About me

• PhD student at Crawford School of Public Policy, ANU.

• Lecturer at Politeknik APP Jakarta, currently on leave.

About this paper

• Part of my dissertation.

• click s.id/krisna-aifis for the slides.

• click here for the pre-print.

• Indonesian economic growth has been struggling to reach the pre-1998 level.

• Development plans still trying to emphasize on manufacturing to carry Indonesian economy.

• Globalization helps developing nations import capital, intermediate goods and know-how, exactly what Indonesia need.

• This paper focuses more on the importance of trade policies on imported intermediate goods to Indonesian manufacturing growth.

• Some findings:

  • restrictive import regulations hurt firms productivity.
  • Smaller firms affected more heavily.

• Intermediate inputs help firms' backward participation in Global Value Chain (World Bank 2020).

• Help firms access better inputs (Amiti and Konings 2007; Bas and Strauss-Kahn 2014; Castellani and Fassio 2019; Ing, Yu and Zhang 2019).

• Help firms innovate and be more productive (Fernandez and Farole 2018, Pane and Patunru 2019).

• Improves access to more lucrative foreign markets (An and Maskus 2009; Cadot et al. 2013; Fugazza, Olarreaga and Ugarte 2017)

• Closer ties of economic relationship also benefits non-economically.

• Follows 1,512 firms from 2008-2012.

• Survei Industri (SI)

  • Output, factors, ownership.
  • Highly unbalanced, many problematic zeroes.

• Integrated customs data

  • Firms' import, HS-8 detail, source countries.
  • Only use matched SI-Customs firms.

• Tariff collected from scrapping MoF's regulations.

• NTMs from UNCTAD TRAINS.

  • 7 broad categories, mainly SPS and TBT.

• Follows approach by Amiti and Konings (2007) and Pane and Patunru (2019).

• Using Levinsohn and Petrin (2003) algorithm to get marginal productivity of labour (l), energy (n), intermediate input (m) and capital (k):

$$y_{it}={\beta }_0+{\beta }_l{l}_{it}+{\beta }_n{n}_{it}+\varphi (m_{it},k_{it})+{\mu }_{it}$$

• Predict TFP of each firm i at time t:

$$TF{P}_{it}=y_{it}-{\hat{\beta }}_l{l}_{it}+{\hat{\beta }}_n{n}_{it}+{\hat{\beta }}_m{m}_{it}+{\hat{\beta }}_k{k}_{it}$$

• Tariff is scrapped from Ministry of Trade's regulations

  • around 7 of them consisting almost 10,000 tariff lines
  • Customs data include source country to assign specific tariff
  • Tariff coverage ratio is calculated per firm basis

• Same treatment for NTMs

  • NTM data is count data. So it's not percent.

$$C_{\theta it}=\frac{\sum T{P}_{\theta scit}{V}_{\theta scit}}{\sum V_{\theta scit}}$$

• Lastly, log TFP is regressed as the function trade policies, ownership, firm fix effect and industry fix effect.

$$tf{p}_{it}={\gamma }_0+\sum _{\theta }{\gamma }_{\theta }{c}_{\theta it}+\sum _{\theta }{\delta }_{\theta }{c}_{\theta it}\ast l_{it}+F{O}_{it}+{\alpha }_i+ISI{C}_i+{\eta }_{it}$$

• The term ${\delta }_{\theta }$ shows the different impact of trade policies among firm with different size, measured in the number of labor.

• A percent increase of tariff correlates with a 0.371% reduction in firms' TFP.

  • Positive tariff*l suggest a less severe impact for large firms.
  • These results are consistent without Fix effect.

• NTMs are somewhat less important except for licensing (mostly quota restrictions) and competition (high presence of SOEs)

• Value-added per Labor (VaL) has the opposite impact.

  • It may be the case that firms reduce labor even more than output, hence $\frac{Va\downarrow }{L\downarrow \downarrow }\Uparrow$
  • To test for this, I use percent change in labor as my dependant variable and rerun the regression.

• Firms' employment is shrink with higher trade restrictions.

• Correlations with NTMs are more significant.

• Fix effect make the coefficient higher.

• This paper provides yet more evidence that protectionism hurts manufacturing.

  • it leads to lower productivity and employment.

• Additionally, bigger firms may have better ways to dampen the impact.

• Overall, strategy to use manufacturing to improve the economy using trade policy would end up as a disaster.

• Targeting CAD may not be ideal.

  • bad debt management can't be solved by protecting from import.

• TFP variable may reflect market power

• NTMs have mixed results.

  • Count data isn't ideal
  • May have different impact for different types of NTMs
  • May have different impact on different goods
  • More in-depth studies are needed

• The sample is quite restrictive.

  • Importers are different compared to general firms
  • the number of importers is very small

• Data isn't perfect.

  • SI have many problems
  • Customs have not go beyond 2012 yet