Masato Abe, PhD

Economic Affairs Officer

Trade and Investment Division

United Nations ESCAP

Bangkok, Thailand

Multi-dimensional Review of Myanmar

Myanmar Business Survey:

Objectives, Methodology & Results

Outline

• Overview of the Multi-dimensional Review

• Business survey

– Research design

– Methodology

– Implementation

• A guest speaker

• (Select results)

My background

• Merchant / farmer’s family

• Marketing, business administration & economics

• Japanese, Singaporean & US businesses

• Sales, marketing, supply chain management, logistics & procurement

• UN; Asia-Pacific, Southeast Asia

• New Myanmar SME Law

1. Overview of the Multi-dimensional Review

Objectives

• Identify the binding constraints in various fields to development in Myanmar

• Suggest a strategy with specific policy recommendations

• Propose action plan

– National authorities

– Economic and social actors

– International donors

Research phases

2012-13 2013-2015 Present Future

Biz Survey

Research team

• ESCAP – Trade and Investment Division, Business and Development

Section

• OECD (lead organization) – Development Centre

• UMFCCI – Myanmar Business Survey Team

• MNPED – Focal-point ministry

• Cooperating agencies – The Asia Foundation, GIZ, Japan, Mekong Institute, South

Korea, Switzerland, UNIDO – Others

Key topics

• Macro economic stability • Human resources • Industrial transformation

– Agriculture – Manufacturing – Services

• Business environment • Financing for development • Social development • Education • Others

Research method • Data collection

– Non-structured qualitative interviews with various stakeholders • Government • Business • Academia • NGOs • International donors

– Statistical data

• Literature review – Multi languages – Burmese

• Team discussion • Cross examination

– Follow-up interviews

• Team and individual missions – Approx. 300 person days

Research outputs

• OECD Development Pathways: Multi-dimensional Review of Myanmar – Volume 1. Initial Assessment (2013)

– Volume 2. In-depth Analysis and Recommendations (2015)

• Myanmar Business Survey – Myanmar Business Survey 2014: Survey Results (2014)

– A comprehensive version (forthcoming)

• Other ESCAP studies – Business and Development in Myanmar (2015)

– AEC Business Handbook (forthcoming)

URL addresses

• http://www.keepeek.com/Digital-Asset-Management/oecd/development/multi-dimensional-review-of-myanmar_9789264202085-en#page1

• http://www.keepeek.com/Digital-Asset-Management/oecd/development/multi-dimensional-review-of-myanmar_9789264220577-en#page1

• http://ebac.unescap.org/news/release-myanmar-business-survey-2014-results

2. Business Survey

• First-ever national wide survey for business

• All sectors, sizes, legal structures, ownerships , nationalities and States/Provinces

• Conducted by ESCAP/OECD/UMFCCI

• Donors: Japan, Germany, Switzerland, The Asia Foundation, UNIDO & Mekong Institute

• 3 months for data collection

Background & project objectives

• Lack information on business conditions and environment

– Not much information in 2012-13

• Assess the status of the business community

• Identify major obstacle in identifying effective policy recommendations

Seven key issues

1. Business enabling environment • Legal regulatory framework

• Infrastructure

2. Entrepreneurship (partially covered)

3. Access to finance

4. Human resources

5. Innovation and technology

6. Business development services (partially covered)

7. Market & resource access

Differences from other business surveys

• WB/IFC – Business enabling environment (BEE) – Enterprise survey (trade/market access focus)

• 650 samples • Missing data • Descriptive

• German – Smaller scale – Geographical limitation – Narrower focus

• CSO/MOI – Annual surveys for industrial estates under MOI

• JETRO (and IDE) – Small scale and geographical limitation

• Other academic surveys

What is a business survey?

• A detailed study of a business community, market or geographical area to gather data on attitudes, impressions, opinions, behaivours or satisfaction level

• Polling a section of the population

• Generalization and inference

Why do we conduct a survey?

• Collecting data could be expensive and time consuming

• It is often impossible to observe all subjects in the entire population (i.e. what we call a census)

• Rather a researcher will usually study a small sample, sometimes involving complex survey method(s) to help cut down on the expenses and time

The foundations of the business survey

• Target population => Myanmar business • Sampling frame

– List of all the elements in the population with some chance of being selected

• Sampling methods • Sampling instruments

– Interview • Structured, semi-structured or non-structured

– Questionnaire – Others

• Cost and time

Inference using survey method

Target population and sampling frame

Myanmar business

Target population

• Cover all types, size, ownership and nationality of businesses in Myanmar – Foreign enterprises in Myanmar

– State owned enterprises

– Military enterprises

– Large enterprises

– Joint ventures/partnerships

– Small and medium enterprises

– Micro enterprises

– Cooperatives

– Informal sector

Structure of Business Sector

Foreign Enterprises

State-owned Enterprises

Military Enterprises

JV

JV

JV

JV

Biz Associations

Myanmar Enterprises

JV

JV

Privatization

Partnerships

Cooperatives

Target sectors

• Agribusinesses (but not subsistence farmers)

• Manufacturing

• Services

• Trading

• Also cover various sub-sectors in each of the sectors

Constitution of enterprises

Source: IFC (2009) Self-employed

Informal Sector

Most businesses are small or informal

Number Percentage

Large enterprises

721 0.6%

SMEs

126,237 99.4%

Total

126,958 100.0%

Informal sector

620,000 83% of entire

business

Source: President’s Office (Nay Pyi Taw News, 14 January 2013)

SME density is low

Source: OECD , 2013.

Different

types of

businesses

in Myanmar

Type Number Per cent Registration /

supervision

Legal framework

State-owned

enterprises

632 0.6% Various governmental

offices

State-owned Economic Enterprises Law (1989)

Myanmar enterprises 30,135 28.5% MNPED/line ministries Myanmar Companies Act (1914) (and Companies

Rules (1940); Companies Regulations (1957));

Myanmar Citizens Investment Law (1994)

Foreign enterprises 1,686 1.6% MNPED/line ministries Myanmar Companies Act (1914) (and Companies

Rules (1940); Companies Regulations (1957));

Foreign Investment Law (2012)

Partnerships 1,072 1.0% MNPED/line ministries

Development

Partnership Act (1932)

Joint ventures

between state owned

enterprises and local

enterprises

163 0.2% MNPED/line ministries Myanmar Companies Act (1914) (and Companies

Rules (1940); Companies Regulations (1957));

Special Company Act (1950); Myanmar Citizens

Investment Law (1994); Foreign Investment Law

(2012)

Business associations 57 0.1% MNPED/MOC Myanmar Companies Act (1914) (and Companies

Rules (1940); Companies Regulations (1957)); Law

relating to Forming of Organizations—NGO Law

(1988)

Large sized

manufacturing

enterprises

4,808 4.6% MOI Burma Companies Act (1914); Private Industrial

Enterprise Law (1990); Special Company Act (1950)

Manufacturing SMEs 38,424 36.4% MOI Myanmar Companies Act (1914); Private Industrial

Enterprise Law (1990); Special Company Act (1950)

Micro sized

enterprises

13,309 12.6% Ministry of Cooperatives Promotion of Cottage Industries Law (1991); Law

Amending the Promotion of Cottage Industries Law

(2011)

Cooperatives in the

primary sector

9,581 9.1% Ministry of Cooperatives Cooperative Society Law (1992)

Hotels and tourism

enterprises

5,790 5.5% MNPED/MOHT Myanmar Hotels and Tourism Law (1993)

Total 105,657 100.0%

Source: Various public

sources

Smaller food and beverage firms dominate manufacturing in Myanmar

0

5000

10000

15000

20000

25000

Food a

nd beve

rage

s

Constru

ctio

n mate

rials

Cloth

ing

and w

earin

g ap

parel

Meta

l and m

inera

l

Perso

nal g

oods

Raw m

ater

ials

Liter

ature

and a

rts

Consum

er p

roduct

s

Autom

obiles

Indust

rial t

ools an

d equip

ment

Agric

ulture

mac

hinery

Elect

rical

equip

ment

Oth

ers

Large

Medium

Small

Source: Compiled by the author based on the data of CSO (2013).

The Government has developed regional industrial zones & special economic zones as the main development strategy for Myanmar’s industrialization

Mingaladon has been the most successful industrial zone in Myanmar (2009-10)

0

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

4,500

Min

gala

rdo

n

Mye

ik

Hla

ingt

har

yar

Dag

on

seik

kan

Shw

epyi

thar

Dag

on

myo

thit

(Ea

st)

Shw

e P

ou

k K

an

Thar

kayt

a

No

rth

Okk

alap

a

Sou

th D

ogo

n

Yan

gon

Wes

t

Man

dal

ay

Sou

th O

kkal

apa

Pat

hei

n

Mya

un

gmya

Mo

nyw

a

Mei

ktila

Pya

y

Myi

ngy

an

Yan

gon

So

uth

Yen

angy

aun

g

Maw

lam

yin

e

Pak

okk

u

Hin

thad

a

Tau

ngg

yi

Kal

e

0

100

200

300

400

500

600

Sales per firm Gross profit per firm Employee per firm

Source: Compiled by the author based on the data from CSO (2014).

Issues

• We have some broad idea about the status of business in Myanmar

• However, we do not have a comprehensive and reliable list to develop a sampling frame

• Unknown geographical distribution

• Fragmented dataset – UMFCCI memberships

– Other business associations’ memberships

– Industry directories; Yellow Pages

– Government registrations, licenses or permits • Various line ministries

– Population of the States and Regions

– Local information

Questionnaire

GIGO

• Garbage In Garbage Out (GIGO) – Quality of questionnaire is a key success factor

• Topics & questions

• Statistics – Descriptive

– Inference

– Multivariate or advance method

– Data collection • Method

• Instruments

• Management

– Data entry

Two crucial issues

Missing data

&

Response rate

Questionnaire development I • Total design method for surveys (Dillman, 1978)

• Consulted with a number of industrial and country experts

• Reviewed many existing business survey questionnaires – e.g. JETRO, 2009 and 2012; OECD, undated; World

Bank, 2013

• Questions selected or modified from existing business surveys or newly developed – e.g. Abe et al., 2012; DCED, 2008; IFC, 2008; ILO, 2003;

JICA, 2006; MEF, 2005; Transparency International, 2014; WEF, 2013; World Bank, 2013

Questionnaire development II

• Fitted with advanced statistical techniques – Multivariate statistics

• Conducted forward-backward translation – Two professional English and Burmese translators – Two separate questionnaires in Burmese and English

• Pre-tested with 13 business executives in Yangon • Contained 73 questions

– Often use six-point Likert scale instruments

• Designed and printed in a professional looking format – 20 pages in A4 – Introductory statement signed by three agencies

Difficult questions

• Examples

– Turnover

– Profitability

– Corruption

Questionnaire (English) is available at:

http://ebac.unescap.org/news/release-myanmar-business-survey-2014-

results

Sampling methods

Good sample: Overview

• Unbiasedness

• Efficiency

• Consistency

• Sufficiency

Good sampling

Aczel (1993), Figure 5-1 P.167

Bad sampling

Aczel (1993), Figure 5-1 P.167

Biasedness

Aczel (1993), Figure 5-10 P.184

Unbiasedness

Aczel (1993), Figure 5-9 P. 184

Normal and sampling distributions

Normal distribution

Sampling distributions

Good sample

• Free from selection bias as best as possible

• Don’t have non-response bias, over- and under-representation, survivorship bias, etc.

• Have accurate responses to the items of interest

• Good questionnaire design is important in getting accurate information

• Distinguish between sampling and nonsampling errors

How can we conduct good sampling?

Best solution is to randomly choose samples

from the sampling frame which represents the target population

adequately

Probability Theory

Sampling methods

• Probability sample

– Simple random sample

– Stratified random sample

– Cluster sample

– Systematic sample

• Non-probability sample

– Convenience sample

– Snowballing sample

– Conference sample

Simple random sample (SRS)

• The most basic type of sample

• A sample in which observations are “independently” sampled without replacement from a population

• In an SRS, each possible sample of the population of size n has an equal probability of being chosen as the sample

• Designed to be avoid sample bias

Stratified random sample

• An advanced method to SRS • Done by dividing members of the population into

homogeneous subgroups before random sampling • Each subgroup should be mutually exclusive (every

element in the population must be assigned to only one subgroup) and collectively exhaustive (no population element can be excluded)

• Stratified samples often help improve the representativeness of the sample by reducing sampling error

• Thus, its main objective is to increase precision

Cluster sample • A sampling technique often used when a “natural” but

relatively homogeneous subgroupings are evident in a statistical population

• The total population is divided into such subgroups (or clusters) and a simple random sample of subgroups or clusters is selected

• A common motivation for cluster sampling is to reduce the average cost per interview

• Given a limited budget, this can allow an increased sample size (e.g. reduced logistical cost)

• Why use cluster samples? – Constructing a sampling frame list of observation units may be

difficult, expensive or impossible – The population may be widely distributed geographically or may occur

in natural clusters such as households or schools, and it is less expensive to take a sample of clusters rather than an SRS of individuals

Stratified sample vs. cluster sample

Stratified sample

Cluster sample

Heterogeneous subgroups

Homogeneous subgroups

Two-stage cluster sample

Systematic sample

• A statistical method involving the selection of elements from an ordered sampling frame

• The sampling often starts by selecting an element from the list at random and then every kth element in the frame is selected, where k is the sampling interval (sometimes known as the skip)

• Systematic sampling is to be applied only if the given population is logically homogeneous, because systematic sample units are uniformly distributed over the population

• Reliable sampling frame is must

Non-probability samples

• Convenience sample

• Snowballing sample

• Conference sample

• Membership sample

• Others

Survey instruments

• Face-to-face interview

• Telephone interview

• Questionnaires

– Mail

– Internet

– Telephone

– Fax

– Drop-off

• Mixed instruments

• Follow-up

High

Cost

Low

F2F Interview

Mail, fax,

telephone

Email, SurveyMonkey

Wide Sampling coverage Narrow

Coverage vs. cost

Conference/drop-off

Tele-interview

Our sampling method

• Lack of a sampling frame with unknown population – No nation-wide business registration/licensing system – No reliable business/industry directories – Out-of-dated lists of business association memberships

• Sampling methods conducted – To maximize the number of responses and the accuracy of the

survey • Convenience sample • Snowballing sample • Conference sample

• Pre-appointed on-site interviews by trained personnel – 57 minutes on average

• Prepared interview guides – Updated as the survey project was progressed

• Held press conferences and meetings with business associations to seek cooperation

Field works

• Networking

• Team building

• Identification

• Approach

• Persuasion

• Connection

• Communications

• Follow-up

• In-kind support/fund raising

Interview techniques

• Find a prospective informant

• Make an appointment

• Convince him or her to participate in the survey by telling the importance of the project

• Provide benefits

• Develop a bond with the informant

• Avoid missing answers

• Take strategies fit with your character (important)

• Speak with your language (important)

• Manage time

• Start from easy targets

Management issues I

• Logistics

– Cost and time

• Record of interview

• Confidentiality

– Informants/participating companies

• Intellectual property right

– OECD/ESCAP/UMFCCI

Management issues II

• No duplication

• Related data

– Formal system

– Informal system

• Data entry

– Entry sheet

– Procedures

• Journal

– For future activities/projects

Management issues III

• Performance

– Quantity vs. quality

– Monitoring

– Evaluation

– Corrective actions

• Cheating by interviewers

– Statistical screening

– Outliers

– Illogical patterns

Quantitative target

• Approximately 500 duly filled questionnaires per Area Manager (6 of them)

• By mid March

– Daily target

– Weekly target

– Monthly target

• How many interviewers do you need?

• Geographical target

– Townships, areas and industrial estates

• Training needed

Very large target samples: 2,500 to 3,000 businesses

covering all major commercial hubs

Draft sampling plan

No. State/Region Suggested cities Suggested Sample Population

1. Kachin Myitkyina, Bhamo 75

2. Kayah Loi Kaw 30

3. Kayin Myawaddy, Hpa-an 50

4. Chin Hakha, Falam 50

5. Yangon Yangon, 700

6. Bago Bago, Pyay 200

7. Mandalay Mandalay, Pyin Oo Lwin 500

8. Sagaing Monywa, Sagaing 200

9. Mon Mawlamyaing 100

10. Shan Taunggyi, Lashio, Tachilek, Muse 200

11. Ayeyarwaddy Pathein, others 100

12. Magway Magway, others 100

13. Rakhine Sittwe, Kyaukphyu 75

14. Tanintharyi Dawei, Kawthoung 100

15. Naypyitaw Naypyitaw 50

Total 2530

Chin State Hakha

Falam

Sagaing Kalay

Region Tamu

Sagaing

Monywa

Myin Gyan

Kachin Myitkyina

State Bhamo

Shan Lashio

State Muse

Taunggyi

Inlay

Tachilek

Kengtung

Bago Bago

Region Pyay

Magway Magway

Region

Nay Pyi Daw Nay Pyi Daw

Mandalay Mandalay

Region Naung - U

Pyinoolwin

Yangon Region Yangon

Ayeyarwaddy Region

Kayah State

Kayin State Myawaddy

Hpa-an

Mon State Mawlamyaing

Rakhine State Sittwe

Kyaukphyu

Tanintharyi Region Dawei

Kawthoung

10

10

18

17

16

15

14

13

12

11

20

21

22

23

24

25

26

27

29

29

11

12

13

14

15

16

1718

20

2321

22

24

26

27

28

25

28

30

30

31

32

32

33

1

2

3

4

5

6

7

8

9

8

9

3

4

5

6

7

1

2

Pathein

Hlegu

33

34

34

33

Taungoo

19

19

Loikaw

Survey sites

All 14 States/Regions and Nay Phi Taw; 34 cities/towns in total

UMFCCI survey team

The roles of the project members

• Executive Committee Members

• Executive Director

• Project Manager

• Project Assistant

• Area Managers

• Team leaders

• Team Members

• Data Entry Personnel

• ESCAP/OECD Advisers

Two workshops for managers

• Lectures

• Two major exercises for interview in both Burmese and English

– Made an appointment with a Burmese business person for real interviews

• Individual/group works

• In-class discussions

Recruitment of right people

Empowerment of team members

Develop local strategies

Leadership

3. Select survey results

DESCRIPTIVE STATISTICS

• 3055 samples

• Agriculture & mining (17.7%); manufacturing (33.1%); services (49.2%)

• Average 40 employees (from 1 to 4000)

• MSMEs (97.8%)

• Informal sector (14.4%)

• Foreign companies (3.0%)

• 14 years old on average

• Female (29.7%); male (70.3%)

• Exporting (22.5%); importing (38.9%)

Low level of missing data!

THE SURVEY PRODUCES A WELL-BALANCED REPRESENTATION

The number of respondents by sector

0

200

400

600

800

1000

1200

Agriculture, forestryand fishing

Extractive industries Manufacturing Trade Hotels and restaurants Other services

ENTERPRISE SIZES ARE SMALL IN MYANMAR

The number of respondents by size category

0

200

400

600

800

1000

1200

1400

1600

Micro (1-9 employees) Small (10 -49 employees) Medium (50 - 249employees)

Large (250+ employees)

MOST FIRMS IN THE SAMPLE HAVE LONG BEEN ESTABLISHED

The number of respondents by firm age

0

200

400

600

800

1000

1200

1400

1 year 2 to 4 years 5 to 9 years 10 to 14 years 15+ years

CORRUPTION, LACK OF SKILLED LABOUR AND

OF TECHNOLOGY

CONSTITUTE MAJOR BARRIERS FOR BUSINESSES

0%

5%

10%

15%

20%

25%

The percentage of respondents who see the constraint as a very severe obstacle

SIX KEY ISSUES

1) Access to markets, labours, supplies and technologies

2) Financing

3) Regulations and taxation

4) Infrastructures and utilities

5) Conditions for international business

6) Corruption

BUSINESS SENTIMENT: STATE/REGION

-80

-60

-40

-20

0

20

40

60

80

100

ACCESS TO MARKET, LABOUR, SUPPPLY & TECHNOLOGY

-80

-60

-40

-20

0

20

40

60

80

FINANCING

-80

-60

-40

-20

0

20

40

60

80

100

120

CORRUPTION

-100

-50

0

50

100

150

KEY ISSUES: RELATIVE SIGNIFICANCES

-60

-40

-20

0

20

40

60

80

Internationalbusiness

Infrastructures &utilities

Regulations &taxation

Access tomarkets &resources

Financing Corruption

BUSINESS SENTIMENT: SIZE

AEC 2015 • Two third of samples were not aware of the

ASEAN Economic Community (AEC) 2015

MOST FIRMS DO NOT SPEND ON RESEARCH AND DEVELOPMENT

The share of respondents by the size of R&D spending

0%

10%

20%

30%

40%

50%

60%

None 50lakhs

or less

51 -100

lakhs

101 -500

lakhs

501 -1,000lakhs

1,001 -2,500lakhs

2,501 -5,000lakhs

5,001 -7,500lakhs

7,501 -10,000lakhs

Over10,000lakhs

INNOVATION

However, 60% of samples did not invest in R&D at all. Those which conducted R&D spent closely USD40,000 on average per year. Only one third of samples protected their intellectual properties mainly through patent applications.

HUMAN RESOURCES MANAGEMENT

• Increasing pressure to both employment and salary

• Shortage of skilled labours and

professionals – In-house training

– Little investment to HR • Over one-half: No fund allocated

• Average USD 215 per employee per year

– Use services of commercial training institutes

– Room for the Government

– AL, SL and medical support provided

– Few pension scheme

– ICT skills required

2%

27%

4%

32%

24%

9%

2%

Graduate

Undergraduate

Vocational diploma

High school

Middle school

Elementary school

No formal education

SKILL SHORTAGES PERSIST IN ALL SECTORS

The share of respondents that find skill shortages as a major obstacle to business

0%

10%

20%

30%

40%

50%

60%

70%

Total Agriculture, forestryand fishing

Extractive industries Manufacturing Trade Hotels and restaurants Other services

PAID LEAVE IS NOT YET COMMON PRACTICE, THOUGH

The share of respondents providing paid leave by the number of paid leave days

0%

5%

10%

15%

20%

25%

30%

None 3 days orless

4 - 6 days 7 - 10 days 11 - 15days

16 - 20days

Over 20days

THERE IS NO WAGE GAP BETWEEN MEN AND WOMEN

The percentage of people by extent of agreement with whether there is wage differential

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

21% orless

11 - 20%lower

1 - 10%lower

Nodifference

1 - 10%higher

11 - 20%higher

21% ormore

GENDER EQUALITY

SOURCES OF FINANCING

71%

10%

10%

3% 1% 1%

4%

Personal saving

Personal loans

Retained earnings

Commercial banks

State Development banks

Money lenders

Others

• More than one–half required more funds • Two-third do not borrow externally

• 40% of samples do not need external loan

FIRMS THAT HAVE ACCESS TO LOANS TEND TO BORROW SHORT TERM

The share of respondents with external debt by the size of their outstanding loans

0%

1%

2%

3%

4%

5%

6%

7%

8%

Less than 50lakhs

50 - 100 lakhs 101 - 500 lakhs 501 - 1,000lakhs

1,001 - 5,000lakhs

5,001 - 10,000lakhs

10,001 - 50,000lakhs

50,001 -100,000 lakhs

100,001 -500,000 lakhs

Over 500,000lakhs

Short-term loan (1 year or less) Mid to long-term loan (over 1 year)

COLLATERAL IS THE BIGGEST OBSTACLE FOR ACCESS TO EXTERNAL FINANCING

The relative degree of major obstacles for external financing

0

10

20

30

40

50

60

70

Collateral & guarantee Loan procedures Small loan size Interest rate Short loan term Credit rating Disclosurerequirements

A QUESTION TO ALL

-60

-40

-20

0

20

40

60

80

Internationalbusiness

Infrstructure &utilities

Regulations &taxation

Access to makrets& resources

Financing Corruption Overall businesssentiment

Is there any other hidden factor?

Entrepreneurship? Sales and profit?

Productivity?

Implications

• Crucial role of public corruption – Anti-corruption agenda

• National and sub-national development strategies

• Service development for the business sector

• Sectoral development strategies – Agriculture

– Mining

– Manufacturing

– Services

CONCLUSIONS

• More analysis required – Generalization – Sophisticated methods – Comparison with WB’s enterprise survey

• Policy recommendation to be developed – For the Government – For business with sector development strategies – For donors, international agencies and NGOs

• Export and FDI promotion strategies – Supply-side capacity building – Market intelligence – Networking