![Sample our Economics, Finance,Business & Industry journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5931/TOC-Subject-Sample-2021-Economics-Finance-Business-Industry.jpg"})
Abstract
We analyse discrepancy in the trade data between China and New Zealand, where what China reports as its exports to New Zealand and what New Zealand reports as its imports from China reached an alarming level of US$2.5 billion in 2014. We investigate the roles that export-tax and import-tariff avoidance play in explaining this discrepancy. We find strong evidence of export-tax avoidance for China’s exports to New Zealand that explains 11–27% of the missing exports, equivalent to 3.9–8.8% of true export value. We find only weak evidence of import-tariff avoidance in either direction.
Acknowledgements
We would like to thank the seminar participants at the 11th Australasian Trade Workshop in Brisbane, 56th Annual Conference of the New Zealand Association of Economists in Auckland and the Department of Economics and Finance at the University of Canterbury who saw an early draft of this paper. We are greatly indebted to the anonymous referees whose suggestions were highly valuable and helped us improve our paper significantly. All remaining errors are ours.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1 The value of total trade was US$4.9 billion in 2007 and US$15 billion in 2014. Trade figures are taken as an average of what was reported by China and what was reported by New Zealand and includes Hong Kong re-exports.
2 The VAT system is production-based, which means that there are no rebates on capital goods purchased.
3 After correcting for the CIF-FOB difference.
4 For example, Singapore has a very high trade-to-GDP ratio. The high trade-to-GDP ratio for Singapore is because of the city-state’s role as the middleman for firms in developed countries that outsource the assembly part of the production process to low-wage countries such as China. This could be a possible reason for the negative trade discrepancies and should be investigated in further research.
5 Our approach of having different regression models for the northbound and southbound trade between New Zealand and China is in line with the findings of Patnaik et al. (Citation2012).
6 Because there is no variation across goods in the GST rate, we cannot test if this is contributing to underreporting of imports at the New Zealand Customs.
7 We want to thank Frank Staehler for bringing this idea to our attention.
8 This classification system was developed by the World Customs Organisation and, at the 6-digit level, assigns a 6-digit code to each of about 5000 commodities. The 1992 nomenclature is the original version and the system has since been updated every five or six years.
9 In about 40% of these ‘one-sided’ observations China did not report a value for imports and in the other 60% New Zealand did not report a value for exports.
10 In about 25% of these one-sided observations New Zealand did not report imports and in the remaining 75% China did not report a value for exports.
11 NZTtariff(2008) = .75*MFN tariff + .25*min[FTA tariff, MFN tariff].
12 The figure is based on the Epanechnikov kernel function with a bandwidth of 0.4405.
13 This and subsequent tables are reported without the CIF-FOB adjustment as we do not make this adjustment in our econometric model given that this level adjustment does not affect the marginal effects of tariffs and taxes on the discrepancy that our study is interested in.
14
15 See footnote 14.
16 The lower bound comes from the preferred specification in Column 3 (0.856*4.6) and the upper bound comes from Column 2 (1.993*4.6).
17 See Appendix 1 for the calculation based on our regression results using the ER measure of discrepancy. Our regression results using the TG measure of discrepancy produce much smaller estimates of the value of the missing trade attributable to export tariff avoidance, equalling 0.32% of the reported value of exports.
18 While we control for industry-specific factors, there could be other possible reasons that we fail to capture. For example, if environmentally dirty products see changes in technology such that associated inputs are difficult to price internationally, it could drive both reported trade discrepancies and the export tax.
19 This result is obtained by substituting coefficients 0.891 and 2.226 for ChinaExportTax found in Table columns 3 and 2, respectively, into (A.3).
20 Of all missing trade in the southbound trade, roughly 25% have no imports recorded by New Zealand and 75% have no exports reported by China.
21 One possible explanation for the positive sign of the quadratic term for tariff could be as follows: when tariffs increase, it is likely that fewer goods are imported, which could improve the efficiency of customs controls discouraging underreporting.
22 Rauch (Citation1999) defined differentiated products as those not having a reference price or those that do not have prices quoted on organised exchanges. Rauch suggested two definitions, a conservative and a liberal one, in order to account for the ambiguities arising in the classification. The conservative definition minimises the number of commodities that are classified as homogeneous goods, while the liberal definition maximises this number. We follow the conservative definition. According to this definition, for example, animal oils and fats are classified as homogeneous goods, while textile and leather machinery are considered to be differentiated products.