Dynamic productivity growth and its determinants in the Indonesian food and beverages industry

ABSTRACT

This research investigates dynamic productivity growth and its determinants in the Indonesian food and beverages industry decomposing dynamic productivity growth into the contributions of dynamic technical inefficiency change, dynamic technical change, and dynamic scale inefficiency change. The empirical application employs unbalanced panel data of 44 subsectors in the Indonesian food and beverages industry over 1990–2014. To estimate dynamic productivity growth, this research uses a Luenberger indicator accounting for the presence of adjustment costs. The results show that dynamic productivity growth exhibits a decreasing trend. Dynamic technical inefficiency change and dynamic scale inefficiency change contribute positively to dynamic productivity growth, while dynamic technical change contributes negatively. Dynamic productivity growth is affected by the change in industrial concentration, the growth rate of capital intensity, the growth rate of exports, the growth rate of foreign direct investment, and location.

KEYWORDS:

• Dynamic productivity growth
• Indonesian food and beverages industry
• directional distance function
• Luenberger indicator

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

1. Autocorrelation or serial correlation causes bias in the standard errors of the parameters or inefficiency in the parameters of the model without making the parameters biased (see Gujarati 2004). The Driscoll and Kraay method can correct this inefficiency caused by the correlation of the dependent variable over time and between subjects.

2. The Loc variable represents the effect of the location. Thus, this is not presented in the first difference form. The Fisher-type test for the stationarity test will be applied because this research uses unbalanced panel data.

3. For example, the subsector of 10210 is a combination from the subsectors of 10211 and 10212.

4. For example, subsector of 10223 had ∆HHI of 0.80 in 1997, since many firms exited from the industry caused by the economic crisis at the period.

5. The TFPG in the period of 1991/1990 indicates that the TFPG is calculated between the periods of 1991 and 1990.

6. Based on the parameters of the fixed effects of the model, the subsectors having the positive (negative) average TFPG also had positive (negative) parameters of the fixed effecs.

Additional information

Funding

This work was supported by the Direktorat Jenderal Pendidikan Tinggi [PDUPT Dikti].