VII International Shipping and Transportation Services
Author:
International Monetary Fund
Search for other papers by International Monetary Fund in
Current site
Google Scholar Close
Language:
English
Print Publication Date:
15 Sep 1987
Keywords:
BOOK; debit; balance of payments account; current account; investment; transaction; shipment debit; shipment credit; debits need; ocean freight revenue; transportation account; Wages; Transportation; Credit; Balance of payments statistics; Imports; Eastern Europe; Middle East
Download PDF (1.9 MB)

Abstract

INTRODUCTION

1. INTRODUCTION

Although the main task of the Working Party was to explore the origin of the growing discrepancy in the international investment income accounts, within the time and resources available an analysis of the persistent discrepancies in the shipping and transportation services accounts was initiated. This chapter attempts to identify the essential ingredients of any study of these accounts and reports on the results of the preliminary work on this topic undertaken by the Working Party.

Transactions involving the international carriage of merchandise and passengers are recorded in either of two categories in the Fund’s classification of balance of payments statistics: “shipment” or “other transportation.” The “shipment” account records the cost of freight, insurance, and other distributive services. Under the conventions adopted for balance of payments reporting, these services are deemed to be provided to (and paid for by) the importer, regardless of the actual channels of payment. Thus they become international transactions, to be included in the balance of payments, only if they are provided by a resident of a country other than that of the importer. Under “shipment” credits should be reported revenues (gross receipts) from nonresidents—that is, freight on exports and cross-trades—earned by all vessels or other means of transportation operated by residents of the reporting country, regardless of the flag of registry. This includes the earnings of foreign-flag vessels “chartered in,” as well as vessels owned and operated by domestic residents. The converse, of course, applies to “shipment” debits, which should cover freight on imports in vessels operated by nonresidents, including domestic-flag vessels “chartered out” to nonresident operators.

The “other transportation” account includes international passenger fares, expenditures by international carriers in countries other than that of the operator of the carrier, charter fees paid to nonresident shipowners, and other transportation services when provided internationally.

The first of these two accounts has consistently shown a considerable excess of debits in recent years, with no discernible trend from 1979 to 1983. The asymmetry on “other transportation” has been small (Table 64).20 It is important to note that there is a relationship between the “shipment” account and that part of “other transportation” that consists of expenditures of foreign carriers in domestic ports—bunker fuel, ships’ stores, port charges, and the like. However, the relationship is not necessarily the same from country to country, nor stable over time, when reflected in balance of payments accounts. It will be affected by the volume of transactions that do not enter the balance of payments (freight on imports carried in, and expenditures in domestic ports of, nationally operated carriers), the geographic composition of freight earnings (with a high ratio of foreign port expenditure to revenues, for instance, possibly accompanying a high proportion of revenues from cross-trades), and the composition of the revenue by type of vessel (liner, bulk dry cargo, tanker).

Table 64

INTERNATIONAL SHIPPING AND TRANSPORTATION ACCOUNTS, 1979–83

(In billions of U.S. dollars)

article image
Source: International Monetary Fund, Balance of Payments Statistics Yearbook, Vol. 36 (Washington, 1985), Part 2.

While the debits on “shipment” may be overstated by some countries,21 it is virtually certain that the large discrepancy on “shipment” reflects mainly the omission, on the credit side, of the revenues of a large portion of the world shipping fleet—principally the earnings of fleets operated by Hong Kong, Eastern European,22 and Greek enterprises—from recorded balance of payments data. Hong Kong and Eastern Europe are not included in the Fund’s balance of payments statistics, so far as the “other transportation” accounts are concerned. The Greek balance of payments data exclude the operations of the Greek fleet,23 one of the largest in the world. There are a few countries that report no shipment credits, in spite of the fact that they have ships under national registry. And there is a possibility that some countries do not capture in their statistics the transactions of all vessels under domestic operation, particularly foreign-flag ships “chartered in” to domestic operators.

In any event, the relative stability over time of the ratio between the discrepancy on shipment (fourth line of Table 64) and total “shipment” debits indicates that the principal cause of the discrepancy has probably been correctly identified as the “missing” fleets. The ratio between recorded “shipment” debits and the total insurance and freight component of world trade has also been fairly constant over the 1979-83 period, ranging between 75 and 78 percent. These results are imprecise, in that the insurance and non-maritime freight components are included. But they imply that there has not been much variation in the proportions of world trade carried by three major categories of carriers, namely (1) the importing countries’ own vessels; (2) vessels operated by other countries, figures for which are included in the Fund’s data; and (3) the “missing” fleets.

In order to address these issues, the Working Party requested 26 countries, including all the major maritime countries supplying balance of payments data to the Fund, to complete a questionnaire giving a detailed breakdown of the “shipment” and “other transportation” accounts in their balance of payments statistics. (The questionnaire is reproduced in Appendix VII.)

In the meantime, the Fund’s Bureau of Statistics conducted an extensive examination and revision of the data it collects from member countries on the ratio between the c.i.f. and f.o.b. values of their imports. These data, as revised,24 provide the only available measure, albeit an indirect one, of the freight and insurance on imports into all countries other than the Union of Soviet Socialist Republics and other economies not reporting to the Fund.

In this chapter, we present our analysis of existing asymmetries in the shipping and transportation accounts. While we believe we can identify the causes of most of the discrepancies, much more work is needed to improve reporting by countries and to generate precise adjustment factors where necessary’. This means testing each country’s entries in the balance of payments accounts for internal consistency with the data on exports and imports, and with whatever information is available, from domestic or external sources, regarding international transportation—size of the national fleet, freight rates, chartering practices, and other relevant factors. More specific recommendations for further research are presented in the concluding section of this chapter.

2. SHIPMENT

In order to determine how much of the asymmetry on shipment account can be attributed to the revenues of fleets entirely omitted from the Fund’s statistics (mainly Greece, Hong Kong, and Eastern European countries that were not members of the Fund—plus a few reporting countries which, although they possess merchant marines, show no shipment credits), it is necessary first to evaluate the reported data. This involves checks on several elements of the debit and credit sides of the accounts. Also, we need to distinguish, to the extent possible, the revenues and expenses of marine transport from those of other means of transportation.

a. Evaluation of Debits

As already indicated, the debits on shipment exceeded recorded credits by $31.8 billion in 1983. While it might be assumed, as a first approximation, that this amount represents primarily unreported shipping revenues, the accuracy and appropriateness of the debit totals also need to be tested. That is, do the debits include items that would have no counterpart in a “shipment” credit reportable, in principle, either by a reporting country or by the operators of the “missing” fleets? For instance, if a country overestimated the amount of freight included in the c.i.f. value of its imports, it would ipso facto understate the f.o.b. value of its imports. The partner countries would, in principle, report a higher export value and lower shipping receipts. However, the total current account discrepancy would not be affected.

It has been suggested that the data on the relationship between the c.i.f. and f.o.b. values of imports, as published in the Fund’s International Financial Statistics Yearbook, do in fact overstate the implied insurance and freight costs, especially for developing countries. This may be the case, but the evidence is not conclusive. It is true that, over the last decade or so, the c.i.f./f.o.b. ratio has declined significantly for industrialized countries (reflecting mainly a relative decline in freight rates compared with merchandise values) and has not declined for the developing countries. In part, at least, the latter phenomenon reflects a great increase in the proportion of Middle Eastern countries’ imports in the world total, coupled with the fact that freight and insurance rates to that area are, in general, higher than to most of the rest of the world.

Aside from the c.i.f./f.o.b. problem, and since we are focusing our attention at this point on missing ocean freight revenues,25 our “shipment” debits need first to be purged of payments made for other purposes—mainly for (1) insurance and (2) freight carried on other means of transport.

(1) Insurance

In principle, insurance debits should include only payments to nonresident insurers by importers of the merchandise. It seems likely that this would occur mainly when goods were invoiced on a c.i.f. basis. In that case the exporter would probably place the insurance in his own country, resulting in a payment by the importer to a nonresident insurer through the exporter. On the other hand, in the case of f.o.b. or f.a.s. billings, it seems likely that importers would in most cases place their insurance domestically. In response to our questionnaire, countries accounting for 48 percent of world shipment debits reported insurance debits equal to 4.8 percent of total shipment debits. This ratio reflects (1) the ratio of insurance costs to the f.o.b. value of the merchandise and (2) the share of the insurance placed abroad. Based on this sample, it can be estimated that reported shipment debits of $78.7 billion include insurance debits of $3.8 billion, 4.8 percent of the total.26

(2) Other Means of Transport

Respondents to our questionnaire survey reported total shipment debits of $5.6 billion relating to non-maritime means of transport. As a measure of the total included in the Fund’s data, this is obviously a minimum figure; receipts of $6.6 billion were reported on the questionnaires. However, because of the wide variety of means of transport involved, there seems to be no practical way to expand it to a world total. As a preliminary step, therefore, we assume that reported world shipment debits include $5.6 billion of payments for nonmaritime shipping services.

(3) Other Adjustments

There are also certain missing debits that should be added because they are reportable in principle as credits, either by countries actually reporting credits or by the operators of the “missing” fleets, whose earnings we are trying to measure. First are the shipment debits related to exports from the “world” (as defined in the Fund’s Direction of Trade Statistics) to the nonmarket countries that were not Fund members. Such exports, on an f.o.b. basis, totaled $54.2 billion in 1983;27 total freight costs might be conservatively estimated at 4.5 percent of the f.o.b. value of the merchandise, which is approximately equal to the freight on imports of the industrialized countries (5.0 percent for freight and insurance less, say, 10 percent of this total for insurance), or $2.4 billion.28 As a rough estimate we assume that one half of this amount may have been paid to maritime operators of other countries. Similarly, allowance must be made for shipment debits on imports into two other economies not included in the balance of payments data—Hong Kong and Bermuda. Hong Kong’s imports were reported at $24.1 billion, and exports of the “world” to Bermuda were $0.4 billion, according to Direction of Trade Statistics. Freight on imports to these two economies might have been about 8 percent (based on the c.i.f./f.o.b. tables in International Financial Statistics). If half came in foreign vessels, the implied debit would be 4 percent of $24.5 billion or $1.0 billion.

Among countries reporting balance of payment statistics, some reported no shipment debits; perhaps they reported their imports c.i.f. or included the freight debits in some other classification in the balance of payments. These were mostly relatively small countries in Africa and Asia; their total imports were $8.3 billion. Again, at 4 percent, freight on imports carried in foreign vessels would be $0.3 billion.

(4) Summary

Thus adjustments to arrive at world ocean shipment debits that would match reported or reportable credits would be as follows:

Table 65

ADJUSTMENTS TO SHIPMENT DEBITS, 1983

(In billions of U.S. dollars)

article image

Council for Mutual Economic Assistance.

b. Evaluation of Credits

In order to arrive at an estimate of reported “shipment” credits on maritime trade, similar adjustments also have to be made on that side, where the recorded total is $46.9 billion.

(1) Insurance

It seems likely, both on a priori grounds and on the basis of the questionnaire returns, that total shipment credits contain a smaller proportion of insurance premium receipts than the debits—based on our questionnaire responses, the difference was 2.2 percent. Allowance for this factor would thus amount to $1.0 billion.

(2) Other Means of Transport

Countries responding to the questionnaire reported total nonmaritime shipment credits of $6.6 billion. Again, this amount is taken as representing the world total, although it is certainly a minimum.

(3) Summary

Allowance for both of these factors would reduce reported credits to $39.3 billion (Table 66), with a remaining discrepancy on maritime freight of $32.5 billion.

Table 66

ADJUSTMENTS TO SHIPMENT CREDITS, 1983

(In billions of U.S. dollars)

article image

It may be noted that acceptance of this result implies that the revenues of all ships operated by reporting countries, including foreign-flag vessels chartered in, are, in fact, included in the reported totals. All but one of the countries responding to our questionnaire stated that to be the case, at least in principle. The validity of this implication of complete reporting needs to be tested by further research.

c. Evaluation of Adjusted Discrepancy in Shipment Data

If the adjusted excess of debits on marine freight of $32.5 billion is correct, it must represent the revenues earned by fleets not covered by the Fund’s data from carrying goods other than their own imports. The only independent means of testing the reasonableness of this result that seems feasible is to relate the revenues of the two groups—those included in the Fund’s data and those not so included—to the amount of tonnage operated by each group. However, there is no independent compilation of maritime shipping that conforms exactly to the country-of-operation basis required for balance of payments accounting. Data on world shipping by flag of registry and type of vessel, and also by various types of ownership, are maintained by Lloyd’s,29 but Lloyd’s statistics do not identify the current employment of ships under charter from one company to another, mainly because changes occur so frequently. However, it may be possible to obtain special tabulations from Lloyd’s that fit the balance of payments requirement fairly well. The principal difficulty is the prevalence of chartering, especially of vessels registered under so-called flags of convenience (primarily Liberia and Panama, but now including a growing list of countries or territories) but also of vessels registered in the principal maritime countries themselves. Moreover, there are intricate chains of ownership and many varieties of management and operating arrangements. Consequently, no simple criterion, such as flag of registry or ultimate ownership (if the latter could be reliably determined), would yield the correct allocation of shipping tonnage among countries.

One of the main tasks to be accomplished in solving the problem of the shipping account discrepancy will be to find a reliable means of allocating the operating world tonnage by country, so that the accuracy of the reported earnings can be verified and the missing credits can be assigned to the appropriate nonreporting countries or economies. Pending the availability of such data, it may be useful to illustrate the necessary estimating process by using data published by the United Nations Conference on Trade and Development (UNCTAD) that distribute the tonnage of five open-registry economies (Liberia, Panama, Cyprus, Bermuda, and The Bahamas) to the countries of “true management.”30

It seems unlikely that this definition coincides in all cases with country of operation as defined for balance of payments purposes. For instance, the UNCTAD data make no allowance for chartering among major maritime countries themselves, nor for ships registered under flags of convenience other than the five “open-registry” countries just named. Nevertheless, the use of UNCTAD data presumably brings one closer to the tonnage operated by individual countries than do the simple flag-of-registry statistics, and even closer if—as is done here—countries are combined in large groups, so that cross-chartering is covered, at least in part. Using Lloyd’s data adjusted by the UNCTAD allocations, revenue per registered ton can be calculated and a comparison made between “reported” and “missing” earnings per ton. However, to make such a comparison one must include not only the international revenues from exports and cross-trades but also the revenues earned in carrying imports and coastwise traffic of the countries operating the ships—revenues which are not included in the balance of payments.

Including the UNCTAD reallocations, the tonnage associated with countries not included in the Fund’s reported total of shipment credits in 1983 totaled 195 million deadweight tons (DWT), out of world tonnage of 648 million DWT. Although, as already indicated, the allocation by country in the UNCTAD data may not fit the balance of payments criteria, it seems clear that the largest missing fleets (including nationally registered vessels) are those operated by Greece, Hong Kong, and the U.S.S.R. Using the UNCTAD-adjusted information, inadequate as it may be, to illustrate the estimating process, we can proceed as follows.

To estimate the earnings of the “reported” and “missing” fleets from carrying the imports of their respective countries of operation (so as to eliminate them since they are not “international”), the following procedure was used:

article image

Corrected in 1986 IFS Yearbook.

Based on opinions of insurance industry specialists.

In addition to the caveat expressed in Subsection 2 a (2) of this chapter, this figure should, in principle, include freight on imports passing through third countries earned by carriers of the importing countries—for example. German (Fed Rep) trucks carrying imports from the United States through the Netherlands Such earnings are part of the freight component of the c.i.f value but, like freight on imports transported in domestic vessels, are not a balance of payments item.

Council for Mutual Economic Assistance.

Excluding trade among the CMEA countries.

(It was not possible, on the basis of available information, to estimate earnings from coastwise (domestic) traffic or, to accomplish the same end, to eliminate coastwise shipping from the tonnage figures.)

Thus we arrive at total and average earnings of the “reported” and “missing” fleets as shown in Table 67.

Table 67

DERIVATION OF FLEET EARNINGS PER REGISTERED TON

(In billions of U.S. dollars)

article image
Note: Excludes earnings of members of the Council for Mutual Economic Assistance (CMEA) on intra-CMEA trade.

The differences in the averages per DWT are, on their face, unlikely, and probably reflect primarily the inadequacy of the UNCTAD data as a basis for establishing the tonnage actually operated by each country, rather than large errors in estimating gross shipping earnings of the two groups.

It is quite possible, for instance, that the UNCTAD data reflect country of ultimate beneficial ownership rather than of actual operation, where these are different. For instance, an international oil company might have a tanker fleet operated by a Panamanian subsidiary. The operations of such a company would, in all probability, not be reflected in the Panamanian balance of payments statistics. In such a case, the UNCTAD data might classify the vessels by the country of the parent oil company (e.g., United States, United Kingdom), whereas, following the preceding analysis, it would be proper to consider the vessels as part of the “missing” fleet. To the extent such situations occur, the tonnage attributed to “reported” fleets is overstated, and earnings per registered ton understated, and there are opposite effects on “missing” fleets.

The foregoing analysis has focused on estimating the international maritime earnings of the “missing” fleets, which we have placed at S32.5 billion. In the process, we have, in effect, ignored the (estimated) discrepancy on insurance of minus $2.8 billion (debits, $3.8 billion; credits, $1.0 billion) and the discrepancy of $1.0 billion on shipment by nonmaritime means of transport (debits, $5.6 billion; credits, $6.6 billion). The combined discrepancy on these two accounts of minus $1.8 billion (which, to repeat, cannot be known but only roughly estimated from our questionnaire data) would remain.

But we have brought the maritime freight part of the “shipment” account into balance by assuming that the excess of debits, after adjustment, measures the international earnings of the “missing” fleets.

Table 68

ADJUSTMENTS TO MARITIME SHIPMENT ACCOUNT

(In billions of U.S. dollars)

article image

Council for Mutual Economic Assistance.

3. OTHER TRANSPORTATION

In the preceding section, we estimated the freight earned by maritime operators whose activities are not reflected as credits in the Fund’s balance of payments statistics. By the same token, the expenditures of these fleets in the ports of countries other than their own are not reflected as debits in the “other transportation” account. We deal with this subject in this section, together with other possible adjustments to this account. The data for this account as reported are as follows: credits, $70.2 billion; debits, $73.6 billion; and net discrepancy, -$3.4 billion (Table 64).

This account consists of three main components: passenger fares, charter hire, and port expenditures. Both ocean shipping and other means of transport are included. To arrive at port expenditures, the other two components must be deducted from the total.

Fortunately, passenger fares are reported separately to the Fund by most countries; some countries include them in the travel, rather than the transportation, account, and some include them, properly, in “other transportation” but do not report them separately. Reported fares were virtually balanced at the following levels: credits, $18.1 billion; debits, $18.0 billion.

Charter fees are not reported separately to the Fund but can be roughly estimated on the basis of information furnished on our questionnaires. The questionnaire respondents accounted for 74.7 percent of total shipment credits. Assuming that both charter payments and charter receipts were related more closely to total shipment earnings than to any other available number, charter payments and receipts as reported in the questionnaire were divided by 0.747 to arrive at estimated world totals of $10.0 billion and $3.3 billion, respectively. It can be assumed that the excess of debits of $6.7 billion is an approximate measure of charter payments to owners of flag-of-convenience ships. Thus, it would be appropriate to add $6.7 billion to “other transportation” receipts, since such countries have clearly not included them in their balance of payments reports.

Deducting passenger fares and (estimated) charter fees from the total leaves estimates for port expenditures by all means of transportation, which are shown in Table 69.

Table 69

DERIVATION OF PORT EXPENDITURES

(In billions of U.S. dollars)

article image

As indicated above, the “missing” fleets had total estimated freight revenues of $34.9 billion in 1983. Presumably the port-expenditure debits related to these earnings were also omitted from the Balance of Payments Statistics Yearbook data. On the other hand, at least as a first approximation, it is assumed that the countries receiving these port expenditures would generally have reported them as a credit. (However, port-expenditure credits may have been widely understated.)

The only reasonably reliable data available on the relation between total shipping revenues and expenditures in foreign ports are those derived from our special shipping questionnaire.31 The results (for sea transportation only) are summarized in Table 70.

Table 70

SHIPS’ EXPENDITURES IN FOREIGN PORTS

article image

On exports, imports, cross-trades, and passengers.

56 percent in 1984.

The major part of the business of the two economies whose residents are the major owners of our “missing” fleets—Hong Kong and Greece—presumably arises in third-country trade. As a rough estimate, we may place their port-expenditure figure at the high end of the range, say 60 percent, or $20.9 billion (based on total revenues of $34.9 billion), an amount which can appropriately be added to “other transportation” debits.

On the other hand, there seems to be evidence of gross underreporting of “other transportation” credits, presumably mainly port expenditures, especially by developing countries. Inquiries by the Working Party indicate that the proportion of total ships’ revenues used for port expenditures, including the cost of bunkers, averages well over 50 percent. Thus, for any country, one would expect port-expenditure receipts to be equal to at least half of the amount reported or reportable as shipping debits. On this assumption, we can increase port-expenditure receipts by $13.0 billion.32 (It may be noted that for many countries, the ratios were much more than 50 percent, and in some cases more than 100 percent.) In addition an allowance of $1.1 billion was made for port expenditure receipts of countries not included in the Fund data (CMEA countries, Hong Kong, etc.).

4. GEOGRAPHIC ALLOCATION

Adjustments to the “shipment” and “other transportation” accounts relating to the revenues and expenses of the “missing” fleets have been allocated by country according to the “true management” tonnage as adjusted by the UNCTAD data. The charter fee credits of $6.7 billion are not allocated to a particular country group, although it is likely that a significant part—assuming that our analysis is reasonable—should be attributed to Panama, a major offshore banking center, and Liberia. The underreporting of port expenditures of $14.1 billion was derived from detailed country data and thus could be accurately allocated. The remaining (small) adjustments to the “shipment” account (Table 68) were also attributable to specific areas. The results are summarized in Table 72.

Table 71

ADJUSTMENTS TO “OTHER TRANSPORTATION” ACCOUNT

(In billions of U.S. dollars)

article image
Table 72

GEOGRAPHIC DISTRIBUTION OF ADJUSTMENTS TO “SHIPMENT” AND “OTHER TRANSPORTATION” ACCOUNTS, 1983

(In billions of U.S. dollars)

article image

5. SOME IMPORTANT RESERVATIONS

The estimates presented in the three preceding sections of this chapter do not pretend to be definitive, but only to illustrate procedures that could be followed to identify and measure the erroneous or missing elements in existing data. The following caveats need to be kept in mind.

a. Reliability of c.i.f./f.o.b. Data

While the c.i.f./f.o.b. ratios, as published in the Fund’s International Financial Statistics Yearbook, were not used directly in calculating any of the adjustments to the balance of payments entries, they were used above to estimate freight revenues on imports in national vessels. Nevertheless, it may be that the actual balance of payments debits on “shipment” account and the c.i.f./f.o.b. data are both overstated—especially for some of the developing countries.

To illustrate the problem, the calculations presented earlier were reworked on the assumption that the ratio of insurance plus freight costs to the f.o.b. value of merchandise imports for developing countries (other than those in the Middle East)33 declined from 1973 to 1983 by 28.4 percent, the same as the decline in the ratio for industrial countries. Under this assumption, total insurance and freight for those developing countries, and for the world, would have been $6.3 billion less. An unknown part of the $6.3 billion would have been applicable to freight on imports transported in national carriers. Nevertheless, if the entire $6.3 billion were deducted from adjusted shipment debits,34 Table 67 would be changed to read as follows:

article image
Note: Excludes earnings of members of the Council for Mutual Economic Assistance (CMEA) on intra-CMEA trade.

Thus even with this (probably excessive) adjustment, the average earnings per million DWT of capacity are still not compatible with what would be expected on an a priori basis—namely, that the “missing” fleets, which are comprised, to a larger-than-average extent, of bulk cargo and tanker vessels, should have smaller-than-average earnings.

b. Reliability of UNCTAD Data as Measure of Nationalities of Ships’ Operators

Some reasons for lack of confidence in these data were given in the preceding sections of this chapter. It may also be noted that since the status of any individual vessel may change from voyage to voyage, it is likely to be very difficult to obtain the relevant data for, say, an entire year.

c. Lack of Adequate Data on Capacity Utilization

Even if all vessels could be correctly assigned to the proper country of operation, the amount of tonnage actually employed in international trade at any given time would also be difficult to estimate. In recent years, large amounts of tonnage have been laid up; data on laid-up tonnage are available by flag of registry, but so far as the Working Party is aware, not by nationality of operator. If the laid-up tonnage were proportionately greater for the “reporting” than for the “missing” fleets, the average earnings per ton employed would be greater for the former and less for the latter than is shown above. Also, some countries, notably the United States, employ a substantial amount of nationally registered shipping in coastwise trade.

6. SUMMARY AND RECOMMENDATIONS

In this chapter we have attempted to identify and to establish some magnitudes for what appear to be the major missing elements in the “shipment” and “other transportation” accounts. In part this was accomplished by sheer assumption; as noted, there is no reliable external evidence to support our estimate of the earnings of fleets omitted from the Fund’s data. However, we believe that the existing asymmetries are reasonably well explained by our analysis.

Nevertheless, the Working Party has not been able to explain the transportation discrepancies with the degree of confidence that can be attributed to its resolution of the investment income discrepancies. Relatively more research by the Fund staff remains to be accomplished before statistical procedures for reducing the discrepancies can be put in place. Our recommendations, therefore, are necessarily somewhat less specific than those applicable to investment income.

(1) Careful consideration should be given to the possibility of enlisting the aid of the Greek and Hong Kong authorities in developing estimates for earnings and port expenditures of their respective fleets. The Hong Kong authorities collect considerable data from resident shipping companies. Even if those authorities do not receive the exact information needed for balance of payments entries, they should be in a better position than, say, the Fund staff, to produce acceptable estimates. The Greek situation may be more difficult, since many “Greek” shipowners apparently do not reside in Greece, but there are indications that some useful data exist. Another possibility would be to regard vessels operated by nonresident Greeks as part of the national shipping of the actual countries of their owners’ residence. However, it is not likely that such a procedure would be acceptable to the authorities of those countries.

(2) As noted earlier, further refinement of the shipping discrepancy analysis will require the accumulation of much better data on the shipping tonnage actually operated by each country. The Working Party recommends that the Fund pursue such research, using outside experts as necessary.

(3) A proper evaluation of the missing shipment credits is at least partly dependent on the ability to arrive at an estimate for the total freight and insurance bill of the world. This, in turn, can at present only be estimated by the use of the c.i.f./f.o.b. factors compiled by the Fund and published in its International Financial Statistics Yearbook. We recommend that these data be carefully monitored, and especially that their consistency with balance of payments data be established and maintained.

(4) We also recommend that responsibility for monitoring the transportation data be centralized in some individual in the Balance of Payments Division of the Fund’s Bureau of Statistics, who would also be responsible for making estimates of “missing” data for inclusion in the World Economic Outlook. The field is complex, and it would not be worthwhile to ask all the country specialists in the Balance of Payments Division to become experts on shipping and transportation statistics, even on those supplied by the countries they concentrate on.

(5) As indicated above, it appears that the credit entries for port expenditures are understated by a significant number of countries, perhaps because of outright omissions of certain types of receipts (such as bunker fuel deliveries)35 or some netting in the estimating process. It is recommended that this possibility be investigated by the Fund, perhaps with a clearly focused one-time survey. Such an investigation could also serve as a stimulus to encourage more accurate reporting of the transportation accounts. The Working Party survey could help in designing an inquiry. All countries reporting “other transportation” credits that were significantly less than 50 percent of “shipment” debits would be candidates for inclusion in such a survey.

(6) Further progress on resolving the shipping discrepancy will require information on the industry that is likely to be available mainly from industry sources. We recommend that consultations with experts be held to establish more precise specifications for the information required and to explore the feasibility of acquiring such data from country or industry sources.

Within Same Series

Other IMF Content

Other Publishers Content

  • Collapse
  • Expand
Cover Report on the World Current Account Discrepancy
Report on the World Current Account Discrepancy
ISBN:
9780939934904
Publication Date:
15 Sep 1987
Publisher:
International Monetary Fund
DOI:
https://doi.org/10.5089/9780939934904.071
Pages:
203