The development of the Indonesian service sector: a first analysis of the estimates

Trucks and busses

Unfortunately no information is available about revenue and operating expenses of trucks and busses. Therefore some very bold assumptions have to be made. We assume that, because more people are needed (for among others loading/unloading, administration, etc.) to operate a bus or truck, the value added of these means of transportation are twice as high as that of a taxi. This is exactly in line with the assumptions made by Neumark (1954, p. 377) and Muljatno (1960, p. 176) who both had access to information of the Ministry of Communications and Ikatan Motor Indonesia (Indonesian Motor Union).⁴
To arrive at constant price series the following procedure is followed.

• 
  Traditional road transport is extrapolated based on an urbanisation index.
  

• 
  Taxis/cars are extrapolated based on the number of cars in Indonesia.
  

Value added in current prices is obtained by assuming that value added in this sector followed the development of the cost of living index. This index is taken from van Leeuwen (forthcoming). The constant price series is inflated by this index.

For the period 1950-2000 we rely on the input-output tables which are published every five years since 1975. These estimates are extrapolated based on an index of number of motor vehicles. Current price estimates are obtained by inflating the constant price series by the consumer price index.

2.1.3 Water transport

During the colonial period the most important player in shipping transport was the Koninklijke Paketvaart Maatschappij (KPM). In 1930 it was estimated that KPM covered 42.9 per cent of total intra-island shipping traffic or almost 80 per cent of intra-island shipping under the flag of the Netherlands-Indies (Statistiek van de Scheepvaart over het jaar 1930). Since the profit and loss accounts of this company are available it is possible to estimate its value added for the period 1900-1940.

In the 1960s estimates were based on Bank Indonesia’s statistics of Indonesian owned and chartered tonnage in interinsular and ocean trade. Gross value added at constant prices was obtained by deflating current price estimates by the cost of living index.

From the 1980s onwards output and current prices were estimated by multiplying the number of cargoes and passengers transported by tariff per unit of cargo and passenger, respectively. Average output data were derived from shipping enterprise reports, while data on cost structure were based on Input-Output tables. Cargo and passenger data were provided by the National Shipowner Association (INSA), BPS and other sources. Gross value added at constant prices was calculated by extrapolation using weighted composite indices of cargoes and passengers transported.

Just like with the other sub-sectors, these underlying reports and surveys have never been published. It is also remarkable that, besides loading and unloading statistics, other data on cargo and passengers or systematic data on the number of ships sailing under the Indonesian flag have never been published by BPS. So, again, for the period 1950-2000 the I-O tables serve as the main starting point from which estimates for the other years are arrived at by extrapolation.


2.1.4 Air transport

The value added of air transport can also be estimated with the production approach. In 1928 the Dutch Netherlands-Indies Airlines (Koninklijke Nederlandsch-Indische Luchtvaartmaatschappij, KNILM) started its operation. Based on its annual reports value added of air transport could be estimated for the period 1928-1940.⁵

For the period after independence the same estimation method has been used. Gross value added was estimated by production approach, based on output and cost structure data that were obtained through airline enterprises survey by BPS. Gross value added at constant prices was computed by using a weighted composite production index of passenger-kilometer and ton-kilometer of cargo transported.

It will be no surprise that these surveys have not been published either. Neither have I been able to find annual reports of any of the airlines, although it is questionable whether this would lead to more reliable estimates than the ones made by BPS. The I-O tables again are the most valuable source for estimation.

2.1.5 Communication

This sector covers the activities of the post, telegraph and telephone offices. During the colonial period the communication sector was in hands of the state-owned enterprise PTT (Staatsbedrijf der Posterijen, Telegrafie en Telefonie). The contribution to national income is obtained as the sum of wages and salaries, interest and operating surplus. The data available for this sector are, like that for rail and air transport, quite comprehensive and are taken from the annual reports of the PTT.⁶

The method of estimation throughout the second half of the 20^th century was also the production approach. Output at current prices was gathered from financial reports of these companies. Value added was also from the financial report in the form of summing wages and salaries, profit or loss, depreciation and other components of the value added. Value added and output at constant prices were estimated by extrapolation.

2.2 Trade

According to the 1993 System of National Accounts (SNA) traders are treated as supplying services rather than goods to their customers by storing and displaying a selection of goods in convenient locations and making them easily available for customers to buy. Their output is measured by the total value of the trade margins realized on the goods they purchase for resale.

2.2.1 Trade margins in colonial time

Unfortunately data on trade margins during the colonial period are almost non-existent. Sivasubramonian in his research on the national income of India in the 20^th century and Horlings for the Netherlands between 1800-1850 were confronted with a similar data problem as well (Sivasubramonian, 2000, p. 338; Horlings, 1995, p. 342).

Smits, on the other hand, gives some very detailed information on trade margins in the Netherlands. According to his calculations for the trade house Van Eeghen the gross profit margin was in the 1850s on average between 25 to 30 percent for imports and the net profit margin between 10 and 12 percent. Compared to trade margins for the NHM, he concludes, this seems to be a rather modest estimate. Furthermore Smits found that the trade margin on imports decreased to around 6 percent in 1890 and 5 percent in 1913 (Smits, 1995, p. 314-315).

However, for the Netherlands-Indies no direct data on trade margins for the colonial period have been found. Research in the archives of Dutch trading houses, such as Internatio, Handelsvereniging Amsterdam (HVA, Trading Association Amsterdam) and the Nederlandse Handelsmaatschappij (NHM, Netherlands Trading Society) has not yielded information that enables one to calculate disaggregated trade margins. Only a very crude way can be adopted to estimate an aggregated trade margin. This is by dividing total turnover by gross profit. The annual reports of Internatio and Hagemeijer provide this information.

This method results in trade margins that vary significantly between an average of 17.6 percent for Hagemeijer and 4.5 percent for Internatio. It is hard to believe that such large and consistent differences really existed. Especially because trends in profitability are fairly similar. Probably the difference is just caused by different accounting methods and the ‘true’ profit margin lies somewhere in between.

2.2.2 Trade margins in Indonesia’s National Accounts