Resource Guide Changing Economy in Indonesia and Foreign Trade Statistics of Colonial Indonesia

By Hiroyoshi Kano

The 15-volume series Changing Economy in Indonesia (hereafter, simply Changing Economy) has been published in the Netherlands beginning in the mid-1970s and was scheduled for completion in the course of 1996. The series covers every type of economic statistic for colonial Indonesia up to the early 1940s. The publisher is the Amsterdam-based Royal Tropical Institute (in Dutch, Koninklijk Instituut voor de Tropen, or KIT). The titles are as follows:
	1.	Indonesia's Export Crops 1816-1940
	2.	Public Finance 1816-1939
	3.	Expenditure on Fixed Assets
	4.	Rice Prices
	5.	National Income
	6.	Money and Banking 1816-1940
	7.	Balance of Payments 1822-1939
	8.	Manufacturing Industry 1870-1942
	9.	Transport 1819-1940
	10.	Foodcrops and Arable Lands, Java 1815-1940
	11.	Population Trends 1795-1942
	12a.	Trade Statistics 1823-1940
	12b.	Regional Patterns in Foreign Trade 1911-1940
	13.	Wages
	14.	Prices (non-rice)
	15.	Forestry
This statistical data has been extracted, processed, and edited from every kind of statistical record from the late 18th century to the early 1940s. The editing task was begun by the late P. Creutzberg of KIT. Creutzberg died while directing the project, and was succeeded by P. Boomgaard, economic historian at Amsterdam Free University (and presently the director of the Royal Institute of Linguistics and Anthropology) who took over direction of the project team established at KIT.

The plan for gathering and editing the massive amounts of documents scattered around colonial Indonesia (the Netherlands East Indies) was first conceived in the 1930s by W.M.F. Mansvelt, chief of the Netherlands East Indies Government Central Statistics Bureau. Creutzberg worked in the Central Statistics Bureau at that time. After Indonesia achieved independence following World War II, he returned to the Netherlands where he applied himself to accomplishing the project conceived by Mansvelt, his former superior. He took a new position at KIT (known before the war as the Colonial Research Institute (Koloniaal Instituut)) where his efforts steadily brought forth results.

The data in the 15 volumes includes a huge amount of data gathered from different time periods and places and compiled according to different methods. Within the various subject matters, it has been, to the greatest extent possible, compiled and edited into standardized time series statistics. As expected in a project expected to require upwards of 20 years to complete, immense amounts of time and labor input were required. Indeed, the results reflect customary Dutch precision and perseverance. No matter how hard one might search, it is not possible to find such refined statistical collections in other Southeast Asian countries.

The statistical collections for Indonesia are a pioneering research resource which, of course, the Asian Historical Statistics Project cannot overlook in its quest to compile a long-term database. By borrowing data from Changing Economy, we should save countless hours of labor. For our purposes, however, there are some problems with the materials. One example concerns Volume 5, National Income. The manuscript written in 1943, over 50 years ago, is put into print by this volume. Thus, the calculation of national income in this volume never reflects the results of data compilation carried out by the Changing Economy team. Further, considered from present standards, there might be problems with the old methods and concepts used in this work to compute macroeconomic calculations. In addition, the period covered is only from 1921 to 1939.

Even if we leave aside macroeconomic calculations of national income statistics and focus on statistical databases for general economic historical research, there are a number of problems with Changing Economy. One source of problems results from the determined efforts to arrange long-term time series statistics. Consequently, the numerical values for items are too assiduously arranged in large categories to the exclusion of detailed matters. Historians are sometimes unable to identify important and decisive changes. Let me turn to foreign trade statistics, which will serve as an example to explain the nature of the problem.

Volumes 12a and 12b of Changing Economy deal with trade. The former covers 1823-1940 (or parts of the period, depending on the item). The following 9 categories appear:

(1) Changes in total value of imports
(2) Changes in total value of exports
(3) Changes in values of imports from regions which are important trade partners
(4) Changes in values of exports to regions which are important trade partners
(5) Changes in import values for most important items
(6) Changes in export values for most important items
(7) Domestic trade
(8) Customs duties on imports and exports
(9) Changes in values of imports and exports of major islands
The second of the two volumes covers only 1911-1940, listing the following ten tables:
(1) Aggregated trade values for selected products by region in Java
(2) Regional balances of trade in the Outer Provinces
(3) Regional composition of major Java-based exports
(4) Regional composition of major non-Java-based exports
(5) Regional composition of main imports entering Java
(6) Regional composition of main imports entering the Outer Provinces
(7) Commodity composition of exports from major ports in Java
(8) Commodity composition of imports entering major ports in Java
(9) Commodity composition of exports from major Outer Provinces
(10) Commodity composition of imports entering major Outer Provinces

There can be no doubt that using the data on changes in trade structure to track changes in the economic structure of colonial Indonesia will yield useful and fascinating results. However, a number of product areas are bundled into such huge categories that their value as historical research materials is rather limited. This is particularly true of the last four charts, including, for export items, raw materials, consumer commercial foodstuffs, and refined oil among import items, and, for import items, foodstuffs, consumer goods, and capital goods. Further, because we cannot obtain cross-data for separate items or regional trading partners, we cannot figure out what items were exchanged among which regions and what time series changes occurred. It is, for example, almost impossible for scholars wishing to investigate changes in sugar export levels from Java to other regions to find the necessary data.

I wondered if it might be possible to go back to the original data in cases where the databases were clearly only partly compiled. To ask about this matter, I made an e-mail enquiry to J. Th. Lindblad of Leyden University's History Department. Dr. Lindblad, a historian who assisted with the editing of Volume 12, directly acknowledged that lack of cross-data constitutes a problem in the database, and indicated that some academic projects would be impossible to realize. That type of cross-data was not formulated during the compilation process.

In addition, Dr. Lindblad told me that the most important raw statistical data which I wanted to use can only be accessed in the Netherlands. He said that using the materials was not so easy as one might think, but that he would provide assistance if the research was important enough for me to make the trip to the Netherlands. The Business Office of the Statistics Project informed me that there was travel money remaining in the budget, and the timing was perfect, so last February, during coldest winter, I headed for the Netherlands.

It was the most severe winter in several years, and frozen canals and snowy landscapes could be seen from the windows of the Royal Institute of Linguistics and Anthropology (KITLV) which is located within Leyden University's campus. At KITLV's library I spent two weeks poring through decades of trade statistics materials and entering information into the laptop computer I had brought along. In the limited amount of time available, I was hard pressed simply to get to the beginning of the 20th century. Further, seeing the actual materials and their condition for the first time convinced me it was true, as I had been told, that it was no easy task to arrange the data for the period prior to the 20th century. The main reason is that some commodities were recorded according to the physical volume of their trade, while the other commdities were recorded according to the monetary value of their trade. Nevertheless, judging from the data I gathered on this trip, one can conduct some quite interesting fact finding.

Let me conclude by reiterating the two main benefits which Changing Economy provides the Historical Statistics project. First, intricate colonial-era economic statistics of every sort have been arranged in comparatively simple format and, for the purpose of formulating macroeconomic estimates, they provide an excellent baseline. The second advantage is that -- even though from a historian's point of view, the arrangement is oversimplified and materials needed for specific research themes are often lacking -- we can nevertheless use the materials to understand what approaches to take in order to deal with missing data. The clues can be deduced from the materials. It would be frustratingly difficult to conduct research if this material were not already collected.

Tokyo University, Institute of Oriental Culture