Export Market Size Matters: The Effect of the Market Size of Export Destinations on Manufacturing Growth

Abstract

Literature contends that the manufacturing sector is crucial for economic development, and it is conventional wisdom that exports drive manufacturing growth. However, it has yet to be established empirically whether the market size of export destinations is an essential factor in explaining diverging regional and sectoral manufacturing growth patterns. This article argues that accessing a few large external markets reduces entry costs, increases expectations of economies of scale, and fosters capital formation. To test this hypothesis, we construct a novel Relative Export Market Size (REMS) index that measures whether the share of sectoral exports destined to large economies in one region is higher than in other regions. Using a PVAR model with fixed effects, we verify the impact of the REMS index on value added, employment, and capital accumulation of 129 manufacturing sectors in 23 regions in Colombia from 1992 to 2017. The obtained results show that exporting to larger markets positively impacts employment, capital formation, and value added per capita of manufacturing sectors at a regional level. This finding indicates that exporting to the world’s largest market helps develop competitive manufacturing sectors.

KEYWORDS:

• Export market size
• manufacturing exports
• trade
• manufacturing growth
• regional growth
• industrialization

JEL Classifications:

• F14
• O14
• R11

Acknowledgements

We thank Alejandro Torres, Jan Priewe, the participants of the 23rd Conference of the Forum for Macroeconomics and Macroeconomic Policies (FMM), and two anonymous reviewers for their comments that helped us to improve earlier versions of this manuscript.

Data availability statement

The data underlying this article are available in Mendeley and can be accessed under DOI:10.17632/yjdt73sw8m.2.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Notes

1 The REMS index ranges from -1 to 1.

2 The Big Push literature argues that simultaneous large-scale government interventions in many sectors are necessary to enable developing countries to industrialise because they lower the investment risk and ensure economies of scale and complementarities between industries (Rosenstein-Rodan, 1943). A common feature of the Big Push models is that they show that ‘complementarities between industrializing sectors work through market size effect’ (Murphy et al., 1989b, p. 1024).

3 Although empirical studies back this hierarchy theory to some extent, their results rather suggest that the most popular destinations for exporting firms are not the largest markets but the closest ones. Eaton et al. (2011: Table 1), for example, report that for French firms Belgium/Luxembourg, Germany and Switzerland are the most popular export destinations, while the USA is only placed at rank seven; and Lawless (2009: Table 6) finds that the UK is by far the most popular market for Irish firms.

4 EAM is based on firm-level data. However, it is not advisable to use microdata or highly disaggregated data, given that at levels above 4-digits DANE distorts the data to ensure anonymity. Similarly, the data from ten regions and that of the sectors 1600, 2414, 2430, 2729, 2922, 2927, 3000, 3210, 3220, 3230, 3312, 3330, 3511, 3512, 3530, 3692 cannot be used due to statistical reservations. Please note furthermore that ISIC’s product categorisation changed twice during the sample period (Rev.2 before 2000, Rev.3 from 2000-2012, and Rev.4 from 2012 onwards). To make the categories comparable over time, they are all brought to Rev.3. To combine Rev.2 with Rev.3, it is assumed that establishments that existed before 2000 did not change their main sector, i.e. they received the same classification for the period 1992–1999 than they had in the year 2000 (71% of the sample). The rest is classified in line with product concordance tables from DANE (25%) or excluded from the sample (4%). Similarly, to combine Rev.4 with Rev.3, existing establishments received the same categorisation during 2012–2017 as they had in 2011 (93%), while the rest is classified according to the correlative tables (6%) or excluded from the sample (1%).

5 Available at www.dane.gov.co/index.php/sistema-estadistico-nacional-sen/normas-y-estandares/nomenclaturas-y-clasificaciones/tablas-correlativas.

6 All values are brought to constant 2014 COP by using the industrial Producer Price Index at ISIC’s two digit-level. Exports are reported as FOB in USD, while EAM data is reported in COP. To make the data comparable, we transform the USD values into COP with exchange rate data reported by the Colombian Central Bank.

Additional information

Notes on contributors

Thomas Goda

Thomas Goda is Professor of Economics at the School of Finance, Economics and Government at Universidad EAFIT (Medellin, Colombia). His main research areas are International and Development Economics, Economic Inequality, and Political Economy. His research has been published in journals such as Socio-Economic Review, Cambridge Journal of Economics, Journal of International Financial Markets, Institutions and Money, Development and Change, Social Indicators Research, and Review of World Economics.

Santiago Sánchez González

Santiago Sánchez González is a junior researcher with working experience as an economic consultant at the Inter-American Development Bank. His main research interests are Macroeconomics, Political Economy, and Urban Economics. His research has been published in the repository of the Inter-American Development Bank and journals such as Transportation Research Part A, Transport Policy, and Sustainability.