Abstract
I estimate the cost of being landlocked on exports using a structural gravity model. The empirical challenge of doing so is to estimate a country-specific variable, being landlocked, in the presence of exporter and importer fixed effects. To do so I follow two alternative approaches, each of which models the exporter fixed effect as a function of country-specific variables and average trade costs. Both approaches show a substantial “landlocked penalty”, with landlocked countries on average exporting 27–41% less than non-landlocked countries over 2005–2014, all else equal. I further demonstrate that such a penalty is driven primarily by developing countries. Indeed, whilst I find no landlocked penalty in high-income countries, the penalty was over 40% in developing countries. The difference between the two sets of countries is likely driven by the income level of transit countries, and the ease with which exporters can access the coast. Developing landlocked countries are generally constrained by limited infrastructure, inefficient logistics services and lengthy border delays. This constraint appears to be increasing over time; the landlocked penalty was higher in 2014 than in both 2005 and 2010.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Throughout the paper, I use the terms “developing” and “low-income” to mean all countries that are not classified as “high-income” by the World Bank. Those countries classified as “middle-income” by the World Bank are therefore included here within “low-income”. I show in the robustness tests—Table 5—that the landlocked penalty applies to both middle-income and the remaining low-income countries, although the effect is smaller for the middle-income countries [although the coefficients are not significantly different (in 2014)].
The multilateral resistance term captures a country’s ability to trade with all countries. That is, trade between i and j will depend not only on the trade cost between the two countries, but also each country’s propensity to trade with other countries. If j is located very close to a number of large exporters for example, it will tend to import less from i because there is lots of competition for j’s market. Excluding the multilateral resistance term overlooks this fact, and wrongly models bilateral trade as dependent only on the trade cost between countries i and j.
Indeed, omitting the multilateral resistance terms and including distance as the only trade cost factor in \(\tau _{ij} \) gives the most basic form of the gravity equation, with bilateral trade determined by the size of the two economies and the distance between them.
Standard errors are clustered by exporter i due to the presence of \(\sigma _i \) in the error term.
This number is smaller than 180x194 as not all countries report trading with each other.
The number of observations for low-income landlocked countries increased from 2057 in 2005 to 2434 in 2014 for example.
If aggregate trade (the mean of imports and exports) is used here instead of exports, the results are very similar.
ln(GDP per capita) is positive and significant if ln(GDP) is omitted, and the landlocked dummy remains negative and highly significant.
Productivity estimates are taken from the International Labour Organisation “labour productivity” series and are in constant $2005. Output per worker is calculated as GDP/number of employed persons. GDP figures are taken from national accounts and the ILO states that “estimates of employment are, as much as possible, for the average number of persons with one or more paid jobs during the year”. The data is available http://www.ilo.org/global/statistics-and-databases/lang--en/index.htm and meta-data is available http://www.ilo.org/ilostat-files/Documents/description_PRODY_EN.pdf.
I use those countries listed as “Least developed country: UN classification” in the World Bank country classifications, available https://datahelpdesk.worldbank.org/knowledgebase/articles/906519-world-bank-country-and-lending-groups.
This is a substantial figure, as the mean cost for non-landlocked countries is $4620.
Note that for computational reasons, these robustness tests are estimated using 2014 data only. The \({\bar{z}} _i\) variables are included throughout, but are not presented for brevity.
References
Anderson, James, and Eric van Wincoop. 2003. Gravity with gravitas: Solution to the border puzzle. American Economic Review 93 (1): 170–192.
Anderson, James, and Eric van Wincoop. 2004. Trade costs. Journal of Economic Literature 42 (3): 691–751.
Arvis, Jean-Francois, Gael Raballand, and Jean-Francois Marteau. 2010. The cost of being landlocked. Washington, DC: The World Bank.
Baldwin, Richard, and Daria Taglioni. 2007. Trade effects of the euro: A comparison of estimators. Journal of Economic Integration 22 (4): 780–818.
Collier, Paul. 2008. The bottom billion: Why the poorest countries are failing and what can be done about it. Oxford: Oxford University Press.
Collier, Paul, and Jan Willem Gunning. 1999. Why has Africa grown slowly? Journal of Economic Perspectives 13 (3): 3–22.
Eaton, Jonathan, and Samuel Kortum. 2002. Technology, geography and trade. Econometrica 70 (5): 1741–1779.
Faye, Michael A., John McArthur, Jeffrey Sachs, and Thomas Snow. 2004. The challenges facing landlocked developing countries. Journal of Human Development 5 (4): 31–68.
Gallup, John Luke, Jeffrey D. Sachs, and Andrew D. Mellinger. 1999. Geography and economic development. International Regional Science Review 22 (2): 179–232.
Head, Keith and Thierry Mayer. 2015. Gravity equations: workhorse, toolkit and cookbook. In Handbook of International Economics, ed. Elhanan Helpman, Kenneth Rogoff and Gita Gopinath Vol 4, 1–740. Amsterdam: Elsevier.
Head, Keith, Thierry Mayer, and John Ries. 2010. The erosion of colonial trade linkages after independence. Journal of International Economics 81 (1): 1–14.
Hummels, David. 2007. Transport costs and trade in the second era of globalization. Journal of Economic Perspectives 21 (3): 131–154.
Limao, Nuno, and Anthony Venables. 2001. Infrastructure, geographical disadvantage, transport costs, and trade. The World Bank Economic Review 15: 3.
Silva, Santos, M.C. João, and Silvana Tenreyro. 2006. The log of gravity. The Review of Economics and Statistics 88 (4): 641–658.
USAID. 2010. Transport and logistics costs on the Tema-Ouagadougou corridor. West Africa Trade Hub Technical Report 25.
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
Table 5 re-estimates the one-step model (i) using a Poisson pseudo maximum likelihood (PPML) estimator (Silva et al. 2006), (ii) using a PPML estimator but only including positive export observations,Footnote 12 (iii) dropping all exporters with a GDP or population greater than two standard deviations above the low-income landlocked average, (iv) scaling exports by diving by the average of exporter and importer GDP, (v) including landlocked dummies for the importing nation and (vi) splitting the “low income” group into “middle income” countries and the remaining “low income” countries Table 6 presents a list of landlocked countries.
Rights and permissions
About this article
Cite this article
Moore, A.J. Quantifying the Landlocked Trade Penalty using Structural Gravity. J. Quant. Econ. 16, 769–786 (2018). https://doi.org/10.1007/s40953-017-0106-3
Published:
Issue Date:
DOI: https://doi.org/10.1007/s40953-017-0106-3