occasional paper
on
economic statistics
Measuring Services Economy
in Singapore:
Emphasis on Short-term Indicators
Singapore Department of Statistics
October 1998
This paper should not be quoted without the permission
of the Singapore Department of Statistics
Papers in this Occasional Paper Series are intended to provide an informal
means for the preliminary dissemination of ideas to stmulate research
discussion and analyses. The views expressed herein are preliminary and are
based on data presently available. Comments and suggestions are welcome.
MEASURING SERVICES ECONOMY IN SINGAPORE
EMPHASIS ON SHORT-TERM INDICATORS
I INTRODUCTION
1 Services and manufacturing are the twin engines of growth in the
Singapore economy. Accounting for two-thirds of GDP, the services sector
comprises a diverse mix of economic activities, ranging from traditional
services such as entreport, wholesale and retail trade in the commerce sector to
emerging services such as IT, communications and media, healthcare and
logistics.
2 Unlike traditional services, the emerging services clusters are not only
complex by nature, they also span across different industrial sectors, making
measurement difficult. With their rapidly increasing contribution to the
economy, the need for accurate and timely statistical indicators to assess their
growth and performance is widely recognized. The compilation of such
statistics poses tremendous challenges to official statisticians and raises
interesting and problematic conceptual issues.
3 This paper1 presents an overview of how the Singapore Department of
Statistics (DOS) measures the services economy in Singapore, with special
emphasis on short-term indicators for assessing the growth and performance of
the various services industries. Measuring services is inherently difficult.
Meeting the growing demand for timely and accurate short-term statistical
indicators of the performance of services industries is even more difficult,
aggravated by the constraints of limited resources and the avoidance of
excessive respondent burden. These issues are discussed in this paper.
II OVERVIEW OF SINGAPORE’S SERVICES ECONOMY
4 Comprising commerce, transport and communications, financial and
business services as well as community, social and personal services, the
services sector has increased its share of GDP from 63 per cent in 1988 to 67
per cent in 1997. Conversely, the manufacturing sector’s share of GDP has
declined from 29 per cent to 23 per cent. The share of the remaining sectors
(agriculture, mining and quarrying, utilities and construction) has remained
constant at around 10 per cent. During this period, GDP in current prices has
grown from S$52 billion in 1988 to S$143 billion in 1997.
1
An earlier version of this paper was presented at the 13th Voorburg Group Meeting on
Services Statistics held in Rome from 21 – 24 September 1998. The Voorburg Group was established
in 1986, after an initiative by Statistics Canada and the United Nations Statistical Office (UNSO).
The group focuses on both the conceptual and practical issues related to the measurement of services
statistics.
5 Traditional services industries are found mainly in commerce (entreport,
wholesale and retail trade, hotels and restaurants), and transport and
communications. While the commerce sector has maintained its share of GDP
at about 17 or 18 per cent, the share of transport and communications has
declined from 13 per cent in 1988 to 10 per cent in 1997 (Table 1).
TABLE 1: PERCENTAGE DISTRIBUTION OF GROSS DOMESTIC PRODUCT
AT CURRENT MARKET PRICES
1988 1997
Manufacturing 29 23
Construction 6 9
Commerce 17 18
Transport & Communications 13 10
Financial & Business Services 23 29
Financial Services 9 13
Business Services 14 16
Other Services 10 10
6 The sector with the largest proportion of non-traditional activities,
namely the financial and business services sector, is also the sector with the
largest increase in its share of GDP in the last decade (from 23 per cent in 1988
to 29 per cent in 1997). The financial services sub-sector has increased its
contribution to GDP from 9 per cent in 1988 to 13 per cent in 1997 (Table 1).
The growth of this sub-sector has previously been based on traditional banking
and financial activities including offshore banking through Asian Currency
Units (ACUs), stock broking, foreign exchange trading and insurance services.
However, the importance of new and non-traditional activities such as fund
management, derivative trading (futures, options, swaps), bond and security
trading has increased rapidly in recent years. On-going reform of the financial
services sub-sector coupled with the rapid pace of financial innovation will
ensure the importance of financial services in Singapore’s future economy.
7 Spurred by the emergence of new and emerging business services,
particularly IT-related and media services, the business services sub-sector has
also grown rapidly increasing its share of GDP from 14 per cent in 1988 to 16
per cent in 1997.
8 Two main factors account for the rapid growth of business services.
Firstly, as more manufacturing companies outsource activities that were
formerly done in-house, related business services such as legal counselling,
accounting, management consultancy, advertising and logistics have grown
rapidly.
9 Secondly, Singapore has been actively promoting high value-added,
knowledge-based and internationally exportable services since they have strong
2
growth potential. These include service clusters such as IT, communications
and media, logistics, education and healthcare.
10 For example, for the IT, communications and media cluster, both multi-
national corporations (MNCs) and promising local companies have been
encouraged to undertake high value-added activities in Singapore including
software product development, applications development, research and
electronic commerce. In the area of communications and media, companies are
encouraged to develop content, as well as to introduce new and innovative
services in Singapore and deliver them to the region.
III MEASURING SINGAPORE’S SERVICES ECONOMY
Comprehensive Annual Data
11 Comprehensive annual data are drawn from four major surveys
conducted by DOS and the Monetary Authority of Singapore (MAS). DOS
conducts the annual surveys on commerce, services and the international trade
in services while MAS conducts the survey of financial institutions. These
surveys collect comprehensive income and expenditure data, which are used as
inputs in the estimation of national accounts. A comparison of the data items
proposed in the model survey presented and discussed at Voorburg Group’s
meetings show that data items for most of the 11 modules of the model survey
are available from our annual surveys. This is particularly the case for
traditional services.
12 However, for new and emerging services, particularly the IT-related
services, information on selected modules are not available from our annual
surveys. Table 2, which presents the comparison for the IT sector, shows that
our annual surveys do not currently yield information on packaged software
revenue. DOS will improve our annual surveys to ensure the collection of
more comprehensive data pertaining to new industries such as the IT-related
industries and logistics. For example, questions on electronic-commerce (e-
commerce) have been included in the current round of surveys of commerce
and services for the first time.
3
TABLE 2: COMPARISON OF EXISTING DATA WITH ITEMS
PROPOSED IN THE MODEL SURVEY
Module Topic Availability
1 Revenues Available, Annual Survey of Services
2 Goods and services used in operations Available, Annual Survey of Services
3 Purchases of goods and services for resale Available, Annual Survey of Services
4 Inventories Available, Annual Survey of Services
5 Accounting procedures Available (accrual basis)
6 Exports Available, Trade in Services
7 Imports Available, Trade in Services
8 Packaged software revenues Not identified separately
9 Employment Available, Annual Survey of Services
10 Fixed assets, additions and disposals Available, Annual Survey of Services
11 Software research and development Partially available, Annual Survey of
Services
Inadequacy of Existing Industrial Classification
13 The inadequacy of existing industrial classifications to cater to new and
emerging economic activities is well recognised. Indeed, classification-related
issues are discussed at great length at Voorburg Group meetings. Economic
activity in Singapore national accounts is currently classified on the basis of the
Singapore Standard Industrial Classification 1990 (SSIC 90). Recognising the
inadequacy of SSIC 90, DOS has recently issued a revised and updated
classification, Singapore Standard Industrial Classification 1996 (SSIC 96,
Table 3).
14 The SSIC 96, which was adapted from the United Nations Industrial
Standard on Industrial Classification (ISIC) Rev 3, took into consideration
changes in Singapore’s industrial structure. Placing greater emphasis on
services, SSIC 96 comprises 17 sections, or nearly twice the 9 major divisions
in SSIC 90 (Table 3). Sections pertaining to services have been expanded, with
services previously classified within divisions and major groups upgraded to
the section level. SSIC 96 will be implemented in Singapore’s national
accounts late this year.
15 However, with the rapid development and increasing complexity of new
and emerging services, there is a growing recognition that the conventional
framework of industrial classifications (as represented by ISIC) impedes
attempts to statistically measure these new services, which span across several
different industries. An example is the logistics sector whose activities include
transportation, distribution, inventory control and communications. These
activities would be assigned to several different industrial classifications under
a conventional classification.
4
TABLE 3: COMPARISON OF SSIC 1990 AND SSIC 1996
SSIC 1990 SSIC 1996
Division Section
1 Agriculture, Fishing, Forestry & Hunting A Agriculture, Hunting and Forestry
B Fishing
2 Mining and Quarrying C Mining and Quarrying
3 Manufacturing D Manufacturing
4 Electricity, Gas and Water E Electricity, Gas and Water
5 Construction F Construction
6 Commerce G Wholesale and Retail Trade
H Hotels and Restaurants
7 Transport, Storage and Communications I Transport, Storage and Communications
8 Financial, Insurance, Real Estate and J Financial Intermediation
Business Services
K Real Estate, Renting and Business Services
9 Community, Social and Personal L Public Administration and Defence
Services
M Education
N Health and Social Work
O Other Community, Social and
Personal Services
P Domestic Work Activities
Q Extra-Territorial Organisations
16 An alternative framework is adopted by the recently introduced North
American Industrial Classification System (NAICS), which groups industries
according to differences in industry process, whereas ISIC groups industries
according to differences in industry output.
17 Thus, while the implementation of SSIC 96 in Singapore’s national
accounts will enable the presentation of statistical indicators for more detailed
services industries, DOS has begun to review, modify and expand SSIC 96.
Discussions with relevant government agencies will be held to ensure that the
revised SSIC has sufficient level of details to enable data to be recompiled
according to the appropriate clusters. In undertaking this revision (expected to
be completed in the year 2000), DOS will examine closely the NAICS with the
view to assessing its appropriateness for Singapore.
IV SHORT-TERM INDICATORS
18 The above shows that fairly comprehensive annual survey data are
available to measure Singapore’s services economy. These data would be
supplemented by the inclusion of additional data items, particularly those
suggested in the model survey that are not currently collected in our annual
surveys. In addition, there is also a need to revise our industrial classification
to ensure that new and emerging economic activities are adequately covered.
5
Modification to our surveys and revisions to our industrial classification are
required in order to develop a sound and reliable statistical framework for
measuring services in Singapore.
19 However, while the foundations for such a framework have been firmly
laid, there is the additional need for timely, accurate and reliable short-term
economic indicators to assess economic performance. Timeliness and scarce
resources dictate that the development of these indicators cannot be simply
based on survey returns.
Administrative Data
20 In developing these indicators, DOS has relied as far as possible on
administrative data which has the advantages of being available on a high
frequency (usually monthly), more comprehensive (usually almost complete
coverage) and less labour-intensive than surveys. A wide range of short-term
indicators (both volume as well as deflated value indicators) is derived from
administrative by-products arising from the operations of several government
agencies (see Annex 1). Some examples are:
a. Trade data
Singapore is the first country in the world to introduce a
nation-wide comprehensive electronic data interchange
system (TRADENET) for trade documentation.
TRADENET is highly interactive and processes trade
declarations within 30 minutes, with some under 5
minutes. The system facilitates international trade and
provides very comprehensive monthly information of our
external trade, which are then used to assess the
performance of entreport and wholesale and retail trade.
Trade data are normally available within one week from
the end of the reference period.
b. Tourism-related data
Taxes imposed on hotels and food & beverage
establishments as well as other administrative data from
the Singapore Tourism Board (STB) are generally
available about 4 weeks after the end of the reference
period. Monthly data on room days occupied, visitor
arrivals, average occupancy rate, hotel room and food and
beverage revenue are used to assess the performance of
hotel and restaurants, which are sub-components of the
commerce sector.
6
c. Transportation data
Administrative data from relevant agencies such as the
Civil Aviation Authority of Singapore, Maritime and Port
Authority of Singapore, Singapore International Airlines
and the Land Transport Authority are used to derive
indicators on passenger-kilometres, aircraft landings,
vessel arrivals, container throughput, number of taxis, etc
to assess the performance of the transport sub-sector.
These monthly/quarterly data are generally available 4 – 6
weeks after the end of the reference period.
d. Banking Schedules
Administrative data pertaining to Bank’s operations such
as assets and liabilities are also obtained from the
Monetary Authority of Singapore. Indicators such as loans
and advances of commercial banks, merchant banks and
finance companies as well as assets and liabilities of Asian
Currency Units (ACUs) are used to assess the performance
of the financial services sub-sector. These
monthly/quarterly data are also available about 4 weeks
after the end of the reference period.
21 DOS is also exploring the feasibility of developing additional indicators
from the greater exploitation of data pertaining to the Goods and Services Tax
(GST), which was introduced in 1994. Countries such as New Zealand and
Canada have used GST statistics extensively as short-term economic indicators.
Survey Data
22 Since administrative data are often insufficient, they would have to be
supplemented by survey data. Unlike the annual surveys, which are large-scale
and comprehensive, these higher frequency (usually quarterly, but sometimes
monthly) surveys collect minimal information from a relatively small number
of establishments, which nevertheless account for a large share of total value
added in the industry. Data from these surveys are used to derive key
economic indicators for the relevant services industries. Some examples are:
a. Monthly Surveys of Retail Sales & Catering Trade
These monthly surveys have been conducted since 1985.
They collect basic data on turnover, which enable the
compilation of the monthly retail sales and catering trade
indices. These indices, available with a 4 weeks lag, are
used to assess the performance of domestic trade and
restaurants within the commerce sector.
7
b. Quarterly Surveys of Wholesale Trade
This survey was introduced in 1996 to complement the
above two surveys. The data collected are used to compile
the quarterly wholesale trade index which, together with
the retail sales index, enable a more complete assessment
of domestic trade. The index is available with a 6 – 7
weeks lag.
c. Quarterly Business Expectations Surveys
These quarterly surveys on services and commerce have
been conducted since 1975. The surveys currently have a
sample size of about 1,250 establishments covering
transport and storage, business services, real estate and
financial services.
The surveys collect, in addition to qualitative assessment
of business expectations, quantitative data for a small
number of key economic and financial variables. These
include operating receipts, expenditure and employment.
The qualitative data, which are released in the Quarterly
and Annual Economic Surveys, are used to assess business
sentiments. The quantitative data are used as crucial
inputs for the compilation of quarterly national accounts
estimates, and some components of the composite-leading
index. Results from these surveys are available with a
one-month lag.
d. Quarterly Surveys of Financial Institutions
The Monetary Authority of Singapore’s quarterly survey
of financial institutions collects data on selected income
and expenditure items, including the amount of fees and
commissions received by banks, ACUs as well as stocks,
shares and bond brokers. These data are used in
conjunction with monthly data from the financial markets
and the banking schedules to assess the performance of
financial services. The information is available with a 4
weeks time lag.
e. Survey of Quarterly National Income Estimates
For statutory boards and larger private companies in
sectors, which are not covered in other surveys, DOS
conducts a quarterly survey of national income estimates
to collect basic data on income and expenditure items.
The results are also available with a 4 – 6 weeks lag.
8
V ON-GOING AND FUTURE DEVELOPMENT
Measurement of Output
23 While the judicious use of administrative and survey data has enabled
the compilation of a fairly comprehensive range of short-term indicators,
particularly for traditional services, DOS recognises the need for the
development of additional data sources and indicators for the new, emerging
and non-traditional services industries. Further, within traditional services
whose output are largely provided at non-market prices, e.g. education services,
medical and healthcare, as well as public administration, there is an increasing
recognition for the development of new output-based indicators to replace the
traditional input-based indicators. Table 4 shows selected industries, which
pose the most measurement difficulties.
TABLE 4: SELECTED SUB-SECTORS AND INDUSTRIES OF SERVICES
Financial Intermediation Education
Real Estate, Renting & Business Activities Community, Social & Personal Services
Software development & consultancy Sewage & refuse disposal
Research & development Trade Union
Legal services Recreational & cultural activities
Accounting services Library & museum
Advertising services Sports & country clubs
Public Administration & Defence Health & Social Services
Economic development & promotion services Medical services
Defence Social services for the handicapped
Judiciary Child care services
24 Insofar as traditional services generally provided by the public sector,
e.g. public administration, defence, education, medical and healthcare, the
problems associated with the current international practice of valuing their
output on the basis of the value of inputs used, or the costs of productions are
widely recognised. However, the difficulties of developing appropriate output
measures for these services are not easy to resolve. Statistical agencies in the
advanced economies are actively developing new methodologies to resolve
these difficulties. These methodologies invariably require the collection of
more detailed information. Their experiences will serve as a useful reference
for us, as we seek to improve our statistical indicators to measure the output of
this group of services.
9
25 With the rapid emergence of new, non-traditional services within
business services, the development of additional short-term indicators for
business services is both a priority and a challenge. Even for traditional
services such as legal, accounting and advertising services, it is necessary for
new and more appropriate indicators to be developed to replace some of the
existing input-based indicators such as employment and provident fund
contributions. Some indicators being considered are the volume and value of
property transactions for legal (conveyancing) services, and advertising revenue
for advertising services.
Better Data Coverage of New Industries
26 The existing coverage of the quarterly business expectations surveys
(BES) is not sufficient to cater for new, emerging services industries,
particularly those which are still relatively small. These include several of the
IT and media-related services. However, considerations of timeliness and the
need to minimise respondent burden suggest that it may not be appropriate to
re-design and expand the BES. Instead, a new quarterly survey of business
receipts will be launched. This survey will focus on the collection of basic
information on business receipts and expenditure (i.e., it collects less
information than the BES). Its coverage will not overlap with that of the BES,
but will instead supplement the BES, particularly in new and emerging
services.
27 Information from the BES and the quarterly surveys of national income
estimates will be combined with those to be collected from the quarterly survey
of business receipts to develop a new quarterly business receipts index (QBRI)
which will be used to assess the performance of business services. While these
services (excluding real estate and rental) currently account for only about 5 or
6 per cent of GDP, their importance is expected to increase very rapidly. With
the additional data, we will also be developing the Index of Services Output,
which is analogous to the Index of Industrial Production for the manufacturing
sector compiled by the Economic Development Board (EDB). This index will
enable better measurement of services output in volume terms and will provide
a better gauge of the services sector performance.
28 With the growing importance of electronic commerce (e-commerce),
priority is given to ensuring adequate coverage of this information in our
statistical system. For a start, some basic data items pertaining to e-commerce
have already been included in our current annual survey of commerce and
services for reference year 1997. These include the following:
a. type of e-commerce activities (exchange procurement
information or sales of product/services);
10
b. type of payments accepted (credit cards, cheques, cash
cards, etc); and
c. type of e-commerce transactions (business-to-business or
business-to-consumer) transactions.
29 In recognition of the importance of a proper statistical framework on the
measurement of e-commerce, DOS will organise an International Statistical
Institute (ISI) Cutting Edge Conference on “The Measurement of Electronic
Commerce” in December, 1999. This conference seeks to provide a forum for
academics, practitioners and statisticians to exchange ideas with the view of
developing a common understanding of the definition and measurement of e-
commerce, including how their impact on the economy can be assessed.
30 DOS will further study carefully the definitions of the information and
communication technologies (ICT) industries agreed at the meeting of the
OECD/ICCP Statistical Panel in June this year2. The development of statistical
indicators for the ICT sector will provide a useful reference for the
development of a statistical framework for the measurement of e-commerce.
An Occasional Paper will be issued on this study.
Exploitation of IT and Reduction of Respondent Burden
31 A notable feature of Singapore’s statistical system is the high degree of
exploitation of IT for data collection and processing. The greater exploitation
of IT is particularly crucial in view of our scarce manpower resources, and the
need to minimise respondent burden. The most recent example is the launching
of the collection of statistical returns through the Internet. The electronic
transfer of returns (ETR) was launched in March 1998 for the BES.
32 ETR enables faster processing with less manpower resources with the
data submitted by respondents directly updated into our database. Online
validation checks are instituted into the system to ensure data accuracy.
Authentication (through digital signature) ensures data integrity and
confidentiality with only authorised respondents being permitted to submit
returns directly to the database. DOS will extend ETR to other surveys.
2
‘Towards a definition of ICT Commodities’ by Eurostat discussed at the 13th Voorburg Group
meeting.
11
VI CONCLUSION
33 The measurement of services is inherently difficult. The need for
timeliness, the scarcity of resources and the necessity to reduce respondent
burden has made the development of short-term indicators for services even
more difficult. Notwithstanding these difficulties, DOS has developed a fairly
comprehensive range of timely short-term indicators which are used both to
assess the performance of services, and as direct inputs into the compilation of
Singapore’s national accounts estimates. These indicators are generally
adequate for traditional services.
34 However, the emergence of new and non-traditional services industries
and the rapid pace of financial innovation has resulted in the urgent need for
further additional and more appropriate short-term indicators for financial and
business services. In taking a strong initiative to develop indicators in these
areas, DOS has adopted an approach that is characterized by the following:
a. close collaboration with the appropriate government
agencies, e.g. MAS (for financial services) and the EDB
(for business services);
b. integrated approach in the use of multiple data sources
with maximal use of administrative data, as well as data
from several independent but related surveys; and
c. greater exploitation of IT.
Singapore Department of Statistics
October 1998
12
Annex 1
SELECTED SHORT-TERM INDICATORS FOR SERVICES
INDUSTRIES
Industry Indicators Frequency
Commerce
Entrepot Trade Re-export trade statistics Monthly
Foreign Wholesale Trade Index Quarterly
Domestic Trade Trade margins Monthly
Retail Sales Index
Domestic Wholesale Trade Index Quarterly
Hotels Room days occupied Monthly
Visitor arrivals
Average occupancy rates
Hotel room revenue
Restaurants Food & beverage revenue Monthly
Catering Trade Index
Transport & Communications
Land Transport Bus & MRT fare receipts Monthly
Number of taxis
Number of hired vehicles
Water Transport Container throughput Monthly
Sea cargo handled
Vessels arrival
Financial statements of shipping lines Quarterly
Air Transport Passenger-kilometres Monthly
Freight tonne-kilometres
Aircraft landing
Air cargo handled
Financial statements of airlines Quarterly
Services Allied to Transport Consignments handled (courier services) Monthly
Air passenger departures (travel agencies)
Number of cruise passengers Quarterly
Communications Direct exchange lines Monthly
International telephone calls in minutes
Internet services
Mobile phone and pager usage
Postal articles handled
Financial Services
Commercial banks Loans & advances Monthly
Fees & commissions Quarterly
Asian Currency Units Assets & liabilities Monthly
Fees & commissions Quarterly
Merchant Banks Loans & advances Monthly
Fees & commissions Quarterly
Finance Companies Loans & advances Monthly
Fees & commissions Quarterly
13
SELECTED SHORT-TERM INDICATORS FOR SERVICES
INDUSTRIES
(continued)
Industry Indicators Frequency
Financial Services (continued)
Stocks, shares & bond brokers Value of shares transacted Monthly
Fees & commissions Quarterly
FOREX Trading Value of transactions Monthly
Futures Trading Volume of futures contract (Singapore Monthly
International Monetary Exchange)
Life Insurance Premiums & claims Quarterly
Financial statements
General Insurance Premiums & claims Quarterly
Financial statements
Business Services
Real Estate Developers' margins Quarterly
Occupied space of commercial &
residential buildings
Financial statements from Housing Monthly
Development Board & Jurong Town
Corporation
Legal Services Property transactions (Conveyancing) Monthly
Number of provident fund contributors
Advertising Advertising Revenue Monthly
Number of provident fund contributors
Engineering & Technical }
Services }
IT Services }
Management Consultancy }
Business Representative } Number of provident fund contributors Monthly
Offices & Head/Regional } Data on revenue from Quarterly
Offices } Business Expectations Survey
Accounting Services }
14
