On overinvoicing of exports in Pakistan.

Mahmood, Zafar ; Azhar, Mohammad

1. INTRODUCTION

Pakistan launched its programme of industrialisation under the influence of import-substituting (IS) strategy of advancing development. The IS strategy, pursued for many decades, however, has not satisfied the goals of the industrial policy. Dissatisfaction with the outcome of IS policies has lead to a partial shift from the IS strategy to the export-promoting (EP) strategy.

In the process of this strategic shift, Pakistan has offered many attractive incentives to encourage the production of export-oriented manufactured products. Whereas export incentives have helped the country to manage a respectable growth rate in exports, exporters have exploited the weaknesses of the incentive system and developed some unfair export practices (by overinvoicing the value of transaction). These illicit practices result in a significant financial loss to the country and undermine the effectiveness of the policies to achieve their stated goals. At the same time, many exporters who do not indulge in such practices have to suffer losses, as their bargaining position in the marketplace is affected adversely.

The menace of overinvoicing of exports calls for immediate attention of policy-makers, because without recognising it they may limit the scope of the export-promoting schemes and the trade liberalisation programme. Ideally, the implications of export overinvoicing should be integrated with the usual policy prescriptions. Likewise, it is important to determine the presence and magnitude of export overinvoicing, so that policy-makers are reminded of the extent of the problem.

Despite this well-recognised problem in Pakistan, no systematic study is available on the subject. To fill this gap, this paper shows its presence and estimates the geographic and product-wise patterns in export overinvoicing. The paper concludes on a set of policy recommendations to curb the menace of export overinvoicing.

The layout of the paper is as follows. Section 2 explains reasons for overinvoicing of exports, while Section 3 describes export overinvoicing and quantifying through the empirical approach. Export-promoting policies adopted during the period of the study are discussed in Section 4. In Section 5, empirical findings are reported. Finally, Section 6 gives policy suggestions.

2. EXPLAINING THE OVERINVOICING OF EXPORTS

Why exporters indulge in unethical and unfair trade practices? An exporter is tempted to overinvoice exports (1) if (say) the duty drawback (2) rate is higher than the premium on foreign currency that he has to purchase from the kerb market to meet the export-earning surrender requirement of the State Bank. When there are no foreign exchange controls, so that all exchange transactions take place only within an insignificant range legally permitted around the parity value, i.e., the premium is nil, then there is a clear incentive to overinvoice exports in the presence of a scheme of duty drawback. Where, however, there is exchange control, and the exporter must surrender all declared export receipts to the State Bank, the exporter will have to purchase illegal foreign exchange in the kerb market to an amount excess of the declared over actual value of exports. In this case, if the kerb market premium on foreign exchange is less than the rate of duty drawback, then again the exporter will be induced to overinvoice exports. There is, however, some risk (of being caught by law enforcement agencies) attached to getting involve in illegal activities. Hence, overinvoicing of exports will not cease unless the differential between the duty drawback and the premium in the kerb exchange market is greater than the risk factor evaluated by the exporter.

3. METHODOLOGY

Under normal circumstances, one would expect a trading partner country's statistics to show excess of carriage and freight (c & f) import values over the corresponding free on board (f.o.b.) export values of the same trade commodity. But if the observed discrepancy is in the reverse direction, and there is no other reason for the discrepancy, one may conclude the presence of overinvoicing of exports. This inference will be more certain if it can be established (i) that for these commodities the duty drawback rates are higher than the kerb market premium rate on foreign exchange and (ii) that these commodities are such that it is relatively easy to overinvoice them because of the nature of the products.

Following Bhagwati (1964, 1967); Mahmood (1997); Mahmood and Nazli (1999); Naya and Morgan (1969) and Simkin (1970), we use the partner-country-data comparison technique to test for overinvoicing of exports. In this technique, cost, insurance and freight (c.i.f.) import value of the partner country are compared with the free on board (f.o.b.) export value of the concerned country to find unexpected discrepancies in exports. Using this approach, export overinvoicing is defined in the following way:

MIS = MIC - XP*AD

Where,

MIS = Misinvoicing of exports.

MIC = C.i.f. imports of industrial countries from Pakistan.

XP = F.o.b. exports of Pakistan to industrial countries.

AD = Adjustment factor defined as c.i.f.-f.o.b, ratio.

MIS < 0, implies overinvoicing of exports.

To show such discrepancies, we use data available in the 'Commodity Trade Statistics' of the United Nations and 'Direction of Trade Statistics' of the International Monetary Fund. The data for c.i.f.-f.o.b, adjustment factor are obtained from International Monetary Fund (1984, 1992, 1994).

One should also note the following alternative reasons (3) that can be put forward to explain the unusual discrepancies:

1. The method presumes the faking of invoices to occur only in one country. If both countries fake invoices, it becomes impossible to make a case for overinvoicing of exports. One-sided fake invoicing can, however, be assumed given strict enforcement of regulations in the partner countries, especially the developed ones.

2. There can be 'mis-allocations' of the same traded item by commodity and country. These inconsistencies can arise from both genuine customs mistakes and differences in classifications adopted by trading partners. Cross-checking of commodities by countries enables one to determine a possible explanation of unusual discrepancies.

4. EXPORT-PROMOTING POLICIES

Realising the distortionary effects of the import-substitution policies, Pakistan has pursued export-promoting policies since the late 1950s. One of the prominent tools of this policy adopted in 1959 was the export bonus scheme, which introduced multiple exchange rates in the country. (4) This scheme was abandoned in 1972 after the devaluation of Pak rupee. In order to promote export of manufactured goods, since the late 1970s, an elaborate export incentive system was introduced. This system included: (i) exemption of exports from sales tax and central excise duty, (ii) duty drawback scheme covering sales tax, central excise duty, and customs duty on inputs used in the production of exports, (iii) taxation of export of domestically produced raw materials, (iv) concessionary export finance and export credit guarantee scheme, and (v) income tax rebates.

Out of the above list of incentives, three measures induce exporters to indulge in export overinvoicing: concessionary export finance, duty drawback on exports, and income tax rebate on profits from exports. Concessionary export finance provides short-term pre-shipment and post-shipment finance in local currency to direct exporters. Limited availability of concessional export finance induces exporters to overinvoice the value of exports so that they can obtain a large size of loan from the bank. In 1994, the annual interest rate on concessional loans was 11 percent, with a maximum loan period of 180 days. (5) In contrast to this, the market interest rate was 22-23 percent. About 70 percent of exports fell within the concessionary export credit scheme. In 1993, US$ 2.7 billion of export credits were extended to exporters. Export refinance borrowings are either exempted from the payment of excise duty or taxes are included in the export rebates.

Pakistan allows drawback on customs duties and sales taxes on imported inputs, and on excise duties and sales taxes on indigenous inputs. Rebates for 54 broad industrial groups covering more than 250 products have been standardised as a specified percentage of the f.o.b, value of export or a specific amount per unit of goods exported. The customs duty drawback rate, as a percent of manufactured exports on which duty drawback is applied, was 12.5 percent in 1994. (6) The gap between the duty drawback rate and the exchange rate premium attracts exporters to indulge in overinvoicing exports. This is mainly due to the fact that with little effort, corrupt exporters with the connivance of corrupt government and bank officials manage to receive drawbacks from the exchequer. They, in fact, exploit the weaknesses of the duty drawback system (DDS) while indulging in unfair trade practices. The major weaknesses of the DDS are as follows:

1. Duty drawback rates are not automatically adjusted to take account of additional taxes imposed by the government from time to time.

2. Due to fixed nature of the system, exporters in general are unable to draw back the full amount of taxes and duties paid by them.

3. The Revenue Division sometimes confronts exporters with an arbitrary downward revision in export prices without any prior notification.

Up to 1988, the government allowed income tax rebate at the rate of 55 percent of the profits earned through exports of manufactured goods. In 1989, the government introduced a three-tier system, in which the income tax rebate was graduated with the degree of processing. On semi-manufactured goods, such as cotton yarn, the income tax rebate was 25 percent. On manufactured exports, in general, it was 50 percent. But on the exports of some manufactured goods, viz., leather garments, engineering and electrical goods, the rebate was 75 percent. In 1993, exports of furniture, doors, and windows were also allowed 75 percent income tax rebate. Later on, the income tax rebate rate was increased to 90 percent in the case of all goods receiving a rebate of 75 percent. Thus by showing large export value out of their total domestic production, exporting firms earn a large income tax rebate.

5. EMPIRICAL ANALYSIS

Alternative explanations of unusual discrepancies in exports, described in Section 2, show the practical difficulties which one faces when drawing conclusions about overinvoicing of exports. Nonetheless, this paper shows that there is enough evidence to provide an explanation for overinvoicing of exports in Pakistan. A comparison of the official exchange rate and the kerb market exchange rate suggests that exporters lose by paying about 4.5 percent more in the kerb market to buy foreign exchange in order to make payments to the State Bank but gain in terms of duty drawbacks by about 12.5 percent by overinvoicing exports. (7)

The analysis begins by establishing that if overinvoicing of exports takes place at the aggregate level, then it can unambiguously be established at the refined level of Standard International Trade Classification (SITC), such as at three-digit or four-digit levels. For the aggregate level, Table 1 reveals that exporters overinvoiced exports to the tune of US$ 2.4 billion over the period 1984 to 1994, i.e., on average US$ 240 million per annum. This amount is equivalent to about 28 percent of the total export value of any recent year in Pakistan.

The above estimates of export overinvoicing are further confirmed from unpublished estimates of the Revenue Division on amounts deducted as overclaimed for duty drawbacks (see Table 2). On the basis of the current level of duty drawback claims made by exporters and the scrutiny made by the Collectorate staff, the potential revenue loss prevented in Karachi was US$ 29.3 million, or 25 percent of the total claims filed during the year 1997-98. The Karachi Collectorate accounts for about 50 percent of the nationwide rebate claims but represents about 80 percent of the high-risk exports. Interestingly, whereas these figures confirm the presence of export overinvoicing by the official source, the low size of fake invoicing shown by these figures confirms weak enforcement of the law by the Revenue Division.

Aggregate estimates of export overinvoicing are subject to two limitations: (i) they only portray the aggregate picture and hence hide many facts regarding geographic and product-wise patterns in export overinvoicing, and (ii) they are net of export underinvoicing and hence under-report the actual size of export overinvoicing. Because of these limitations, we provide below estimates of export overinvoicing with a focus on geographic and product-wise patterns.

For the sake of analysis, we consider only those products which are generally overinvoiced and have a major stake in exports; namely, leather garments (SITC-612), cotton cloth (SITC-652), man-made woven fabrics (SITC-653), made-up textiles (SITC-658), linen (SITC-6584), women's garments non-knit (SITC-843), undergarments knitted (SITC-846), headgear non-textile fabrics (SITC-848), surgical goods (SITC-872), and sports goods (SITC-894). These products accounted for 42 percent of the total exports in 1994. (See Table 3.)

All of the selected products testify to overinvoicing of exports. The last column of Table 4 shows that for three selected years, (8) 1984, 1992 and 1994, overinvoicing of exports is quite visible: US$ 54.82 million in 1984, US$ 413.67 million in 1992, and US$ 455.78 in 1994. In 1994, the degree of overinvoicing of exports was about 7 percent. For all the selected years and for all the selected products, exports were overinvoiced, with the exception of made-up textiles in 1984, women's garments non-knit in 1984, and sports goods in 1992. Four major export products overinvoiced in 1994 were cotton cloth (US$ 89.8 million); women's garments non-knit (US$ 85.5 million), made-up textiles (US$ 85.5 million), and man-made woven fabrics (US$ 60.0 million).

Thirteen major importers of selected products included in the analysis are: Australia, Belgium, Canada, France, Germany, Hong Kong, Italy, Japan, Netherlands, Sweden, Singapore, UK, and USA. Collectively, these countries accounted for 62.4 percent of the total exports in 1994. Overinvoicing of exports in every product is taking place with every major trading partner of Pakistan (see Table 4). Overinvoicing of exports in case of each selected product and for every selected year is found for Australia, Belgium, Canada, and Hong Kong. However, for other countries, in case of a few products, export overinvoicing was not present in each year. For instance, in the case of Japan, overinvoicing was not present in headgear non-textile fabrics (in 1984) and in sports goods (in 1992 and 1994). Likewise, one can notice such cases for other countries. Despite the absence of overinvoicing of exports in a few cases, the overall overinvoicing status remains unchanged. In fact, if these cases are excluded from the analysis, the size of overall overinvoicing will go up further.

Finally, in order to confirm that the discrepancies reported above only arise because of overinvoicing of exports, offsetting discrepancies were thoroughly checked. First, after summing up all countries by individual commodity groups, there remain discrepancies. This confirms that there are no 'mis-allocations' of the same export good by commodity and country. Secondly, by including only those countries which have relatively free trade and strict enforcement of regulations, and by using a relatively large number of products, a strong attempt has been made to eliminate the effects of other factors which may give rise to the discrepancy. Alternative reasons for these discrepancies do not seem to provide a satisfactory explanation and lead to the conclusion that they are there due to the overinvoicing of exports.

6. CONCLUSIONS AND POLICY RECOMMENDATIONS

This paper has shown the existence of discrepancies in selected export statistics for which the only explanation appears to be overinvoicing. This argument can be supported with the evidence of significant differences between the duty drawback rate and the premium on the kerb market foreign exchange rate, which turned out to be 8 percent. To this effect, if the impact of income tax rebate and concessional export finance on account of export overinvoicing is added, then this percentage will further rise. These convincing indications of the presence of overinvoicing of exports lead us to put forward the following policy recommendations:

1. The World Trade Organisation (WTO) does not treat duty drawbacks at final stages and early stages of export production as "export subsidy". Therefore, the introduction of an efficient duty drawback system will give tariff-free and indirect tax-free status to exports, and will be instrumental in curbing export overinvoicing. For this development the following measures need to be introduced:

(a) Increase inspection and examine all export products subject to specific duty drawback rates.

(b) Re-adjust promptly all the announced and revised duty drawbacks and other policy measures. This is crucial to exporters for negotiating export contracts with foreign buyers. Downward revisions of drawback rates should be made effective after a sufficient time lag to allow full accommodation of import costs based on earlier higher duties and tax rates.

(c) Adopt a method of determining duty drawbacks based on proper inputoutput coefficients, aided by professional engineers and accountants. (9)

(d) Introduce a mechanism to ensure proper recording of export prices of rebateable products. This will protect exporters from the occasional arbitrary downward adjustments of export prices by the DDS administration and prevent the overinvoicing of exports to benefit from duty drawback.

(e) Introduce export product sample testing to verify blend ratios. Testing should be outsourced to well-reputed testing laboratories.

(f) All bank credit advice (BCAs) should be channelled through the State Bank of Pakistan and be duly authenticated before drawback payments are made. This will reduce the risk of forged/fake BCAs which reflect higher realisation of export proceeds or submission of altogether fake BCAs covering ghost exports.

(g) Use the individual drawbacks for major imported inputs used to produce each export item, and fixed drawbacks for miscellaneous imported inputs.

(h) Fully document all the export operations, i.e., processing, exporter profiling, export examination, and drawback claim payment at all export stations.

2. Shift emphasis of Export Finance Scheme from providing preferential interest rate loans to a few large direct exporters to an emphasis on providing easy access to export finance for all exporting activities at the market (competitive) interest rate, on the basis of confirmed export orders. The overall amount allocated for the purpose should be raised to accommodate all targeted exports.

3. Income tax rebates to exporters amount to a subsidy on exports; the WTO members may contest these. The Government of Pakistan should review this policy. This should enable the government not only to fulfil its general obligations to the WTO but also lessen the problem of export overinvoicing.

4. The Government should take appropriate measures to eliminate the Hundi system, which is one of the major sources of finance for exporters who overinvoice exports.

5. Eliminate redundant tariffs as a first step towards lowering or removing all tariffs. The Government should continuously review its tariff schedule. All of this should reduce the administrative burden of drawback administration.

6. There is a need for a comprehensive policy on customs bilateral pact with our major trading partners. The mutual administrative assistance in customs matters should aim at sharing intelligence and other trade-related information between any two countries to curb fake invoicing. The idea here is to curb unhealthy trade practices that are detrimental to growing global integration of markets.

7. Export contract value should be made the basis of duty drawbacks. If the invoice value of the exporter is as per the contract value with the trading partner, there will be no need for trade verification.

8. The Government should make a greater use of penalties covering ineligibility for duty drawbacks to those exporters who indulge in overinvoicing.

REFERENCES

Bhagwati, J. N. (1964) On the Underinvoicing of Imports. Bulletin of the Oxford University Institute of Economic and Statistics 26, 389-397.

Bhagwati, J. N. (1967) Fiscal Policies, the Faking of Foreign Trade Declarations, and the Balance of Payments. Bulletin of the Oxford University Institute of Economics and Statistics 29, 61-77.

Bhagwati, J. N. (1974) Introduction. Bhagwati, J. N. (ed.), Illegal Transactions in International Trade. Amsterdam: Elsevier. 1-6.

International Monetary Fund (Various Issues) Direction of Trade Statistics. Washington, D. C.

International Monetary Fund (1982, 1992, 1994) International Financial Statistics. Washington, D. C.

Mahmood, Z. (1997) Determinants of Underinvoicing of Imports in Pakistan. Journal of International Development 9:1, 85-96.

Mahmood, Z., and H. Nazli (1999) Estimates of Unrecorded Private Capital Movements. Economia Internationale 52:1, 79-90.

Naqvi, S. N. H. (1966) Allocative Biases of Pakistan Commercial Policy. The Pakistan Development Review 6:4, 465-499.

Naya, S., and T. Morgan (1969) The Accuracy of International Trade Data: The Case of Southeast Asian Countries. Journal of American Statistical Association 64, 452-467.

Pakistan, Government of (1997) Economic Survey, 1996-97. Economic Advisor's Wing, Finance Division, Islamabad.

Pakistan, Govemment of (Various Issues) Export Policy Order. Ministry of Commerce, Government of Pakistan.

Revenue Division (1999) Drawback Claims: Karachi Export Collectorate (1997-98). Government of Pakistan. (Unpublished).

Simkin, C. G. F. (1970) Indonesia's Unrecorded Trade. Bulletin of Indonesian Economic Studies 6.

United Nations (Various Issues) Commodity Trade Statistics. New York: United Nations Statistical Office.

Authors' Note: We wish to thank Dr Nadeem A. Burney for his useful comments on an earlier draft of the paper. Comments by anonymous referees of the PDR are highly appreciated. The views expressed in the paper are those of the authors and do not necessarily reflect the views of the Institutes at which they work.

(1) Major instruments used for export overinvoicing include: misdeclaration of quantity, misdeclaration of value, misdeclaration of blend ratios, and presentation of forged bank credit advice.

(2) Export incentives that are misused to overinvoice exports are: duty drawbacks, concessional export finance, and income tax rebate.

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

(4) For a detailed discussion of the export bonus scheme, see Naqvi (1966).

(5) The interest rate on concessionary export credit has been increased over time. Export finance was available at only 3 percent interest rate up to 1986. It was raised to 6 percent in 1987. The interest rate was further increased to 7 percent in 1991, to 8 percent in 1992, and to 11 percent in 1994. Later, it was lowered to 9 percent but again raised in 2001 to 10.5 percent.

(6) To this rate of duty drawback, one can add the effect of higher concessional export finance obtained through export overinvoicing, and that of rebate earned on account of income tax. Obviously, the effective benefit earned through export overinvoicing on account of all these leakages will be much higher than 12.5 percent. Non-availability of the required data keep us from estimating the total effective rate of this benefit.

(7) The current kerb-market premium is about 4.5 percent. It came down from 22 percent in 1981, when the country had a fixed exchange rate system, to 8.7 in 1988, when a managed floating exchange rate was in force [Mahmood (1997)]. Now, with the introduction of inter-bank exchange market, the kerb-market exchange rate has further come down. Thus cost of overinvoicing has gone down over this time.

(8) The year 1984 is selected because this was the year when trade liberalisation began to have its initial impact. 1992 is selected because widespread trade liberalisation programmes were initiated in Pakistan that year. The year 1994 is selected as the last year for which data are available.

(9) The recent announcement of the establishment of Input-Output Coefficient Organisation is a right step. The successful experience of East Asian countries in this regard suggests that the work of this organisation should enable the Revenue Division to adopt more practical and effective methods to determine the value of drawbacks on exports.

Zafar Mahmood, Chief of Research, Pakistan Institute of Development Economics, is currently with Kuwait Institute for Scientific Research. Mohammad Azhar is Associate Staff Economist at the Pakistan Institute of Development Economics, Islamabad.

Table 1
Export Misinvoicing in Pakistan (Million US Dollars)

Year MIC XP AD MIS

1984 1210 1199 1.095 -102.91
1985 1529 1376 1.095 22.28
1986 1965 1888 1.095 -102.36
1987 2487 2510 1.095 -261.45
1988 2789 2690 1.095 -161.93
1989 2849 2736 1.095 -146.92
1990 3441 3400 1.095 -291.00
1991 3724 3637 1.095 -257.52
1992 3959 4001 1.095 -421.10
1993 3960 3933 1.095 -346.63
1994 4504 4448 1.095 -366.56

Note: (-) Value means overinvoicing of exports.

Table 2
Drawback Claims: Karachi Export Collectorate (1997-98)

(Million US Dollars)

 Amount
 Sanctioned and Amounts
No. of Exporters Amount Paid after Deducted as
Making Claim Claimed/Filed Scrutiny Over-claimed

(a) 78,144 66.96 66.96 --
(b) 14,933 15.60 15.01 -0.59
(c) 9,552 34.29 5.62 -28.67

102,628 116.85 87.59 -29.26

(Million US Dollars)

 Ratio of
 Refund
No. of Exporters Payments
Making Claim (%)

(a) 78,144 100.0
(b) 14,933 96.2
(c) 9,552 16.4

102,628 75.0

Source: Revenue Division (1999, unpublished).

Table 3
Exports by Commodity Prone to Overinvoicing

(Million US Dollars)

Commodity Year
 1984 1992 1994

Textiles 395.97 950.18 1080.81
Garments 191.87 773.71 847.52
Surgical Goods 41.16 70.44 71.65
Sports Goods 45.81 99.46 160.61
Total Exports 2575.01 4999.40 5167.27

Source: Economic Survey, 1997-9X.

Table 4
Sire of Overinvoicing of Exports by Olajor Importing Countries( from
Pakistan): Three- and Four-digit Level of SlTC

(Thousand US Dollars)

Commodity Country
Year
 Hong
 Japan Kong Belguim Germany Itally

Leather
Garments (612)

1992
1994 -871

Cotton Cloth (652)

1984 529 -1132 -2964
1992 -2090 -12345 -9514 -5304 -5495
1994 -5038 -10042 -22264 -3486 -3886

Man-made N'oven
Fabrics (653)

1984
1992 -11020 1623 10450
1994 -3291 -12642 43 8753

Made-up
Textiles (658)

1984 -1826
1992 -5226 -2142 -18405 -18405 -7267
1994 -2897 -8440 -4873 -1329

Linen (658x)

1984 -1364
1992 -3277 -5533 -7897 -6197
1994 -778 -6506 4780 -3057

Women's Garments
Non-lmit (843)

1984 256
1992 -14410 -2870
1994 -1943 -14012 466
1984 -232
1992 -9440
1994 -2890 -10841 640

Headgear Noutextile
Fabric (848)

1984 -3248 -199
1992 -280 -2402 -3683 -1537
1994 284 -3325 -997 -2360 1101

Surgical
Goods (872)

1984 -3367
1992 -1789 432
1994 -1102 -811 -341

Sports
Goods (894)

1984 -540 -440
1992 1405 -1268 -195 -103
1994 2618 -717 -3995 -8653 -1618

Total

1984 -- 529 -- -11453 -3603
1992 -9468 -14487 -36394 -59700 -13451
1994 -6913 -17375 -59676 41084 -3551

Commodity Country
Year

 OSA Ftamz UK Austra-lia Netherland

Leather
Garments (612)

1992 -2627
1994 -2295

Cotton Cloth (652)

1984 -866 -7680 -9370 -2864
1992 3656 1167 -3854
1994 -8044 13689 -22684 -2861 -5371

Man-made N'oven
Fabrics (653)

1984 -935
1992 -21352 -16186 -3327
1994 26139 -1033 11712 -4239 -6889

Made-up
Textiles (658)

1984 8226 -2309 -7184 -1334
1992 -9389 2541 -2665 -12270
1994 29477 -4011 -7908 -3441 -16553

Linen (658x)

1984 2677 -2100 -2083
1992 8093 -1060 -2201 -10881
1994 -2102 2675 -10624 -1528 -17GG1

Women's Garments
Non-lmit (843)

1984 10017 731 25
1992 -62507 -9039 -5215
1994 -52931 -2GG9 6852 -3053
1984 -1695 -666 -568 219
1992 -26740 -155 1187
1994 -20732 -4223 -1722 1031

Headgear Noutextile
Fabric (848)

1984 697 67 129 -352
1992 -32201 -1534 -3825
1994 3083 2174 2308 -227 -3069

Surgical
Goods (872)

1984 -2862 -11592
1992 -20245 -1354
1994 20359 -1107 -2621 -780 2355

Sports
Goods (894)

1984 X33 1240 -12231
1992 3504 4254 -994 -689
1994 -7161 2724 -7285 -953 -2735

Total

1984 14367 -11448 -38103 2864 -1492
1992 -159808 -30851 -- -13041 -37067
1994 -166157 2869 62100 -13029 -6717

Commodity Country
Year

 Sweden Singapore Canada Total

Leather
Garments (612)

1992 -2637
1994 -3166

Cotton Cloth (652)

1984 -24327
1992 -4427 6535 -146;1
1994 -3903 -11306 -4609 -89805

Man-made N'oven
Fabrics (653)

1984 -935
1992 -6929 -2966 -49707
1994 -7373 -2402 -88 -60012

Made-up
Textiles (658)

1984 -165 108
1992 -2222 -4274 -73058
1994 -1366 -377 -2927 -65499

Linen (658x)

1984 -288 -3158
1992 -2858 -3401 -35212
1994 -935 -236 -1094 -26516

Women's Garments
Non-lmit (843)

1984 10978
1992 -6357 -100398
1994 5535 -85528
1984 -2942
1992 -11925
1994 -287 767 -2475 -17073

Headgear Noutextile
Fabric (848)

1984 4300
1992 2146 -1888 -18205
1994 2042 667 347

Surgical
Goods (872)

1984 -17821
1992 -23820
1994 -29474

Sports
Goods (894)

1984 -12404
1992 300 5914
1994 9 149 -1438 -29055

Total

1984 -753 -- -- -54820
1992 -13990 -- 25411 -413669
1994 -11526 -13605 -17833 -455781

Note: Despite our best efforts, data for the U.K. for the year 1992
could not be obtained.