Uluslararası ticaret tutarsızlığının ölçümü: Türkiye'nin ihracat farklılıkları için üç aşamalı ampirik bir çerçeve

Abstract

Ayna ticaret istatistikleri, uluslararası ticaret raporlamasının tutarlılığını değerlendirmek için yaygın olarak kullanılmakta; ancak ihracatçı ve ithalatçı ülkelerin raporladığı değerler, değerleme yöntemleri, raporlama uygulamaları, yeniden ihracat ve kurumsal farklılıklar nedeniyle sıkça birbirinden ayrışmaktadır. Bu çalışma, 2013–2024 dönemi için 204 partner ekonomiyi kapsayan UN Comtrade verilerini kullanarak Türkiye'nin ihracatındaki ayna farklılıklarını analiz etmek amacıyla bir Uluslararası Ticaret Tutarsızlığı (ITI) çerçevesi geliştirmektedir. Ampirik tasarım; betimleyici, açıklayıcı ve öngörücü yöntemleri üç aşamada bütünleştirmektedir. İlk aşamada kümeleme analizi, ticaret ortakları arasındaki heterojen farklılık rejimlerini belirlemektedir. İkinci aşamada, Türkiye'nin en büyük yirmi ihracat pazarı için panel lojistik regresyon, farklılık riski ile Dünya Bankası Dünya Kalkınma Göstergeleri'nden elde edilen ekonomik ve kurumsal göstergeler arasındaki ilişkiyi değerlendirmektedir. Üçüncü aşamada, WEKA ortamında uygulanan lojistik sınıflandırma ve Random Forest modelleri, on katlı çapraz doğrulama ile öngörü sağlamlığını sınamaktadır. Bulgular, ayna farklılıklarının ortaklar arasında tekdüze değil yapısal nitelikte olduğunu göstermektedir. Ticaret hacmi ve seçilmiş ülke özellikleri, geniş yönetişim göstergelerine kıyasla farklılık riskiyle daha yakından ilişkilidir; makine öğrenmesi modelleri ise orta düzeyde öngörü performansı sunarak ekonometrik bulguları genel olarak desteklemektedir. Çalışma, ayna ticaret farklılıklarının doğrudan ticaret usulsüzlüğü kanıtı olarak değil, raporlama riski göstergesi olarak yorumlanması gerektiği sonucuna ulaşmaktadır.

Keywords

• Ayna Ticaret İstatistikleri
• Uluslararası Ticaret Tutarsızlığı
• UN Comtrade
• Panel Logit
• Makine Öğrenmesi

Measuring international trade inconsistency: a three-stage empirical framework for Türkiye's export discrepancies

Abstract

Mirror trade statistics are widely used to assess the consistency of international trade reporting, yet exporter- and importer-reported values frequently diverge because of valuation conventions, reporting practices, re-exports, and institutional differences. This study develops an International Trade Inconsistency (ITI) framework for analysing mirror discrepancies in Türkiye’s exports using UN Comtrade data for 204 partner economies over 2013–2024. The empirical design integrates exploratory, explanatory, and predictive methods in three stages. First, clustering identifies heterogeneous regimes of discrepancy across trading partners. Second, a panel logistic regression for Türkiye’s twenty largest export destinations evaluates the association between discrepancy risk and economic and institutional indicators from the World Bank’s World Development Indicators. Third, logistic classification and Random Forest models implemented in WEKA assess predictive robustness through ten-fold cross-validation. The results show that mirror discrepancies are structured rather than uniform across partners. Trade volume and selected country characteristics are more closely associated with discrepancy risk than broad governance indicators, while the machine-learning models provide moderate predictive performance and broadly support the econometric findings. The study concludes that mirror trade discrepancies should be interpreted as reporting-risk indicators rather than direct evidence of trade misreporting or fraud. By combining classification, econometric analysis, and predictive validation, the proposed framework contributes to statistical reconciliation, customs risk assessment, and research on international trade data quality.

Keywords

• Mirror Trade Statistics
• International Trade Inconsistency
• UN Comtrade
• Panel Logit
• Machine Learning

References

1. Bhagwati, J. N. (1974). On the underinvoicing of imports. In J. N. Bhagwati (Ed.), Illegal transactions in international trade: Theory and measurement (pp. 138–147). North-Holland.
2. Biswas, A. K., & Sengupta, S. (2011). Tariffs and imports mis-invoicing under oligopoly. Economics and Politics, 23(2), 254–270. https://doi.org/10.1111/j.1468-0343.2011.00384.x
3. Biswas, A. K., von Hagen, J., & Sarkar, S. (2022). FDI mismatch, trade mis-reporting, and hidden capital movements: The USA–China case. Journal of International Money and Finance, 120, Article 102534. https://doi.org/10.1016/j.jimonfin.2021.102534
4. de Boyrie, M. E., Nelson, J. A., & Pak, S. J. (2007). Capital movement through trade misinvoicing: The case of Africa. Journal of Financial Crime, 14(4), 474–489. https://doi.org/10.1108/13590790710828181
5. Carrère, C., & Grigoriou, C. (2015). Can mirror data help to capture informal international trade? (FERDI Working Paper No. P123). Fondation pour les Études et Recherches sur le Développement International. https://ferdi.fr/en/publications/can-mirror-data-help-to-capture-informal-international-trade
6. Chan, B., Ng, I., & Chung, N. (2023). Anomaly detection in trade declarations using deep learning techniques: A risk-assessment approach to identify misclassification and incorrect valuation. Statistical Journal of the IAOS, 39(4), 825–834. https://doi.org/10.3233/sji-230081
7. Chen, C., Jiang, Z., Li, N., Wang, H., Wang, P., Zhang, Z., Zhang, C., Ma, F., Huang, Y., Lu, X., Wei, J., Qi, J., & Chen, W.-Q. (2022). Advancing UN Comtrade for physical trade flow analysis: Review of data quality issues and solutions. Resources, Conservation & Recycling, 186, Article 106526. https://doi.org/10.1016/j.resconrec.2022.106526
8. Farhad, M. H., Madsen, J. B., Jetter, M. K., & Williams, A. (2023). The mystery of misreported trade: A novel method to identify distorted trade figures. The World Economy, 47(2), 748–778. https://doi.org/10.1111/twec.13451

9. Ferrantino, M. J., & Wang, Z. (2008). Accounting for discrepancies in bilateral trade: The case of China, Hong Kong, and the United States. China Economic Review, 19(3), 502–520. https://doi.org/10.1016/j.chieco.2008.02.002
10. Fisman, R., & Wei, S.-J. (2004). Tax rates and tax evasion: Evidence from 'missing imports' in China. Journal of Political Economy, 112(2), 471–496. https://doi.org/10.1086/381476
11. Forstater, M. (2018). Illicit financial flows, trade misinvoicing, and multinational tax avoidance: The same or different? CGD Policy Paper No. 123. Center for Global Development. https://www.cgdev.org/publication/illicit-financial-flows-trade-misinvoicing-and-multinational-tax-avoidance
12. Gara, M., Giammatteo, M., & Tosti, E. (2019). Magic mirror in my hand… How trade mirror statistics can help us detect illegal financial flows. The World Economy, 42(11), 3120–3147. https://doi.org/10.1111/twec.12840
13. Hamanaka, S. (2012). Whose trade statistics are correct? Multiple mirror comparison techniques: A test case of Cambodia. Journal of Economic Policy Reform, 15(1), 33–56.https://doi.org/10.1080/17487870.2012.657827
14. Hayakawa, K. (2020). Departure months and discrepancy in mirror trade data. IDE Discussion Paper 795. Institute of Developing Economies. https://www.ide.go.jp/English/Publish/Reports/Dp/795.html
15. Hong, K. P., & Pak, S. J. (2017). Estimating trade misinvoicing from bilateral trade statistics: The devil is in the details. The International Trade Journal, 31(1), 3–28. https://doi.org/10.1080/08853908.2016.1202160
16. Javorcik, B. S., & Narciso, G. (2008). Differentiated products and evasion of import tariffs. Journal of International Economics, 76(2), 208–222. https://doi.org/10.1016/j.jinteco.2008.07.002
17. Javorsek, M. (2016). Asymmetries in international merchandise trade statistics: A case study of selected countries in Asia and the Pacific (SD Working Paper No. 02/April 2016). United Nations Economic and Social Commission for Asia and the Pacific (ESCAP). https://www.unescap.org/sites/default/files/SD_Working_Paper_April2016_Asymmetries_in_International_Trade_Statistics.pdf
18. Kellenberg, D., & Levinson, A. (2019). Misreporting trade: Tariff evasion, corruption, and auditing standards. Review of International Economics, 27(1), 106–129. https://doi.org/10.1111/roie.12378
19. Linsi, L., & Mügge, D. (2019). Globalization and the growing defects of international economic statistics. Review of International Political Economy, 26(3), 361–383. https://doi.org/10.1080/09692290.2018.1560353
20. Linsi, L., Burgoon, B., & Mügge, D. (2023). The problem with trade measurement in international relations. International Studies Quarterly, 67(2). https://doi.org/10.1093/isq/sqad020
21. Linsi, L., Burgoon, B., Mügge, D., & Ruíz, S. (2023). Mirror trade database v2 (1948–2021). Harvard Dataverse. https://doi.org/10.7910/DVN/HOOTBB
22. Markowicz, I., & Baran, P. (2020). Discrepancies between mirror data on intra-community trade: The case of Poland. Econometrics, 24(1), 1–11. https://doi.org/10.15611/eada.2020.1.01
23. Markowicz, I., & Baran, P. (2021). Mirror data asymmetry in international trade by commodity group: The case of intra-community trade. Oeconomia Copernicana, 12(4), 889–905. https://doi.org/10.24136/oc.2021.029
24. Morgenstern, O. (1963). On the accuracy of economic observations (2nd ed.). Princeton University Press.
25. Triepels, R., Daniels, H., & Feelders, A. (2018). Data-driven fraud detection in international shipping. Expert Systems with Applications, 99, 193–202. https://doi.org/10.1016/j.eswa.2018.01.007
26. United Nations. (2024). UN Comtrade database. https://comtrade.un.org
27. World Bank. (2024). World Development Indicators. https://databank.worldbank.org