An Overview of the Proximate Determinants of Economic Growth in Indonesia Since 1960^*

* Parts of this article are based on Aldino Musyawwiri’s master’s thesis (Musyawwiri 2018), which was supervised by Murat Üngör. The authors thank Mark Millin for proofreading the paper. The views expressed herein are those of the authors and not necessarily those of the institutes to which they are affiliated.

Abstract

In this paper, we address two questions. First, what determined the growth of GDP per worker in Indonesia from 1960 to 2014? We examine Indonesia’s economic performance, using a growth accounting framework. We show that economic growth during the Soeharto era after 1975 was mainly determined by an increase in capital accumulation. Negative growth in total factor productivity (TFP) during the Asian financial crisis was more noticeable in Indonesia than in comparable ASEAN countries. In Indonesia, the contribution of TFP growth turned persistently positive after 1999. Second, what are the key determinants of the GDP per worker differences between Indonesia and the United States? Using data from the recently updated Penn World Table database and employing a levels accounting method, we find that the gap in physical capital deepening between the two countries is of declining importance in explaining the gap in labour productivity between Indonesia and the United States. We then compare our findings with data from the World Bank’s Changing Wealth of Nations 2018.

Penulis mengajukan dua pertanyaan. Pertama, apa yang menentukan pertumbuhan PDB per pekerja di Indonesia pada periode 1960-2014? Penulis melihat kinerja ekonomi Indonesia dengan menggunakan kerangka growth accounting. Kami menunjukkan bahwa pertumbuhan ekonomi pada rezim pemerintahan Soeharto setelah tahun 1975 sangat ditentukan oleh kenaikan akumulasi kapital. Pertumbuhan produktivitas (total factor productivity, TFP) negatif selama krisis keuangan Asia paling nampak terjadi di Indonesia. Kontribusi pertumbuhan TFP berubah positif dan persisten sejak 1999. Kedua, apakah yang menjadi determinan kunci dari perbedaan PDB per pekerja antara Indonesia dan AS? Penulis menggunakan data terkini dari database Penn World Table untuk mengaplikasikan metode levels accounting dan menemukan melemahnya peran kesenjangan pendalaman modal fisik dalam menjelaskan perbedaan produktivitas tenaga kerja di Indonesia dan AS. Penulis kemudian membandingkan hasil temuan dengan data dari publikasi World Bank, Changing Wealth of Nations 2018.

Keywords:

• Indonesia
• capital accumulation
• growth
• productivity
• ASEAN

JEL classification:

• O11
• O47
• O57

Notes

* Parts of this article are based on Aldino Musyawwiri’s master’s thesis (Musyawwiri 2018), which was supervised by Murat Üngör. The authors thank Mark Millin for proofreading the paper. The views expressed herein are those of the authors and not necessarily those of the institutes to which they are affiliated.

1 Booth (1998), Dick et al. (2002), Hill (2000a), Robison (2009), and van der Eng (1992, 2009, 2010) provide detailed perspectives on Indonesia’s long-term economic growth.

2 Using the same data source as in this paper, Barker and Üngör (2018) provide a detailed study on Vietnam’s convergence experience.

3 Indonesia’s economy was the largest in terms of nominal GDP, with more than $.9 trillion in 2016. It was followed by Thailand ($.4 trillion), then the Philippines, Singapore, and Malaysia ($.3 trillion each) (World Bank 2018a).

4 All data are from World Development Indicators (World Bank 2018a).

5 databank.worldbank.org/data/download/site-content/OGHIST.xls.

6 We use the variable GDP per capita in 2017 $ (converted to the 2017 price level with updated 2011 PPPs) from the March 2018 version of the Conference Board Total Economy Database.

7 All PWT versions are available at https://www.rug.nl/ggdc/productivity/pwt/.

8 PWT 9.0 reports a time series for labour share between 1960 and 2014 for Indonesia. The minimum (maximum) figure is .43 (.46) in the sample period. The reported figures for each year are very close to .45.

9 Confrontation (Konfrontasi) was a foreign policy developed by President Sukarno, which used a combination of diplomatic, economic, and military pressure to force a foreign adversary to accept a negotiated settlement on Indonesian terms (Lim 2015). Sodhy (1988) provides a historical and political background of Indonesia’s confrontation with Malaysia.

10 The movement claimed it had acted to prevent a planned coup d’état by a CIA-backed ‘Council of Generals’ and that it remained loyal to President Sukarno (Robinson 2018, 6).

11 Robinson (2018, 3) states that an estimated half a million members of the Indonesian Communist Party and its affiliated organisations were killed between late 1965 and mid-1966.

12 See also Kim and Lau (1994), Krugman (1994), Young (1994), and Aswicahyono and Hill (2002).

13 https://www.pbs.org/wgbh/pages/frontline/shows/crash/etc/cron.html.

14 https://www.economist.com/node/9401752.

15 https://www.brookings.edu/opinions/the-asian-financial-crisis-20-years-on-lessons-learnt-and-remaining-challenges/.

16 There is much literature on the causes of the AFC. See Corsetti, Pesenti, and Roubini (1999) for a comprehensive review, and Hill (1999, 2000b) for a perspective on Indonesia.

17 https://www.pbs.org/wgbh/pages/frontline/shows/crash/etc/cron.html.

18 In addition to the contraction in the real economy, Indonesia’s financial system shrank in the aftermath of the crisis. Domestic credit to the private sector fell from 57% in 1997 to less than 20% in 1999, which was significantly below Indonesia’s ASEAN peers (World Bank 2018a).

19 Using PWT 9.0, we calculate that the average labour share for 1960–2014 was .57 for Malaysia, .42 for the Philippines, .43 for Singapore, and .43 for Thailand. These figures are used in our growth accounting exercises and are reported in tables 2 and 3.

20 Natural disasters are an omnipresent risk to the resilience of growth in Indonesia. OECD (2015, 17) reports that close to .2 million people died as a result of natural disasters between 2000 and 2014.

21 Levels accounting is a popular technique to address the sources of the large differences in income observed between the richest and poorest economies in the world. See Klenow and Rodríguez-Clare (1997), Hall and Jones (1996, 1999), Caselli (2005), Hsieh and Klenow (2010), Jones and Romer (2010), and Jones (2016).

22 In other words, variable factor shares create a classic index number problem. A levels accounting with the Cobb–Douglas production function is invalid without the assumption of the same factor shares across countries (Sturgill 2014; Campbell and Üngör 2018; Harchaoui and Üngör 2018).

23 Our choice is supported by studies specific to Indonesia (van der Eng 2010; Dutu 2016).

24 Appendix A1 provides an exercise with a more flexible formulation that does not assume a common labour share across countries.

25 https://www.rug.nl/ggdc/docs/what_is_new_in_pwt90.pdf.

26 https://datacatalog.worldbank.org/dataset/wealth-accounting.

27 Data are reported in constant 2014 dollars at market exchange rates. A country-specific GDP deflator is used for all wealth components to bring the values to constant 2014 dollars.

28 The data in tables 4 and 5 are based on the Excel Country Tool, which uses the underlying data of the Changing Wealth of Nations 2018 study (https://datacatalog.worldbank.org/dataset/wealth-accounting).

29 https://stats.oecd.org/Index.aspx?DataSetCode=EAMFP.

30 Data on natural capital include 14 types of subsoil assets, comprising fossil fuels (hard coal, soft coal, gas, and oil) and minerals (bauxite, copper, gold, iron ore, lead, nickel, phosphate, silver, tin, and zinc) (Cárdenas Rodríguez, Haščič, and Souchier 2018).

31 It is important to note that subsoil assets may capture only a small part of Indonesia’s natural capital stock, leaving out many other aspects, such as land, water, natural forests, and renewable resources, such as fish stocks (see Brandt, Schreyer, and Zipperer 2017 for a discussion of this issue in the context of other countries). The Changing Wealth of Nations 2018 dataset does not include fish assets either.

32 However, both accounting exercises are useful diagnostic tools, since cross-country (growth) regressions are subject to concerns about possible endogeneity bias, multicollinearity, and reverse causality. Both accounting exercises avoid the concern about endogeneity by applying externally estimated production parameters to variations in a country’s production inputs (Hanushek and Woessmann 2015, chapter 4.4).

33 Measuring human capital by the (average) years of schooling may not be an appropriate representation of the actual knowledge acquired (Hanushek and Woessmann 2008, 2015).

34 http://blogs.worldbank.org/education/latest-pisa-results-seven-key-takeaways. Singapore topped the PISA 2015 rankings in all subject areas.