Under-invoicing of imports: a case study of Pakistan.

Mahmood, Zafar ; Mahmood, Riaz

I. INTRODUCTION

Most of the developing countries, including Pakistan, set high tariffs and stringent quantitative restrictions (QRs) to protect their domestic industries from foreign competition and to raise tax revenues. Both tariffs and QRs, due to the weak enforcement of controls, provide incentives to smuggle goods through illegal channels and to under-invoice imports through legal channels of trade to evade import taxes. In particular, under-invoicing of imports challenge both the abovementioned objectives; that is, the under-invoicing of imports confers much lower protection to domestic industries than that accorded by statutory rates of import duties, while the tax revenues are lost as import taxes are evaded.

Tariff barriers, QRs, and foreign exchange rationing give rise to black foreign exchange markets. Foreign remittances and under-invoicing of exports are the major supply sources of foreign exchange in black markets. On the other hand, the demand for illegal foreign exchange comes largely from the nationals who want to travel abroad, capital flight abroad, and the under-invoicing of imports which requires the purchases of black market foreign exchange to make full payment to the foreign exporter.

Illicit trade activities spoil not only the accuracy and use of foreign trade statistics but also affect the usual policy package of a trade theorist. Any policy package which does not take into account the phenomenon of fake invoicing will not deliver the desired results. Thus ideally fake trade must be integrated into the usual trade policy prescription. At the same time it is also necessary to determine the presence and magnitude of fake trading. We restrict this paper to show whether the phenomenon of incentives-generated under-invoicing of imports exists in Pakistan.

The schematic details of the paper are as follows. Section II deals with reasons and the methodology to determine the under-invoicing of imports. In Section III an empirical analysis is presented. Finally, Section IV concludes the paper with some policy suggestions.

II. REASONS FOR UNDER-INVOICING OF IMPORTS

An importer will under-invoice imports if import duties and premium on quantitatively restricted imports are higher than the black market premium on foreign exchange which an importer pays in order to purchase it from the open market to make full payments to the foreign exporters. (1) In a situation where there are no controls on foreign exchange but trade barriers do exist then clearly the incentive will be to under-invoice imports. There is, however, a risk attached both to the understatement of import value and to engage in illegal foreign exchange transactions. Hence, under-invoicing will not occur unless the differential between the tariff rate and premium on the black market exchange is greater than the evaluated, risk factor.

Normally, one would expect that partner country's statistics to show an excess of c.i.f. import values over the corresponding f.o.b. export values of same traded commodities. But if the observed discrepancy is in the 'perverse' direction, and we do not have any other reason to explain the discrepancy, one may conclude that there is a prima facie evidence for under-invoiced imports. This inference would be intensified if we can show that for these commodities tariff rates are higher than the black market premium on foreign exchange and if the commodities are such that it is relatively easy to under-invoice because several quality differences foul up the concept of an 'international' standard price so that the custom authorities cannot really use a reliable yardstick to avoid faking of invoices see, Bhagwati (1974).

Following Bhagwati (1964, 1967) we use the partner-country-data comparison technique to test for incentives-generated under-invoicing of imports. (2) In this technique c.i.f. import values of the country are compared with the f.o.b. export values of the partner country to find 'perverse' discrepancies. To show such discrepancies, data has been obtained from the "Commodity Trade Statistics" of the United Nations (1981, 1988).

Alternative Explanations of Under-invoicing of Imports

One can note the following additional reasons (3) which can be put forward to explain the perverse discrepancy:

(i) The method presumes the faking of invoices to occur only at one end only. If both countries fake invoices it becomes impossible to make a case for under-invoicing. One-sided fake invoicing can be assumed given the strict enforcement in the partner countries.

(ii) There can be 'misallocations' of the same traded item, by both SITC category and by country. These inconsistencies can arise from both genuine custom's mistakes and differences in conventions adopted by trading partners. Cross-checking with other SITC categories and with other countries enables to sort out a possible explanation of any observed 'perversity' in discrepancies.

(iii) The discrepancies can also arise from the lag between shipment and arrival in the importing country. If the average lag remains unchanged, and the level of export values is not changing, there would of course be no discrepancy between export and import statistics from this source, since the excess of imports over exports to the preceding year's cant over would offset the deficit owing to this year's carry-forward.

III. EMPIRICAL ANALYSIS OF UNDER-INVOICING OF IMPORTS IN PAKISTAN

The above alternative explanations of 'perverse' discrepancies shows the practical difficulty which one faces while drawing any conclusions about under-invoicing of imports. Given certain limitations of the Bhagwati technique, we try to show that there is reasonable evidence which can be presented to show the under-invoicing of imports. In this section, we show several 'perverse' discrepancies for a good number of commodities which therefore seemed to be accountable, prima facie, only by an under-invoicing of imports hypothesis.

In this paper we have included only six major trading partners of Pakistan; namely, France, Germany, Italy, U.K., Japan, and the Netherlands, which accounted for about 40 percent of Pakistan's total imports in 1988. (4)

In order to confirm that the 'perverse' discrepancies reported in Tables 1 to 4 do not arise due to alternative explanations we checked for off-setting discrepancies. By summing over all countries by individual commodity groups we are still left with 'perverse' discrepancies for SITC 8 in 1981 and for SITC 6 to 8 in 1988 and do not find any off-setting evidence to rule out the under-invoicing of imports for these commodities. However, at the 3-digit level we do find some off-setting discrepancies, for instance in 1981 for France and the Netherlands one can note from Table 3 some perverse discrepancies in various sub-categories of manufactured goods, machinery and transport equipment but when we aggregate these by country, as is shown in Table 1 for 1981, then perversity in discrepancy disappear. (5) However, such exceptions do not effect the overall findings of under-invoicing of imports for SITC 6 to 8 for the year 1988. Moreover, by including only those countries which have relatively freer trade and using a relatively large number of commodities we have tried to eliminate the effects of other factors to a large extent which may give rise to the perverse discrepancy. Similarly, the argument of time lag between exports and imports is not valid in the present case and discrepancy can not be explained through this source. This is true because the high proportion of under-invoicing of imports, shown in Tables 3 and 4, is much higher than the 4 percent average annual growth of imports recorded for the 1980s.

We begin our analysis from the l-digit level by establishing that if under-invoicing takes place at this level then it is easy to establish it at a more refined level. At the 1-digit level, from Tables 1 and 2 we can see clear under-invoicing of imports for some commodity groups. Not all of the commodities testify to under-invoicing of imports. Under-invoicing of imports is, however, widespread in chemicals, manufactured goods, machinery and transport equipments, and miscellaneous manufactured goods and is taking place with every country.

Tables 3 and 4 provide the degree of under-invoicing of imports for various sub-categories of chemicals, manufactured goods, machinery and transport equipments, and miscellaneous manufactured goods, covering 1981 and 1988. (6) It may be noted from Tables 3 and 4 that for majority of commodities and for all countries (7) included in the analysis under-invoicing of imports is quite visible and is going up over time. It may also be noted that with the exception of a few commodities generally the extent of under-invoicing of imports is over 30 percent of export value.

The results of under-invoicing of imports are further strengthened by the fact that the large majority of commodities which show perverse discrepancies generally have high import duties (8) ranging between 40 percent for goods such as rubber products, power generating machinery, etc. to 450 percent for motor vehicles which were in excess of the black market premium on foreign exchange-22 percent in 1981 when the country had a fixed exchange rate and 8.7 percent in 1988 when a managed floating exchange rate was in practice (9)--thus carrying an incentive to understate imports. Thus the explanation of under-invoicing of imports stand, up quite well with the trade regime in Pakistan. Moreover, Tables 3 and 4 also reveal that these manufactured goods are such where it is relatively easy to foul up the concept of the 'international' standard price due to quality differences.

IV. CONCLUSIONS AND POLICY SUGGESTIONS

In this paper we have shown the existence of some significant perverse, discrepancies in the import statistics for which the only explanation appears to be under-invoicing. This argument is supported with evidence that significant differences exist between import duties and the premium on black foreign exchange and the fact that the nature of under-invoiced commodities is such that importers can easily dodge the custom authorities. The convincing indications of the presence of under-invoicing of imports leads us to put forward the following policy suggestions:

(i) Any fiscal incentive which would accord less benefits to the undervaluation of capital would make it less attractive to under-invoice imports of machinery/equipments. For instance, a policy such as the accelerated depreciation allowance, although an inefficient way to provide an incentive for increasing investment as it promotes capital-intensive technology in a labour-abundant country, has the ability to offset the under-invoicing of capital goods imports.

(ii) In Order to remove the prevalent under-invoicing practices, the government of Pakistan introduced the system of International Trade Prices (ITP) in 1987 to fix the minimum price Of a product for assessing import duties. While this system may have led to some lowering of under-invoicing in all those commodities where the quality foul-up is not possible, under-invoicing continues in all other commodities. In view of the changing global market a frequent revision and verification of ITP can lead to better enforcement of trade controls.

(iii) Given the rate of import duty if the black market premium on foreign exchange becomes low then it provides a greater incentive to importers to under-invoice. On the other hand, if import duties are simultaneously lowered along with a fall in the premium on black market exchange then the relative strength of the duty savings and premium paid in the black market will determine the extent of under-invoicing. Interestingly, although the average tariff rate in Pakistan has gone down, the import duties on commodities which are under-invoiced have not gone down to the level that it offsets the under-invoicing. In fact, as we noted earlier the under-invoicing of imports has increased over time. It is, therefore, suggested that any policy change to rationalise the tariff structure, aimed at reducing the average rate of import duty, must accompany with a reduction in the under-invoicing of imports.

Comments on "Under-invoicing of Imports: A Case Study of Pakistan"

The two Mahmoods have written a useful paper. The stated object is to show whether the phenomenon of incentives generated under-invoicing of imports exists in Pakistan, and the paper approaches the problem by way of an examination of bilateral trade statistics. The basic message of the paper is that under-invoicing of at least some of those imports that attract high tariffs does occur, the statistical basis for this inference is that for some of the products the imports fall short of the corresponding exports from the bilateral trade partner.

The usefulness of the paper lies in that it draws attention to an evident statistical discrepancy in the foreign trade account. Work on such specific tasks is particularly worthwhile in a time like ours when the basic paradigm (some would say paradigms) of economic theory is being increasingly questioned.

That the balance of payments statistics are amiss is a matter of world-wide concern and the authors do well to make this point in the context of Pakistan. But there can be more than one substantial reason why differences in ,trade statistics arise. And the authors refer to just such a multiplicity of causes, though the thrust of their discussion is that evasion of import taxes is the reason.

I would like to emphasise mechanisms other than under-invoicing for evading or avoiding import tariffs: tariffs can sometimes be avoided with the help of the subjectivity which characterises classification of goods for import purposes. For example a whole range of computers can be seen either as entertainment goods (which would be taxable), or as education machines which are not subject to tariffs. Hence goods attracting relatively large tariffs can be classified as goods attracting low tariffs. The other aspect is that discrepancies can arise from causes other than under invoicing such as for example the possible differences in values created by fluctuations in the exchange rates.

One of the authors assumptions, that there is what they refer to as strict enforcement in the trading partner country and that fake invoicing is possible only in Pakistan seems to run into at least one well known example to the contrary: that of the Matrix Churchill exports from the UK to Iraq. Those present here are probably familiar with the details so I shall not repeat the story.

The general state of trade statistics in different countries can also easily give rise to discrepancies. In Pakistan alone there are at least two main sources of trade statistics and substantial differences characterise these two sets of records. The State Bank maintains its external trade records on the basis of the relevant payments made and sums received. The revenue departments on the other hand compile their external trade records from the shipments actually received and despatched at port.

To the best of my knowledge these two sets of trade figures have never been satisfactorily reconciled.

The authors would also need to satisfactorily explain the discrepancies in the wrong direction. This I feel is necessary to convincingly clinch the argument that they wish to project.

The matter of the discrepancy in relation to the external account is a complex one globally as well. The IMF has for some years now drawn attention to a large but varying discrepancy in the global current account. This discrepancy increased particularly sharply between 1978 and 1982: from US$19.7 billion to US$ 113.9 billion--an almost six-fold increase in a space of just five years? The estimate for 1991 is US$ 121.9 billion. (2) Naturally for any given year the discrepancy is the counterpart of the recorded balance of payments current account deficits for the world as a whole.

Now the IMF discrepancy aggregates the discrepancies on individual external transaction accounts. A major source of this discrepancy is said to be the "large body of cross border assets recognised by the debtor countries but not by the creditors". Work on this aspect shows that the reported capital inflows exceeded outflows by some US$ 300 billion for the period 1977-83 alone. Particularly important for the present paper is the fact that there was a persistent excess of debits on account of shipping (averaging US$35 billion annually over 1977-83). Payments made for shipping are missing. The shipping account relates to the carriage of international trade. (3) Again the finding is relevant to the present context and needs to be considered.

The context of under/over invoicing too has changed since Bhagwati 1974 paper (which paper provides the model for the paper now under review). Worldwide there is an increasing reliance on the market mechanism; in particular a large number of developing economies are now quite liberal in the operation of their external current account as well as the external capital account. Record keeping though has not kept pace with this liberalisation and the number of unreported transactions may have increased--discrepancies are only to be expected.

Changes are evident in Pakistan too: for one thing the establishment of what are called duty free shops enables the import of a whole range of goods more or less legitimately and without duties--hence taking the pressure off the under-invoicing business. Second, the commercial sector in Pakistan has benefited (in terms of an influx of duty-free imports) from the large presence of a semi--nomadic community with considerable experience and some tradition in informal cross--border trade. Again this takes the pressure off the under-invoicing business. Third, there is an evident move towards reduction of tariffs on the part of the government, whatever the reason for that. This too would work to lessen the likely under-invoicing.

These changes need to be explored in relation to the hypothesis of the paper, as they appear to be quite pertinent to the context.

In concluding I would like to complement once again the authors on their choice of topic as well as in making a worthwhile effort at illuminating a relevant area of international money flows. My one reservation though is that the conclusions drawn are stronger than the evidence justifies.

Faizullah Khilji

Corporate Law Authority, Islamabad.

(1) International Monetary Fund, Report on the world current account discrepancy, (IMF, Washington, D.C., September 19S7).

(2) International Monetary Fund, World economic outlook May 1992, (IMF, Washington, D.C., May 1992).

(3) International Monetary Fund, Report on the World current account discrepancy, (IMF, Washington, D.C., September 1987).

Appendix

CODE SPECIFICATION

621 Material of Rubber
625 Rubber Tyres, Tubes
629 Rubber Articles
641 Paper and Paper Board
642 Articles of Paper
651 Textile Yarn
657 Special Textile Fabrics
662 Clay, Refractory Building Products
663 Mineral Manufactures
664 Glass
665 Glassware
673 Iron and Steel Shapes
674 Iron and Steel Plates, Sheets
675 Iron and Steel Hoops, Strips
682 Copper Alloys
684 Aluminum Alloys
691 Structures and Part of Iron, Steel
692 Metal Tanks, Boxes
694 Steel, Copper Nails, Nuts
695 Tools of Iron and Steel
699 Manufactures of Metal
713 Internal Combustion Engine
714 Engines and Motor
724 Textile Leather Machinery
727 Machinery for Food Industry
728 Machinery for Special Industry
736 Metal Working Machinery
741 Heating and Cooling Equipment
742 Pumps for Liquid
743 Pumps Centrifuges
744 Mechanical Handling Equipment
745 Machine Tools Non-Electric
751 Office Machinery
764 Telecommunication Equipment
771 Electric Power Machinery
772 Switch Gears, Parts
773 Electricity Distributing Equipment
775 Electro-mechanical Domestic Appliances
778 Electrical Machinery NES
781 Passenger Motor Vehicles excluding Buses
782 Lorries Special Motor Vehicles
783 Road Motor Vehicles, Buses
784 Motor Vehicle Parts
821 Furniture
881 Photo Apparatus, Equipments
882 Photo Cinema Supplies
885 Watches and Clocks
893 Articles of Plastic

REFERENCES

Bhagwati, J. N. (1964) On the Underinvoicing of Imports. Bulletin of Oxford University Institute of Statistics, November.

Bhagwati, J. N. (1967) Fiscal Policies, the Faking of Foreign Trade Declarations and the Balance of Payment. Bulletin of the Oxford University Institute of Statistics, February.

Bhagwati, J. N. (1974) Introduction. In J. N. Bhagwati (ed) Illegal Transactions in International Trade. North-Holland/American Elsevier.

Kemal, A. R. (1990) Industrial Sector Review of Pakistan. Islamabad: Pakistan Institute of Development Economics. (Mimeographed.)

Naya, S., and T. Morgan (1969) The Accuracy of International Trade Data: The Case of Southeast Asian Countries. Journal of American Statistical Association. Sheikh, M. A. (1974) Underinvoicing of Imports in Pakistan. Oxford Bulletin of Economics and Statistics, November.

Simkin, C. G. F. (1970) Indonesia's Unrecorded Trade. Bulletin of Indonesian Economic Studies.

United Nations (1981, 1988) Commodity Trade Statistics. New York: Statistical Office.

(1) Under-invoicing is done with the help of exporters in the trading partner country. The usual practice is that invoice is made for the amount shown on the Letter of Credit while the exporter receives the remaining portion of the total sale in the form of a bank draft, etc. arranged by the importer through the black foreign exchange market.

(2) A previous study on the subject of under-invoicing of imports in Pakistan covered the years 1965 and 1968, see Sheikh (1974). For other studies on the subject, also see Naya and Morgan (1969) and Simkin (1970).

(3) See, Bhagwati (1964, 1974) for a detailed discussion.

(4) We do not include U.S.A. in our analysis because prior to 1988 U.S.A. trade data also included Puerto Rico which has made the two data sets incomparable for two years.

(5) It may be noted that in many of these 3-digit level commodities under-invoicing of imports may actually exist but the size may not be large enough to result in an overall perverse discrepancy.

(6) We have used a ratio (M-X)/X to show the extent of under-invoicing, where M= c.i.f. value of imports and X = f.o.b. value of exports. A negative sign of the ratio indicates under-invoicing of imports.

The following points must be kept in view while drawing any conclusions from the results: (i) it may be noted from Tables 3 and 4 that some SITC numbers are missing, these are the commodities for which either data are not available or trade is negligible or a commodity is non-traded or a commodity is exportable and (ii) due to differences in denominators sum of these ratios does not lead us to the result we report in Tables 1 and 2.

(7) These countries accounted for 55.16 percent of total Pakistani imports of SITC 5 to 8 in 1988.

(8) Beside import duties, imports upto 1982-83 in Pakistan were licensed on the basis of Free and Tied Lists. Products on Tied List could be imported from specified sources, by specified users or by the public sector only. Products on Free List, which were also subject to QRs, could be imported from anywhere after obtaining a valid import license; in 1981 as many as 406 out of 435 products on the Free List were subject to QRs. By 1983, this number was reduced to only five and at present no product is subject to QRs. In 1983-84, a basic change in the licensing system was introduced when the system of Free and Tied List was replaced by a system of Negative and Restricted List. Products on Negative List can not be imported into the country while those on Restricted List are subject to different restrictions including origin of imported type of users for public sector only, and safety and health standard. The products not appearing on either of the two lists can be freely imported. See, Kemal (1990).

(9) Upto 1982 Pakistan was following the policy of the Fixed Exchange rate but since then Rupee is floating against dollar and linked to a basket of currencies.

Zafar Mahmood and Riaz Mahmood are, respectively, Senior Research Economist and Research Economist at the Pakistan Institute of Development Economics, Islamabad.

Table 1
Under-invoicing of Imports in Pakistan Jan-Dec 1981
(in 000 US Dollars)

 0 1 2

C.I.F. Pak Imports From France 25528 N.A. 1958
F.O.B. Exports of France to Pak 33470 100 1528

C.I.F. Pak Imports From W. Germany 13226 112 22567
F.O.B. Exports of W. Germany to Pak 13196 N.A. 13350

C.I.F. Pak Imports From Italy 2842 N.A. 1084
F.O.B. Exports of Italy to Pak N.A. N.A. 1335

C.I.F. Pak Imports From UK 5425 357 29321
F.O.B. Exports of UK to Pak 3801 927 14755

C.I.F. Pak Imports From Japan 9780 N.A. 61245
F.O.B. Exports of Japan to Pak 7815 N.A. 53580

C.I.F. Pak Imports From Netherlands 17476 N.A. 7619
F.O.B. Exports of Netherlands to Pak 16250 N.A. 4067

 3 4 5

C.I.F. Pak Imports From France 284 N.A. 14193
F.O.B. Exports of France to Pak 553 N.A. 11479

C.I.F. Pak Imports From W. Germany 2350 191 76588
F.O.B. Exports of W. Germany to Pak 2537 140 63788

C.I.F. Pak Imports From Italy 5820 N.A. 16666
F.O.B. Exports of Italy to Pak 1522 N.A. 16792

C.I.F. Pak Imports From UK 1047 480 49474
F.O.B. Exports ofUK to Pak 1240 341 44254

C.I.F. Pak Imports From Japan 339 N.A. 38772
F.O.B. Exports of Japan to Pak 5443 N.A. 41860

C.I.F. Pak Imports From Netherlands 3203 362 54946
F.O.B. Exports of Netherlands to Pak 368 2939 38737

 6 7 8

C.I.F. Pak Imports From France 21161 148331 4715
F.O.B. Exports of France to Pak 29842 46475 13800

C.I.F. Pak Imports From W. Germany 72096 106725 11097
F.O.B. Exports of W. Germany to Pak 63030 122462 14704

C.I.F. Pak Imports From Italy 27046 101755 4187
F.O.B. Exports of Italy to Pak 25985 200192 4856

C.I.F. Pak Imports From UK 48917 179918 12926
F.O.B. Exports ofUK to Pak 48038 169698 14776

C.I.F. Pak Imports From Japan 191821 321207 31751
F.O.B. Exports of Japan to Pak 199206 302188 25562

C.I.F. Pak Imports From Netherlands 6706 20064 1198
F.O.B. Exports of Netherlands to Pak 7967 19075 2123

Note: Where, one-digit level SITC numbers are: 0 = Food and food
products; 1 = Beverages and tobacco; 2 = Crude materials; 3 = Fuels
and lubricants; 4 = Animals and vegetable fats; 5 = Chemicals; 6 =
Manufactured goods; 7 = Machinery and transport equipment; 8 =
Miscellaneous manufactured goods.

Table 2
Under-invoicing of Imports in Pakistan Jan-Dec 1988
(in 000 US Dollars)

 0 1 2

C.I.F Pak Imports From France 20374 N.A. 3946
F.O.B. Exports of France to Pak 17997 102 4668

C.I.F. Pak Imports From W. Germany 11059 N.A. 13427
F.O.B. Exports of W. Germany to Pak 9399 N.A. 8614

C.I.F. Pak Imports From Italy 3374 N.A. 8531
F.O.B. Exports of Italy to Pak 524 N.A. 10421

C.I.F. Pak Imports From UK 4129 788 31938
F.O.B. Exports of UK to Pak 7303 1817 24071

C.I.F. Pak Imports From Japan 7894 N.A. 34066
F.O.B. Exports of Japan to Pak 7316 N.A. 44580

C.I.F. Pak Imports From Netherlands 28420 N.A. 7330
F.O.B. Exports of Netherlands to Pak 23199 209 11599

 3 4 5

C.I.F Pak Imports From France 373 N.A. 27678
F.O.B. Exports of France to Pak 444 N.A. 31375

C.I.F. Pak Imports From W. Germany 3661 4899 133714
F.O.B. Exports of W. Germany to Pak 572 2896 124635

C.I.F. Pak Imports From Italy 585 104 20582
F.O.B. Exports of Italy to Pak 1462 240 24565

C.I.F. Pak Imports From UK 3254 313 109920
F.O.B. Exports of UK to Pak 1855 203 111692

C.I.F. Pak Imports From Japan 1262 110 63108
F.O.B. Exports of Japan to Pak 435 N.A. 59870

C.I.F. Pak Imports From Netherlands 677 1666 48059
F.O.B. Exports of Netherlands to Pak 696 377 39879

 6 7 8

C.I.F Pak Imports From France 34637 180685 4679
F.O.B. Exports of France to Pak 46328 186514 12726

C.I.F. Pak Imports From W. Germany 71811 236791 23187
F.O.B. Exports of W. Germany to Pak 104946 233554 31272

C.I.F. Pak Imports From Italy 21645 143863 4503
F.O.B. Exports of Italy to Pak 35155 194535 8822

C.I.F. Pak Imports From UK 56271 189541 21628
F.O.B. Exports of UK to Pak 58707 224628 35162

C.I.F. Pak Imports From Japan 151116 658700 49855
F.O.B. Exports of Japan to Pak 182086 767890 54674

C.I.F. Pak Imports From Netherlands 7632 33670 2850
F.O.B. Exports of Netherlands to Pak 10894 23193 4409

Note: Where, one-digit level SITC numbers are: 0 = Food and food
products; 1 = Beverages and tobacco; 2 = Crude materials; 3 =
Fuels and lubricants; 4 = Animals and vegetable fats; 5 =
Chemicals; 6 = Manufactured goods; 7 = Machinery and transport
equipment; 8 = Miscellaneous manufactured goods.

Table 3
Extent of Under-invoicing in Pakistan (1981)

SITC Code France Germany Italy

621 NA 0.043 0.859
625 -0.421 0.792 NA
628 -0.141 0.535 -0.223
641 1.575 0.924 0.449
642 -0.534 0.268 -0.429
651 0.105 0.156 0.042
659 NA -0.583 0.188
662 -0.270 0.635 1.199
663 NA -0.420 -0.712
664 NA 0.312 NA
665 0.043 0.270 -0.309
673 1.817 -0.576 NA
674 -0.302 0.576 0.280
675 NA -0.796 NA
682 0.108 0.233 0.470
694 -0.735 -0.052 0.454
691 -0.370 5.998 0.351
692 NA -0.563 NA
694 -0.822 NA NA
695 -0.737 -0.128 -0.673
699 -0.690 -0.368 0.316
713 -0.937 -0.588 -0.526
714 0.581 0.852 0.493
724 -0.474 -0.056 0.096
727 NA 0.356 NA
728 -0.357 0.801 -0.541
736 NA -0.307 -0.442
741 -0.697 -0.595 -0.358
742 NA NA NA
743 -0.456 -0.027 -0.288
744 -0.014 0.133 -0.209
745 NA 0.226 -0.437
751 0.599 1.859 10.000
764 -0.885 -0.812 NA
771 4.273 -0.541 0.751
772 -0.575 0.417 0.211
773 -0.349 0.246 -0.712
775 NA NA 0.344
778 0.844 2.128 0.500
781 0.277 -0.684 NA
782 -0.810 -0.787 5.302
793 NA -0.259 2.793
794 1.937 0.203 0.195
821 1.343 -0.330 NA
881 NA -0.490 NA
882 NA 0.319 NA
885 NA NA NA
893 0.130 -0.082 0.618

SITC Code U.K. Japan Netherlands

621 NA -0.109 NA
625 -0.290 -0.105 NA
628 1.158 -0.511 NA
641 -0.061 -0.247 1.177
642 0.240 1.791 1.107
651 0.471 0.041 -0.151
659 -0.078 -0.636 NA
662 0.485 -0.171 NA
663 -0.203 -0.768 0.140
664 NA -0.408 NA
665 -0.409 -0.020 NA
673 1.930 -0.455 NA
674 0.248 0.091 -0.056
675 -0.804 -0.306 NA
682 -0.743 0.384 NA
694 0.773 -0.741 -0.507
691 -0.488 4.387 NA
692 -0.866 -0.036 NA
694 NA -0.508 NA
695 0.622 -0.237 -0.318
699 -0.259 -0.476 -0.448
713 0.575 -0.716 0.919
714 -0.962 9.357 NA
724 0.101 -0.029 0.350
727 -0.738 NA NA
728 -0.298 0.611 NA
736 -0.781 -0.361 NA
741 0.085 -0.629 -0.740
742 NA NA NA
743 0.295 -0.345 12.208
744 0.526 -0.426 -0.374
745 0.171 -0.384 -0.471
751 2.615 -0.450 -0.385
764 -0.121 0.608 -0.564
771 7.583 5.564 NA
772 -0.095 0.104 -0.052
773 -0.124 -0.290 NA
775 -0.862 -0.290 NA
778 -0.275 -0.013 0.325
781 -0.272 0.062 NA
782 1.052 -0.117 NA
793 1.273 NA NA
794 2.197 0.882 NA
821 -0.234 -0.487 NA
881 NA 6.326 NA
882 NA 0.028 NA
885 NA -0.443 NA
893 4.040 3.217 -0.358

Note: SITC Commodity codes and their specification are reported
in the Appendix.

Table 4
Extent of Under-invoicing in Pakistan (1988)

SITC Code France Germany Italy

621 -0.665 -0.756 -0.141
625 -0.022 -0.231 -0.321
628 -0.397 1.022 -0.132
641 0.059 0.207 -0.242
642 -0.365 2.209 0.026
651 0.928 0.605 -0.151
659 -0.535 -0.267 -0.600
662 NA -0.069 -0.488
663 -0.379 -0.180 -0.193
664 NA -0.105 NA
665 -0.339 0.443 -0.792
673 -0.890 -0.752 NA
674 1.388 1.921 NA
675 NA -0.737 NA
692 NA -0.371 -0.374
684 -0.655 0.124 0.043
691 NA 0.854 -0.681
692 -0.590 -0.883 0.471
694 NA -0.081 -0.930
695 -0.675 -0.389 -0.356
699 -0.461 -0.499 0.034
713 -0.698 -0.035 -0.163
714 0.661 3.839 0.497
724 0.221 0.219 -0.164
727 -0.382 -0.189 -0.699
728 0.716 0.934 -0.517
736 -0.716 2.735 0.138
741 -0.285 0.220 0.052
742 -0.650 -0.186 -0.601
743 -0.487 -0.286 -0.048
744 1.294 -0.642 -0.557
745 -0.384 -0.290 -0.148
751 NA 1.181 NA
764 0.038 0.207 NA
771 0.279 0.142 -0.567
772 -0.702 -0.392 -0.462
773 -0.954 -0.381 -0.219
775 -0.921 -0.910 -0.279
778 -0.647 1.773 2.001
781 -0.537 -0.621 -0.942
782 NA -0.839 -0.984
783 NA -0.498 NA
784 -0.463 -0.732 -0.299
821 2.319 -0.581 -0.650
881 2.410 -0.464 NA
882 NA -0.160 -0.688
885 NA NA NA
893 -0.183 0.568 0.544

SITC Code U.K. Japan Netherlands

621 -0.688 -0.422 1.775
625 0.055 -0.199 NA
628 1.524 -0.783 NA
641 -0.158 0.080 0.100
642 -0.166 -0.608 -0.554
651 0.104 0.202 -0.828
659 -0.029 -0.930 NA
662 -0.099 -0.833 NA
663 -0.138 -0.652 -0.488
664 0.342 -0.507 -0.013
665 -0.667 -0.464 NA
673 -0.424 -0.521 NA
674 0.605 0.800 0.181
675 -0.831 -0.749 NA
692 0.203 -0.145 -0.177
684 0.074 1.126 2.445
691 -0.259 -0.102 NA
692 -0.725 -0.799 NA
694 -0.625 -0.773 NA
695 -0.183 -0.428 -0.131
699 -0.165 -0.341 -0.140
713 0.535 -0.690 NA
714 -0.887 NA 5.363
724 -0.288 -0.441 NA
727 2.561 2.789 0.802
728 1.088 1.572 2.088
736 3.311 5.157 NA
741 0.115 -0.802 -0.702
742 -0.374 -0.367 NA
743 -0.312 -0.673 -0.240
744 0.194 -0.118 NA
745 -0.024 -0.555 -0.814
751 0.787 0.115 0.613
764 -0.212 1.778 0.638
771 0.832 -0.176 0.537
772 -0.366 -0.646 -0.361
773 -0.357 -0.753 NA
775 -0.459 -0.363 NA
778 -0.109 0.282 -0.428
781 -0.420 0.032 NA
782 0.981 -0.117 NA
783 0.535 1.010 NA
784 -0.531 0.028 NA
821 -0.641 -0.517 -0.330
881 NA -0.452 NA
882 3.357 0.092 -0.824
885 NA -0.455 NA
893 1.139 -0.733 -0.675

Vote: SITC Commodity codes and their specification are reported
in the Appendix.