Multinational presence and labour productivity differentials in Indonesian manufacturing, 1975–2001

Abstract

Foreign multinational corporations (MNCs) have accounted for important shares of employment and production in Indonesian manufacturing since 1975, and these shares increased especially rapidly in the early to mid-1990s. These increases were concentrated in the machinery industries and in MNCs with large foreign ownership shares, and continued through the crisis of 1997–98 and beyond, despite apparently large withdrawals of inward foreign direct investment in 1998 and subsequent years. MNCs generally had much higher average labour productivity than local plants and, after controlling for plant-level variation in electricity consumption per employee, size and vintage, we found that these differentials persisted in about three-quarters of the cases examined. However, there was also large variation in MNC presence and in MNC–local productivity differentials across industries and time, with statistically insignificant differentials most common in apparel and footwear, as well as in MNCs with small foreign-ownership shares.

Acknowledgments

The authors would like to thank Hal Hill, Robert E. Lipsey, Oleksandr Movshuk, Fredrik Sjöholm, Wang WenThuen and an anonymous referee for comments on earlier versions of this paper (Takii and Ramstetter 2000, 2003, 2004). It was completed as a part of ICSEAD’s project ‘Foreign Multinational Corporations and Host-Country Labor Markets in Asia’. All opinions and remaining errors are the responsibility of the authors alone.

Notes

¹Several related studies indicate that the presence of MNCs leads to higher labour productivity or total factor productivity in local firms, i.e. to positive productivity spillovers (Blomström and Sjöholm 1999; Hill 1988: 107–17, 1990a, 1990b; Sjöholm 1998, 1999a, 1999b, 2000; Takii 2004, 2005).

²Evidence for Malaysia (Menon 1998; Oguchi et al. 2002) and Thailand (Ramstetter 2002b, 2004) differs from the Indonesian evidence in suggesting that MNCs are often not more productive than local plants in individual industries, but is similar in suggesting that MNCs’ productivity levels are not systematically related to foreign ownership shares in Thailand.

³See Pangestu (2002) for details of foreign investment policy under the New Order regime.

⁴External factors included strong growth and/or appreciating currencies in important home economies such as Hong Kong, Japan, South Korea, Singapore and Taiwan, which made their MNCs bullish on the prospects in Indonesia, other Southeast Asian economies and China. Increasingly sophisticated networks also evolved among manufacturing MNCs in Singapore, Malaysia and Thailand, making it easier for Indonesia-based MNCs to become increasingly involved in these networks by the mid to late 1990s.

⁵This is the most widely used estimate of FDI, and refers to equity and loans from foreign-owned multinational groups to affiliates in Indonesia in which the foreign group has own-ership shares of 10% or more. Indonesian estimates of FDI originate in Bank Indonesia and are taken from International Monetary Fund, International Financial Statistics, 1994 Year-book and January 2005 CD–ROM. This measure ignores outward FDI by Indonesians, estimates of which are available only from 1993, creating the possibility that estimates for earlier years confuse inward and outward FDI. However, outward FDI was probably very small in 1975–92.

⁶Data for MNCs are not published by BPS, but have been compiled by the authors from raw factory-level data for 1975–2001 from the industrial surveys, provided by BPS. See Takii and Ramstetter (2004) for details by MNC ownership group and industry for 1975–2001.

⁷BKPM data come from Investment Coordinating Board (1997) and Board of Investment and State-owned Enterprise Development (2000).

⁸According to the BKPM data (footnote 7) these 26 economies combined to account for 56% of realised FDI stocks in non-oil, non-finance industries in 1997 and 59% in mid-2000. Hong Kong and Taiwan were the major investors not included in this group, and they combined to account for 18% and 17%, respectively, of cumulative FDI in 1997 and mid-2000. Shares of other investors, including joint ventures involving firms from multiple countries, were also large, 24% and 23%, respectively. However, because European and US MNCs account for the vast majority of the FDI in the large oil and gas sector, the 26 economies covered in table 1 probably accounted for much larger shares of total FDI than the BKPM data suggest.

⁹For example, the ratio of raw data-based estimates of manufacturing employment to estimates based on the 1998 backcast data increased from 86% in 1985–89 to 91% in 1991–93 and 95% in 1994–95 (Takii and Ramstetter 2000). See this source for further details on the backcast datasets and comparisons with raw data for 1985–98.

¹⁰See Takii and Ramstetter (2004: 11, 33–34) for details on the method used to correct ownership shares.

¹¹A part of the increase in the share of apparel resulted from the reclassification of some textile products as apparel from 1990 forward.

¹²Shares of minority-foreign plants are not shown in tables 2 and 3 to save space, but can be calculated from the information provided in the tables.

¹³These are the combined shares of office and computing machinery (SITC [Standard Industrial Trade Classification] 75), telecommunications machinery (SITC 76), other electrical machinery (SITC 77), professional and scientific instruments (SITC 87), and photographic and optical equipment (SITC 88) (ICSEAD 2004).

¹⁴Data cited in the text are from Toyo Keizai, Kaigai Shinshutsu Kigyou Souran—Kuni Betsu [Japanese Overseas Investment by Country], various years. The trend toward larger foreign ownership shares appears to have continued thereafter, as several Japanese parents increased or planned to increase their ownership shares in their auto affiliates in Indonesia. These parents include Honda (49% to 51% in 2002), Nissan (35% to 75% in September 2001), Daihatsu (40% to 61.75%, announced in August 2002), Suzuki (49% to 90%, announced in November 2002), Hino (60.15% to 90% of its manufacturing firm, announced in January 2003) and Toyota (49% to 95% of its manufacturing firm, announced in February 2003). Data for auto parents are taken from press releases on the home pages of these companies.

¹⁵Although the period dummies are a reasonable proxy for vintage in plants that have always had 20 or more employees and always reported data in the industrial surveys, this variable also captures the characteristics of plants that first employed 20 or more workers and/or reported data in a period.

¹⁶Some factory identification codes differed between 2001 and 1975–2000. Because this makes it impossible to calculate vintage and year dummies (Ds7585_it, Ds8691_it, Ds9294_it and Dy_it ), 2001 data are excluded from these estimates, and from the revised (March 2005) version of Takii and Ramstetter (2004).

¹⁷We thank Hal Hill and an anonymous referee for this suggestion.

¹⁸It also contrasts with results from related studies suggesting that heavily-foreign MNCs have the highest export propensities, followed by majority-foreign MNCs and then minority-foreign MNCs (Ramstetter 1999).

¹⁹In Thai manufacturing, wholly-foreign MNCs (100% foreign owned) often have lower labour productivity than other MNCs and local plants (Ramstetter 2004), but there is a strong positive correlation between foreign ownership shares and the probability of a plant having a high export propensity (Ramstetter 2002a).