ষ ৪. ৫১-৭৫ লক্ষ ৫. ৭৫ লক্ষ টোকোর উপনর

১৫। আইতডএলতস হনি কিবোর ঋে ছেয়ো হনয়ত ল?

১. একবোর ২. দ্যই বোর ৩. দ্যই এর অতধক

১৬। বযবসো বৃ তদ্ধ বো বযবস্থোপেো উন্নয়নের জেয আনরো তক ঋনের প্রনয়োজে আন ?

১. হযোাঁ ২. েো

১৭। হযোাঁ হনল, ঋে পোওয়োর জেয আপেোর প ন্দেীয় উৎসগুতল তক

১. আইতডএলতস ২. সরকোতর বযোংক ৩. প্রোইনভট বযোংক ৪. তলতজং ছকোম্পোেী ৫. এমএ আই ৬. অেযোেয

১৮। প্রতিষ্ঠোে পতরিোলেোয় তক ছকোম্পোেী আইে ছমনে িনল?

১. হযোাঁ ২. েো

34
Annex 2: LFS – employment growth by sector and self-
employment category
The total employment in each of the chosen sector has been calculated by aggregating the
number of individuals aged 15 years and above employed in each of the components that make
up every sector according to the Gross Domestic Product report published by Bangladesh Bureau
of Statistics. There has been varying rates of growth in employment across the three sectors. The
agro-business industry has experienced the least amount of growth with the employment rates
steadily declining between 2013-2017. This could be an effect of the change in the sectoral
compositions with a shift towards manufacturing and service sectors. In essence, as Figure 20
depicts that the growth in employment in the manufacturing sector has gradually increased
between 2002-2003 and 2013 from 5.41% to 13.93% respectively. Employment in the service
sector has experienced a moderate growth. Unlike the other two sectors, in 2013 the growth rate
of employment in the service sector was negative at 0.87% but then continued to increase in the
other two periods. Between 2016-2017, employment in the service sector grew at almost the
same rate as that for the manufacturing sector.
Figure 20: Employment growth across sectors (%)

Figure 21: Growth rate of self-employment (%)

35
Changes in entrepreneurship in terms of growth is relatively moderate in Bangladesh over the
period of 2000-2017. Between 1999-2000 and 2002-2003, self-employment grew by 3.02% and
till 2013 the growth rates declined and adjusted relatively slowly. However, 2013 onwards self-
employment has grown steadily from 2.41% in 2013 to 4.7% in 2016-2017.
The changes in self-employment over the period of analysis can be conducted separately for
different sectors, urban and rural areas as well as across sexes. Similar to the scenario of total
employment in Bangladesh, females are less engaged in entrepreneurial activities than males
(the difference being significantly large).

36