Financial Performance of Government-Owned Corporations in Less Developed Countries

An earlier version of this paper was presented to a conference on the topic “Is the Business Cycle Obsolete?” organized by the Social Science Research Council and held in London, England, on April 3–7, 1967.


An earlier version of this paper was presented to a conference on the topic “Is the Business Cycle Obsolete?” organized by the Social Science Research Council and held in London, England, on April 3–7, 1967.

The purpose of this paper is to ascertain empirically the financial experience of government-owned corporations, particularly in the less developed world.1 The organizations studied are engaged in producing goods and services in the utility, transportation, communications, petroleum, and other industrial fields. In order to provide a representative sample for all areas of the world, as well as for these industry groupings, a large body of data has been assembled. These data cover 64 government-owned corporations in 26 countries (see Appendix I) for an average of 7 years each; in most instances, the profit and loss statements and balance sheets from the entities served as the source. These raw data have been rearranged into comparable categories (flow of funds, net income, investment, transfers, etc.) for each of the 461 corporate years in the sample.

This information has been used to describe four major facets of the financial performance of the government-owned corporations: (1) their profitability or internally generated supply of funds for investment or other purposes (hereinafter referred to as “flow of funds”); (2) their patterns of investment; (3) the financial burdens or benefits that a government can expect from operation of a government-owned corporation; (4) some of the interrelationships between profitability, investment, and the financing of investment.

I. Data and Methods

Sampling procedure

The data used here are the end result of a sampling process that had several goals, viz., (1) to collect a sample that was representative of the universe of government-owned corporations; (2) to obtain enough observations (each year of experience is considered as an observation) to differentiate as statistically significant a 10 per cent difference in results between groups of observations; and (3) to be able to test the hypothesis that areas and industry groups were similar in performance characteristics.

A universe consisting of balance sheets and profit and loss statements for all government-owned corporations in all less developed countries was not available. Rather, the many published annual reports that were obtainable in Washington, D.C., were drawn on and supplemented by additional information collected in the country concerned. When these published and unpublished reports became available, the data were evaluated and classified (see Table 1). In this process, differences in accounting procedures between corporations and between corporate years of experience were largely eliminated.

Table 1.

Classification of Corporate Data, Example Based on Data for Nigerian Ports Authority

article image
Source: Nigerian Ports Authority, Annual Report, 1961/62 (Lagos, Nigeria). See Appendix II for further details.

The data were grouped by area and type of industry. To differentiate accurately between the results of these groups required obtaining at least 20 observations in each group. Ideally, each base unit, or observation, would have consisted of a corporation with information for enough years (4–5) pooled into one observation to minimize the influence of any particular year’s peculiarities. With 20 observations in each of the 24 groups, this would have implied 480 corporations and data for 2,000–2,500 corporate years.

Such data were not available. Therefore, each corporate year of experience was used as a base unit, or observation, in the statistical analysis. This practice assumed that results for corporate years are independent one from another, both between and within the corporations. An attempt was made to have at least 3 corporations in each group, and more than 5 years’ experience for each corporation.

Even with these compromises, there were occasionally insufficient data to have 20 observations for each group. Moreover, not all groups could be represented because either the required information was not available or the type of corporation for the area concerned did not exist. For example, no time series were available for government-owned petroleum corporations in Africa.2

Classification of data

Table 1 lists the categories into which the raw data were grouped for each corporate year of experience and gives a numerical example. Although these categories are fully described in Appendix II, several facets of the classification are worth noting here. First, depreciation is not included in current expenditures. Not only do accounting practices differ a great deal from corporation to corporation but also book valuations of depreciable assets usually lag behind inflation. Both elements would have introduced gross distortions into the comparisons presented here; hence, only the other current account items were used, the difference between receipts and expenditures so derived being called the flow of funds. Then standard depreciation allowances (based on information gathered independently) were subtracted from the flow of funds to derive what is commonly called net income.

Within revenues and expenditures, the profit or loss on the sale of assets (usually a small item) was included. As investment goods in most cases are not sold until they are virtually useless, the proceeds from this type of sale should not be considered as negative investment but rather as the sale of goods or services adding to the flow of funds to be used for new investment or for other purposes. Ideally, the gross sales price rather than the profit or loss on the sale should be used. (The latter subtracts the book value from the sales price.) Usually, however, assets are fully depreciated when sold, so that the gross sales price and the profit from the sale will be similar.

Most important in the treatment of the raw data was the decision not to estimate unspecified items (except depreciation allowances). For instance, in comparing the financial performance of one corporation with another a valid approach would have been to impute interest for the entire stock of assets as a cost—including an appropriate amount of interest on the government’s equity—even if full interest payments were not being made. Or, to take another example, if a corporation gets cheap imports because of selectively lower tariffs, or by means of a multiple exchange rate, the advantage could be considered a subsidy. Exemptions from internal taxes might also be classified as a subsidy. No allowance was made for such items; only actual flows plus standard depreciation allowances were considered. The main reason for this decision was expediency; balance sheets and profit and loss statements usually provide insufficient guidelines for estimates, and the need for a large sample precluded a more extended search for data.

In most cases, accounting differences between corporations for such items as provisions for reserve funds were successfully eliminated. However, it was not always possible to reconcile exactly operating revenues and expenditures with financing items because of the intricacy of the accounts or gaps in the data. To some degree, these inadequacies are measured by the size of the residual (see Table 1).

The raw data from corporate balance sheets and profit and loss statements were not presented. Rather, all the data were calculated as ratios whose base is the “activity” of the corporation—the average of revenues and expenditures for the particular corporate year (see Table 1). These ratios were used to facilitate comparison among corporations and to avoid problems involved in converting the figures to a common unit of value by applying current exchange rates and appropriate price indices. The disadvantage incurred was that all corporate years were treated as equally important regardless of the size of the corporation.

Of other variables usable as a base, the most likely were net fixed assets, revenues, or expenditures. The most accepted of these measures is net fixed assets. For example, profits are usually calculated as a rate of return on net fixed assets. In countries with high rates of inflation, however, revaluation of net fixed assets often lags behind inflation substantially. In addition, depreciation practices vary between countries and corporations. Hence, net fixed assets were not useful as a base in the present study.

Revenues and expenditures considered separately also had grave disadvantages. Pricing policies vary tremendously between corporations, from monopoly pricing to almost complete subsidization of sales. Revenues alone, therefore, are not a good base. The same types of difficulties pertain to expenditures, as some corporations are more efficient economically than others. By averaging revenues and expenditures, however, many of these difficulties can be avoided. The activity of an extremely efficient corporation following a monopolistic pricing policy will give about the same base as that of a very inefficient corporation that subsidizes consumption, or one of normal efficiency that barely covers costs with revenues. Inflation quickly increases either revenues or expenditures, and thus the activity base is fairly responsive to changes in the external economic environment. For these reasons, the average of revenues and expenditures was chosen as a measure of economic “activity,” and was used as a base for presentation of the data for each corporation.

Thus, as an indicator of profitability, for example, the ratio of flow of funds to fixed capital was not used; rather, the denominator was replaced with activity as defined above. Although the two measures are identical at zero profit, and show either positive or negative signs simultaneously, their size and movements differ, as can be seen in Table 2. If the corporation continuously adjusted its capital stock to fit production, and if ideal capital/output ratios remained constant, the ratio of flow of funds to activity would equal the ratio of flow of funds to capital stock multiplied by a constant factor. If one assumes that capital stock is greater than activity, the ratio of flow of funds to activity would magnify the movements of the more usual ratio of flow of funds to capital stock. Since, in practice, capital stock does not adjust smoothly with output, movements of the ratio of the flow of funds to capital stock will only accidentally follow the same patterns as the ratios using activity as a base, though they will always show the same sign.

Table 2.

Comparison of capital Stock and Activity as a Base for Measuring Performance

article image

The main difference between flow of funds and the usual definition of profit is that the flow of funds does not include depreciation as an expenditure, nor subsidies as revenues.

The base “activity” is a valuable analytical statistic, as well as providing a way around the problem of making comparisons among corporations in different countries, and it aids generalization of the results obtained. It is useful for governments to know how much subsidization, how much investment, etc., a given size of corporation may require. “Activity” is a good measure of size and thus facilitates presentation of answers to these kinds of queries. The results of this study may show, for example, that a certain kind of industry in one of the four areas averaged an investment of 50 per cent of activity. Thus, a government might expect that funds equal to one half the average of revenues and expenditures would be needed for investment financing.

As noted earlier, the corporations are classified by area and by industry. For the various ratios considered (flow-of-funds ratio, long-term debt ratio, etc.), there are enough data to determine statistically when the ratio for any area or industry is significantly different from the overall mean of all observations together. The method used for the separation of high-ratio and low-ratio areas or industries is a variance analysis using a dummy variable linear regression approach.3

II. Results of Empirical Investigation

Flow-of-funds ratios

Table 3 presents the flow-of-funds ratios derived from the sample.4 Not including depreciation allowances, the average ratio for all the industries and years covered was about 8 per cent, that is, the average difference between current revenues and current expenditures was 8 per cent of the average of revenues and expenditures of the corporations. This implies that the corporations generated enough profits to pay for some replacement capital.

Table 3.

Flow-of-Funds Ratios by Area and Industry1

(Per cent of activity)

article image

For list of corporations used in compiling this table, see Appendix I. See also footnote 4 in the text.

A wide divergence existed in the sample between areas and between different types of industries. For the areas, there are broadly two groups of corporations: the European and Latin American corporations in one group, and the African and Asian corporations in the other. The European and Latin American corporations barely covered their current costs, not including depreciation, whereas the African corporations had a flow-of-funds ratio of 20 per cent. Asia can also be considered a high-profit area, as the average return on activity was 16 per cent. These results were statistically significant, at the 90 per cent level or above.

Variation in flow-of-funds ratios between industries in the sample was greater than between areas. The railway industry had the lowest flow-of-funds ratio, not covering current costs on the average. In Latin America, the railways group required an average of 57 cents in external resources for each dollar of activity, or 75 cents for each dollar of revenue. European railways also incurred a rather large loss. The communications group, which includes such industries as telephone, telegraph, and radio networks, as well as several post offices, was barely profitable. The figure for Africa should be discounted, as there is only one communications corporation in the sample. Only the petroleum and electrical enterprises in the sample had positive flow-of-funds ratios in all the areas. The petroleum industry group was most profitable, earning a return of almost 34 per cent on activity. The results for the petroleum entities may be underestimated, as many countries have indirect sales or excise taxes that often substitute for the price that the corporations could otherwise charge. This is also true, but to a lesser extent, for the electrical enterprises. In this context, it is notable that Africa and Asia have the highest flow-of-funds ratios, even though there are no nationalized petroleum corporations in the sample. One of the reasons is that the railways in the countries included in the sample are more profitable than average.

These results were confirmed statistically. The flow-of-funds ratios for railways, communications, and other industries are significantly lower than the over-all mean, while those for the electrical and petroleum enterprises are higher than the mean. Other transport, which includes mostly port authorities, is not significantly different from the mean.

The figures do not explain why a particular classification has a low or high flow-of-funds ratio. The impression is that the African railways are strongly oriented toward freight haulage, and tend to have monopolies in this category. Asian railways have a large number of passengers, but they also have monopolies.5 European and Latin American railways haul freight and passengers but compete with road transport. It seems surprising that the European railways do not make relatively higher profits than those in Latin America, as many of the latter are faced with inflation that continually outruns prices, as well as with notable economic inefficiency in some cases. These factors also seem to affect the electrical enterprises in Latin America. The European petroleum complexes combine retailing and distributing networks with basic refineries to a larger extent than do those in Latin America; this may be one reason for their smaller profits. In addition, in the petroleum complexes there is probably less opportunity for inefficient increases in costs of production through additions to the work force, because the normal cost of labor in a refinery is only 6–7 per cent of the total production expense. This may provide a clue to the differences between industries in Latin America.

Factors peculiar to specific countries cannot be brought out by the analysis. Variations in the flow-of-funds ratios owing to tariff protection and to availability of foreign exchange are examples. The samples for individual countries are usually not large enough to estimate such factors. For areas or industries, however, the sample is representative, and the results presented are judged to be reasonable estimates of the experience that might be expected.

The flow-of-funds ratios discussed above do not include an allowance for depreciation. Depreciation will vary for different industries as well as for each corporation. Table 4 gives estimates of depreciation allowances as a per cent of activity by industry, based on conversations with experts in each industry. The figures in Table 5, showing the net income ratios, were derived by deducting from the figures in Table 3 the estimated allowances in Table 4. None of the areas generated sufficient funds for replacement capital. On a more detailed level, the electrical enterprises in Europe, Africa, and Asia, the petroleum complexes in Latin America, the other transport and other industries grouping in Africa, and the communications entities in Europe and Latin America had positive net income ratios. Over one third of the groupings studied were able to cover operating costs plus estimated depreciation.

Table 4.

Normal Depreciation Allowances and Capital Requirements per Unit of Activity, by Industry1

article image

The figures are derived by taking the ratio of net fixed assets to activity in a sample of “representative” companies. The requirements for being considered “representative” included an up-to-date revaluation of installed capital and operation at reasonably full capacity. Depreciation allowances were computed on the assumption that the average of the ratio of net fixed assets to activity was 50 per cent of gross fixed assets, and by using average length-of-life figures for capital obtained from experts for each type of corporation.

Ratio of net fixed assets to activity.

Average of all other industries.

Table 5.

Net Income Ratios by Area and Industry1

(Per cent of activity)

article image

For list of corporations used in compiling this table, see Appendix I. See also footnote 4 in the text.

Investment ratios

Table 6 shows that the average investment of the corporations in the sample was more than 74 per cent of activity. Latin America with a rate of less than 60 per cent was the lowest, and Asia was the highest of the four areas. The diversity in investment behavior between industries was more substantial, ranging from one in four dollars of activity to one and one third dollar’s worth of investment for each dollar’s worth of activity. These results are affirmed by the statistical analysis, which separates the railways, other transport, petroleum, and communications groups as below the mean, with the electricity and other industries groups above the mean.

Table 6.

Investment Ratios by Area and Industry1

(Per cent of activity)

article image

For list of corporations used in compiling this table, see Appendix I. See also footnote 4 in the text.

Thus, investment costs to be financed by the enterprise either through retained profits or from external sources are large. Table 5 demonstrates that only one third of these groups was able to provide sufficient funds from retained earnings for replacement capital purposes—the need being about 24 per cent of activity. Other forms of financing have obviously been necessary for the large investment ratios given in Table 6. These will be described in more detail below.

If we assume that the average flow-of-funds ratios in Table 3 reflect to some degree marginal operating ratios, then we would expect those corporate groups with high ratios to invest more, ceteris paribus.6Table 7 allows some judgments to be made on this subject, as it uses the data of Table 4 to establish from the investment ratios of Table 6 crude patterns of growth in activity. European corporations, with the lowest flow-of-funds ratios, seem to grow the fastest. This may be the result of queuing. Services that are priced too low will be in much greater demand than if they were priced at nonsubsidized rates, and this demand may take the form of public clamor for more investment. On the other hand, the figures show a generally positive association when the type of industry is considered. Investment in railways resulted in the least growth, and this group was the most unprofitable. The petroleum and electrical enterprises had rapid growth as well as high flow-of-funds ratios. The communications entities followed the patterns of the railways enterprises.

Table 7.

Rate of Growth in Activity Implied by Ratios of Investment and Capital to Activity, by Area and Industry12

(Per cent per annum)

article image

For list of corporations used in compiling this table, see Appendix I. See also footnote 4 in the text.

The method of constructing this table may be illustrated by taking Latin American petroleum industries as an example. Table 6 states that investment was 70 per cent of activity. Table 4 indicates that an increase of 2 units of capital is necessary for a 1 per cent change in activity, and that depreciation is 20 per cent of activity. Thus, 70 per cent minus 20 per cent equals 50 per cent, and 50 per cent divided by 2 equals 25 per cent.

By area, there seems to be an inverse relationship between profits and investment—the higher the profits, the lower the investment. This is probably due to the behavior of the other industries groups, as well as to an inverse relationship within some of the disaggregated groups. Only a further statistical analysis can separate these influences into an orderly pattern.

Notwithstanding the behavioral relationships, the derived over-all rate of growth (as measured by the ratio of capital to activity) of about 15 per cent is impressive and much above the total growth rates in gross national product for any of these areas. The area with the lowest rate of growth is Latin America, with an increase in activity of 12 per cent. An exception to these trends is railways, which have had virtually no expansion.7

Surplus-after-investment ratios

If loan repayment is not included, the total cost (per unit of activity) of owning a corporation for which the government will need to find financing is the flow of funds plus investment costs (regarded as a negative item in this paper). That is, the financing needs of a public corporation are evident only when we consider investment as an expenditure to be financed, in addition to other expenditures. This cost is measured by the “surplus-after-investment” ratios shown in Table 8. When investment is considered as a cost, two thirds of one dollar for each dollar of activity needs to be provided from external sources to finance deficits and/or investment. On the basis of the experience covered, therefore, the developing country cannot expect that, on the average, it will gain financial resources internally by nationalization or creation of government-owned corporations. In the sample, external financing was required in all areas and for all types of industries, except for communications entities in Europe and other transport in Africa. Of course, within the sample, there were particular corporations or particular years in which surpluses were generated, even after investments were covered. These were exceptions, however, and not the general rule.

Table 8.

Surplus-After-Investment Ratios by Area and Industry1

(Per cent of activity)

article image

For list of corporations used in compiling this table, see Appendix I. See also footnote 4 in the text.

By area, the surplus-after-investment ratios were quite different from the ratios (flow of funds) excluding investment. Latin American corporations, which together with the European corporations were in the most unprofitable group when investment was excluded, put least total burden on the sponsoring governments when investment was included as a cost. This would suggest that poor performance on current account is connected with low investment activity. On the other hand, the European corporations, whose current account performance was similar to that of Latin America, required the most outside financing per unit’s worth of activity; in fact, for every unit of activity, almost one unit of outside financing was necessary.8 Africa, which was the most profitable area when investment was not included, fell to second place after investment was included as an expenditure; Asian corporations were almost as burdensome as European. However, the analysis of variance demonstrated that none of the area results was significantly different from the mean for the entire sample.

Variation between industries was again larger than that between areas. The petroleum complexes, which were most profitable before investment was considered, also ranked high among the industries after financing investment expenditure—less than one dollar of outside resources needed to be provided for every four dollars of activity. The analysis of variance demonstrated that the performance of the petroleum corporations was significantly better than the group mean. The other transport entities also did significantly better than the mean for the sample as a whole. Surprisingly enough, the railways’ total surplus-after-investment ratio was greater than the sample average also—almost significantly greater. (As noted earlier, this was due to their low investment activity.) This is in sharp contrast to electrical enterprises, which required more than one dollar of outside resources for each dollar of activity including investment.

Central government transfers ratios

The government-owned corporations have been able to gain resources from foreign governments, from the private sector, from revenue generated internally, and in some cases from their own governments in the form of bona fide loans. In addition to these sources of funds, corporations in all areas, and all industries except one, have required central government transfers (see Table 9).9 The mean level of the transfers to the corporations in the sample was almost 33 per cent of activity: that is, for the 461 corporate years of activity considered, the governments concerned supplied one dollar for every three dollars of activity, on the average. Central government transfers were highest in Latin America, where over two dollars have been supplied for every five dollars of corporation activity. Africa was lowest in subsidy rate—less than one dollar for every seven dollars of activity. European and Asian corporations absorbed approximately one dollar subsidy for every three dollars of activity. As one would expect from these findings, the analysis of variance showed Latin America to be significantly higher, and Africa significantly lower, than the over-all mean. The transfer patterns varied more by industries than by area. The figure for the petroleum group was significantly lower than the mean, and that for the other industries group was twice as high as the mean (also significant). The other industries group used the most transfers—over one dollar for every two dollars of activity. Railways and electrical enterprises were given the next largest amounts, with petroleum, communications, and other transport complexes putting the smallest burden on their governments.

Table 9.

Transfers Ratios by Area and Industry1

(Per cent of activity)

article image

For list of corporations used in compiling this table, see Appendix I. See also footnote 4 in the text.

With the exception of the African other industries group, which consisted of only one corporation, the Latin American railways absorbed more transfers than any other group. Almost one dollar was transferred for each dollar’s worth of activity for these corporations. In contrast, the petroleum complexes in this area needed no transfers. It should be remembered that these figures do not allow for taxes on the sale of petroleum products within the country, or for taxes on imports. If these items were included, the petroleum entities would undoubtedly be shown to have provided a large supply of funds to the government. Next highest on the transfer list is the other industries group in Europe, followed by the electrical enterprises in Latin America; the reasons for these transfers will be considered below.

In Africa, the railways received almost no transfers, and other transport facilities actually provided funds to the government. The same general trends are evident in Asia for these industries. Therefore, it seems possible for governments to own corporations which are self-financing in certain situations. In the sample, this seems to be so for petroleum complexes in both Europe and Latin America, and with transportation facilities in Africa and Asia. Electricity enterprises in Europe received few transfers but needed a large supply of loans.

Behavioral factors

The questions to be examined in this section are as follows: (1) Is investment of the corporation affected by profitability? (2) Are transfers more closely connected with profitability or with investment behavior? (3) Is the amount of investment financed by long-term debt related to profitability?

It would be reasonable to expect a positive answer to the first question if government-owned corporations were to resemble private concerns. A private firm wishing to maximize profits will tend to continue to add to its capital stock up to the point where it expects that further outlays will no longer cover the costs of borrowing, including the opportunity costs of using its own funds. Therefore, present profits often serve as a signal for more investment and provide some of the funds for this investment. If profits continue to grow, the corporation will invest more, unless other factors predominate.

These other factors may be particularly important in the government-owned corporation. For instance, if pricing policy results in queuing (for example, a long waiting list for telephones), and if company profits are low or nonexistent, a government-owned corporation may make rapid investments despite the lack of profitability, given an external source of funds. Or, a government-owned corporation that persists in incurring operating deficits may have large investments as a matter of social policy. For example, new rail lines may be built to open up the interior despite a deficit owing to economic inefficiency, lack of demand, or pricing policy. It would seem, therefore, that the investment and profitability of government-owned corporations would vary more than those of private corporations. There are many other influences which may affect this relationship, viz., type of industry, the location of the corporation, the supply of foreign exchange, the internal capital market.

Chart 1 suggests that there is a positive relationship between profitability and investment. (See Tables 3 and 6.) In disaggregation by industry, a generally positive trend is found, although the other industries group and the electricity group are discretely above the other groups. The railways, the most unprofitable, invest the least. Communications and other transport facilities have about the same flow-of-funds ratios and invest about the same, but more than the railways. The capital/output ratio of the railways is higher than that for the communications and other transport facilities (with the exception of shipping), and this amplifies the trend described. There also seems to be a positive relationship between area profitability and investment.

Chart 1.
Chart 1.

Relationship Between Flow-of-Funds and Investment Ratios, Classified by Area and Industry

Citation: IMF Staff Papers 1968, 001; 10.5089/9781451956221.024.A003

On the other hand, when the ratios are cross-classified by area and industry, there is no relationship. (See Chart 2, and also Tables 3 and 6.) From this cursory examination, it appears that the analysis of investment and profitability depends more on area and industry than on profitability within an area or an industry.

Chart 2.
Chart 2.

Relationship Between Flow-of-Funds and Investment Ratios, Cross-Classified by Area and Industry

Citation: IMF Staff Papers 1968, 001; 10.5089/9781451956221.024.A003

A crude equation relating investment to profitability and lagged profitability, as well as to area and industry, reinforced this conclusion. In this equation, after separating out the influences of area and industry, investment was found to be inversely correlated with current profitability, and positively but insignificantly related to lagged profitability.10 It appears at first, therefore, that investment does not depend on average profitability in these corporations.

What determines the transfers that a corporation receives from the central government? Transfers are required because a corporation either incurs heavy deficits or needs to finance investment. Charts 3 and 4 show the results for these two relationships. As one would expect, there seems to be a negative relationship between flow-of-funds ratios and transfers and a positive relationship between investment and transfers. Chart 3 indicates that transfers are never limited to the flow-of-funds ratio. (See Tables 3 and 9.) All points lie above the diagonal line, which delineates equality of transfers and flow-of-funds ratios. Corporations with positive as well as those with negative flow-of-funds ratios have required transfers. On the level of detail shown in Chart 3, one can say that decreasing flow-of-funds ratios have probably been associated with increasing levels of transfers.

Chart 3.
Chart 3.

Relationship Between Flow-of-Funds and Transfers Ratios, Cross-Classified by Area and Industry

Citation: IMF Staff Papers 1968, 001; 10.5089/9781451956221.024.A003

Chart 4.
Chart 4.

Relationship Between Flow-of-Funds and Transfers Ratios, Cross-Classified by Area and Industry

Citation: IMF Staff Papers 1968, 001; 10.5089/9781451956221.024.A003

Moreover, most corporations get transfers smaller than their investment requirements. As shown in Table 3, corporations on average achieved a surplus of about 8 per cent excluding depreciation allowances, thus generating some funds for investment, which were supplemented by transfers. Chart 4 demonstrates these findings. (See Tables 6 and 9.) The diagonal line indicates equality between investment and transfer ratios. With the exception of European and Latin American railways and African and Asian communications entities, all the corporations received smaller transfers than their requirements for investment. Thus, it would appear that the government-owned corporations have been able to obtain funds for investment from loans as well as transfers. We were successful in explaining the average transfer level using the following equation:


R2 = 0.674


Tt= transfer ratio in time t

E= dummy variable for Europe

LA= dummy variable for Latin America

Af= dummy variable for Africa

As= dummy variable for Asia11

RR= dummy variable for railways

OT= dummy variable for other transport

Pc = dummy variable for petroleum

El= dummy variable for electrical enterprises

C = dummy variable for communications

OI= dummy variable for other industries group11

Pt = flow-of-funds ratio

It = investment ratio

(Pt×It)= interaction effect for flow-of-funds ratio and investment ratio

σPt= variance of flow-of-funds ratio for a particular corporation.

This equation demonstrates that European transfers were significantly lower than the mean and Latin American transfers were significantly higher than the mean. Transfers to the railways were lower than the mean and those to the electrical enterprises were higher.

Changes in the flow-of-funds ratio, and in the investment ratio, and the interaction effects between the two ratios were intimately connected with transfers from the central government. A decrease of 1 per cent in the flow-of-funds ratio caused an equivalent increase in transfers. A change of 1 per cent in investment was associated with a change of only 0.2 per cent in transfers. In other words, changes in transfers and in flow-of-funds ratios are of unitary elasticity, but changes in transfers are inelastic with respect to investment. The interaction effect was also significant.12 Finally, although the coefficient of the variance of the flow-of-funds ratio is not significant, it appears that there is some positive association between fluctuations in the level of profitability and the level of transfers. This may be a result of reaction times between increases in the flow-of-funds ratios and a decline in the level of transfers owing to administrative lags, accounting methods, etc.

It appears that policies leading to more efficient corporate operations will have a dramatic effect on the level of transfers to these entities, but that differences caused by pecularities of an area or an industry will not be so easily eliminated. However, most of the corporations even though profitable required central government transfers, but at the same time were able to finance a significant portion of their investment from other sources. A high flow-of-funds ratio, therefore, does not ensure the government against an outflow of funds to the corporation, but it should substantially reduce transfers.

The final behavioral relationship to be described is between profit-ability and the proportion of investment financed by long-term debt. The outcome of this relationship has important implications for government finance. If corporations with higher profitability are able to obtain more external and internal (nongovernment) long-term borrowing with higher profitability, then governments can use corporate pricing policy or increases in economic efficiency as alternatives to increases in taxes, increased borrowing, or decreases in other government expenditures. In addition, the persistence of this relationship implies that the cost of the resource use of the corporation is borne by consumers of its products and by voluntary lenders, rather than by the general taxpayer, who may derive no benefit from the goods and services produced.

The relationship between the proportion of investment financed through long-term debt and the level of the flow-of-funds ratio is shown in Chart 5, which seems to demonstrate that, as the flow-of-funds ratio increases, the corporations’ ability to get long-term debt also increases. This long-term debt includes external debt, internal nongovernment debt, and bona fide loans from the parent government.

Chart 5.
Chart 5.

Relationship Between Investment Financed Through Long-Term Debt and Flow-of-Funds Ratios, Cross-Classified by Area and INDUSTRY

Citation: IMF Staff Papers 1968, 001; 10.5089/9781451956221.024.A003

There seems to be a difference between the behavior of different areas. European corporations seem to finance investment by borrowing without regard to the level of the flow-of-funds ratio. Latin American corporations seem to finance less of their investment by borrowing as the flow-of-funds ratio increases. For instance, Latin American railways with their extremely high deficits borrow at long term more than enough to finance investments. Latin American petroleum companies, however, with very high rates of profit finance only a small portion of their investment with long-term borrowing. The African and Asian corporations seem to determine the over-all trend most strongly. This is partly because the communications corporations in India are government enterprises, rather than government-owned corporations, and finance their investments by either internal cash generation or transfers from the central government.

The statistical analysis shows, however, that the graphical analysis is spurious, as the relationship cannot be fully specified. The slope is apparently derived from different levels of area and industry effects, rather than from the level of the flow-of-funds ratio. To test this, the following equation was run:


R2 = 0.004

where LTD/I= long-term debt/investment.

The flow-of-funds ratio was not significant at the 90 per cent level, and the variation in the level of the proportion of investment financed by long-term debt was not explained by differences in the level of the flow-of-funds ratio.

In summary, although profitability and the proportion of investment financed externally are linked, these links are not strong, and other variables not specified here have important effects which have not been isolated. However, the over-all level of this relationship is influenced by area and industry effects.

III. Summary and Conclusion

Government-owned corporations, rather than serving as a focal point for collecting financial resources for their own investment or for other purposes, have generally placed a financial burden on parent governments. Although the average flow of funds 13 was 8 per cent of annual activity,14 the need for replacement investment alone is estimated as averaging 24 per cent of activity. Thus, the 8 per cent flow-of-funds ratio furnished insufficient financing for purchases of assets to maintain real production activity, to say nothing of developmental growth.

The corporations in the sample actually invested much more than this 24 per cent depreciation allowance. The average investment ratio (investment to activity) of 74 per cent resulted in an estimated annual growth of about 15 per cent in activity. On the other hand, this investment increased the financing needs to 66 per cent of activity. Therefore, about two thirds of one dollar of resources has had to be provided from sources external to the corporations for each dollar of corporate activity. This is quite contrary to the experience provided by studies of private corporate investment behavior, which show that a large portion of investment needs is financed from retained earnings.

Approximately one half of these financial needs have been provided through transfers from parent governments, and one half through other means, such as bona fide loans from the same parent governments, from other lenders within the country, or from abroad. Thus, the sample indicates that for each dollar of activity about 33 cents has been provided from central government revenues as a supplement for financing the activities of the corporation.

By regression analysis, it was determined that this transfer level was quite responsive to changes in the flow-of-funds ratio, but not so responsive to changes in investment behavior. Thus, active intervention by governments to improve the efficiency or to remove the subsidy element in pricing policies of government-owned corporations could lead concomitantly to smaller transfers from these governments.

In all the ratios, significant differences were recorded between four areas of the world,15 and among the six industrial types included in the sample. The European corporations had the lowest flow-of-funds ratio in the sample, barely covering expenditures (excluding depreciation, investment, and changes in financial assets). Latin American performance was similar, whereas Africa and Asia had flow-of-funds ratios averaging about 18 per cent of activity. Thus, the African and Asian entities were able to provide almost enough internal financing for replacement investment. By industry, only the petroleum and electrical corporations had positive flow-of-funds ratios in all the areas, with petroleum companies averaging almost 34 per cent of activity. The railways and other industries groups had negative flow-of-funds ratios—thus they had to be supplied with funds to cover expenditures. Including an allowance for depreciation, the railways needed two dollars for every five dollars of activity from sources external to the corporation.

It would be reasonable to expect that more profitable enterprises would invest more than would less profitable ones. This pattern was followed most strongly by industries in Latin America and Africa. On the other hand, when corporations of different areas were considered together, there was little positive association between profits and investment. These findings suggest (albeit not strongly) that geographic areas serve as barriers to international lending for investment. In addition, the lack of response of investment to profitability generally suggests that profitability is not an important factor in determining the long-run growth of government-owned corporations.

These statistical findings cannot offer more than intuitive guidance as to the “correct” pricing and investment policies of government-owned corporations. Any such evaluation would entail a detailed survey of the policy objectives in individual countries and their order of priority. It is unlikely that these objectives are the same in all countries and, even if they are, that they rank in the same order. Consequently, it may be a rational policy for a particular form of government-owned corporation, e.g., a railway, to run an operating deficit in one country, and an irrational policy in another.

Nevertheless, the methods of classification and statistical analysis used in this paper may offer an inducement to policymakers to review the performance of government-owned corporations in the light of the general requirements of economic planning. For example, the simple fact that subsidies from the central government budget are widespread and may represent a not inconsiderable proportion of government expenditures is sufficient to raise the question of the social opportunity costs of continuing to subsidize government-owned corporations. This is particularly important if these same funds have alternative uses of high priority in the government’s budget. In addition, the relative merits of tax reduction or, alternatively, a less stimulative government budget should also be contemplated as a substitute for additional subsidization of government-owned corporations.


I. Corporations Included and Years Covered

Railways (Includes the national railway system of each country listed)

  • Argentina, 1961/62-1965

  • Brazil, 1961-64

  • Chile, 1953-58, 1961-64

  • France, 1954-63

  • India, 1949/50-1962/63

  • Italy, 1956-64

    Malagasy Republic, 1959-63

  • Malaysia, 1955-62

  • Mexico, 1960-64

  • Nigeria, 1955/56-1963/64

  • Rhodesia, 1959-64

  • Sudan, 1956/57-1961/62

Other Transport

article image


article image


article image


article image

Other Industries

article image

II. Compilation of Sample Data

In order to evaluate the performance of the public corporations, an extensive effort was made to collect statistical information for a sample of such corporations. The purpose of this Appendix is to explain the process followed in the collection and classification of the basic data, and to present an illustration of the classification.

A statement of financial operations for a series of years was derived for each corporation included in the sample. The basic source of data was either the corporation’s annual report or statistics available to the authors for the country concerned.

The accounting methods of corporations often vary considerably not only from country to country but also from corporation to corporation within the same country; in order to achieve the greatest possible comparability, standardized rules were established for the classification of data.

Classification of Data

The main categories into which the basic financial data have been classified are defined as follows: A. Receipts and expenditures, B. Investment, C. Transfers to or from the central government, D. Changes in working capital/short-term borrowing, E. Changes in long-term debt, and F. Residual.

A. Receipts and expenditures

Receipts and expenditures have been derived from the profit and loss statements of the corporation. Usually the data were on an accrual basis rather than a cash basis. Thus, accounts receivable and payable are included in revenues and expenditures, and are compensated for by either short-term or other financing. The composition of receipts and expenditures follows.

Receipts on current account include (1) all sales revenues, (2) profit from sale of assets, and (3) all other general receipts; they exclude (1) receipts of subsidies from the central government, (2) changes in stocks, and (3) receipts from reserve funds within the corporation.

Expenditures on current account include (1) purchase of materials, labor, etc., used in the production process, (2) payment of interest, (3) subsidies to other enterprises, (4) loss from sale of assets, and (5) other general expenditures; they exclude (1) subsidies to the central government, (2) depreciation, (3) indirect taxes,16 (4) investment made for own account and charged to the profit and loss statement, (5) changes in stocks, and (6) payments to reserve funds within the corporation.

B. Investment

The investment figure is defined as the gross addition to fixed assets during the year, and includes additions to fixed assets proper as well as changes in work in progress, because it is not always possible to separate the two. When these data are not available, the investment figure equals the net change between the gross investment figure of the balance sheet for the year under consideration and the corresponding figure for the preceding year.

C. Transfers to or from the central government

Transfers include those to or from the central government as well as capital contributions and purchase of stock by the central government. Transfers to or from the central government are derived from the profit and loss statements and balance sheets (see long-term debt) of the corporation. Capital contributions and purchase of stock by the central government are calculated as the net change between the outstanding balance-sheet figure for the year under consideration and the corresponding figure for the preceding year.

D. Changes in working capital/short-term borrowing

Changes in working capital proper and changes in working capital outside the corporation were derived. The first includes changes in cash balances and inventories, and the second includes changes in accounts receivable and payable. The figures for these items are the net change between the balance-sheet figure for the year under consideration and the corresponding figure for the preceding year. As it was not possible to separate all the items listed above for some corporations, these two categories have been combined into working capital.

E. Changes in long-term borrowing

This rubric includes the net change in the long-term debt proper and long-term financial investment of the corporations (with institutions other than the central government). Both items are the net change between the balance-sheet figure for the year under consideration and the corresponding figure for the preceding year.

F. Residual

The residual is made up of items that are not listed above or of items that are listed above but that could not be separated for some particular years. In addition, it includes the residual proper. If our classification of data were completely accurate, and if all data were available for each corporation, the residual should be zero. In many cases, however, accounting practices were too intricate to be deciphered completely, and it has not been possible to separate items on the current side and the counterparts to these items on the financing side.


This section demonstrates the techniques followed in deriving the categories shown in Table 1. The balance sheet and profit and loss statement of the Nigerian Ports Authority for the year 1961/62 are used as an example of the type of basic source material that was available. These accounts are presented in Tables 1012; the items used here and in Tables 1012 are designated by letters. The main headings and their numbers correspond to those in Table 1.

Table 10.

Revenue Account for the Year Ended March 31, 1962 of the Nigerian Ports Authority

(In Nigerian pounds)

article image
article image
Source: Nigerian Ports Authority, Annual Report, 1961/62 (Lagos, Nigeria).
Table 11.

Net Revenue Account for the Year Ended March 31, 1962 of the Nigerian Ports Authority

(In Nigerian pounds)

article image
Source: Nigerian Ports Authority, Annual Report, 1961/62 (Lagos, Nigeria).
Table 12.

Balance Sheet as at March 31, 1962 of the Nigerian Ports Authority

(In Nigerian pounds)

article image
article image
Source: Nigerian Ports Authority, Annual Report, 1961/62 (Lagos, Nigeria).
article image
article image
article image


  • Agua y Energía Eléctrica, Memoria y Balance (Buenos Aires, Argentina), 1956/57–1965.

  • Air India, Annual Report (New Delhi, India), 1956/57–1961/62.

  • Baum, Warren C., The French Economy and the State (Princeton, New Jersey Princeton University Press, 1958), xvi + 391 pp.

  • Boiteux, Marcel, Peak-Load Pricing,in Marginal Cost Pricing in Practice, James R. Nelson, editor (Englewood Cliffs, New Jersey, Prentice-Hall, Inc., 1964), pp. 5989.

    • Search Google Scholar
    • Export Citation
  • British Broadcasting Corporation, Annual Report and Accounts (London, England), 1954–63.

  • British Electricity Authority and Area Electricity Board, The British Electricity Authority Report and Accounts (London, England), 1950–62.

    • Search Google Scholar
    • Export Citation
  • British European Airways, The British European Airways Report and Accounts (London, England), 1950–61.

  • British Overseas Airways Corporation, The British Overseas Airways Corporation Report and Accounts (London, England), 1950–61.

  • British Transport Commission, Annual Report and Accounts (London, England), 1951–62.

  • Central African Power Corporation, Annual Report and Accounts (Salisbury, Rhodesia), 1959–64.

  • Central Electricity Board of the Federation of Malaya, Annual Report (Kuala Lumpur, Malaysia), 1958–64.

  • Central Statistical Office, National Income and Expenditure (London, England), 1950–65.

  • Ceylon Transport Board, Statement of Accounts of Statistics and Reports of the Auditor-General and Chairman of the Board (Colombo, Ceylon), 1958–64.

    • Search Google Scholar
    • Export Citation
  • De Neuman, A.M., The Economic Aspects of Nationalization in Great Britain (Cambridge, England, Students’ Bookshops Ltd., 1952), 52 pp.

    • Search Google Scholar
    • Export Citation
  • Einaudi, Mario, Maurice Byé, and Ernesto Rossi, Nationalization in France and Italy (Ithaca, New York, Cornell University Press, 1955), x + 260 pp.

    • Search Google Scholar
    • Export Citation
  • Electricité de France, Rapport d’Activité: Comptes de Gestion (Paris, France), 1949–59.

  • Electricité de France. Travaux d’Investissement (Paris, France), 1949–59.

  • Electricity Authority of Cyprus, Annual Report and Accounts (Nicosia, Cyprus), 1958–64.

  • Electricity Corporation of Nigeria, Annual Report and Statement of Accounts (Lagos, Nigeria), 1952/53–1963/64.

  • Empresa Colombiana de Petróleos, Informe General a la Vicepresidencia Financiera (Bogotá, Colombia), 1951–64.

  • Empresa Nacional de Telecomunicaciones, Dirección Económico-Financiera, Informes Mensuales (Buenos Aires, Argentina), 1960/61–1966.

  • Ferrocarriles del Estado, Memoria Anual (Santiago, Chile), 1952–58 and 1960–64.

  • Gas Council and Area Gas Board, The Gas Council Report and Accounts (London, England), 1950–61.

  • Gas del Estado, Memoria y Balance General (Buenos Aires, Argentina), 1959/60–1965.

  • Gaz de France, Rapport (Paris, France), 1950–61.

  • Hanson, AH., editor, Public Enterprise: a Study of Its Organisation and Management in various Countries (Brussels, Belgium, International Institute of Administrative Sciences, 1955), 530 pp.

    • Search Google Scholar
    • Export Citation
  • Housing and New Towns Corporation, New Towns Act Accounts (London, England), 1951 and 1955.

  • The Karachi Electric Supply Corporation Ltd., Annual Report and the Chairman’s Review (Karachi, Pakistan), 1954–63.

  • Koyck, L.M., Distributed Lags and Investment Analysis (Amsterdam, the Netherlands, North Holland Publishing Company, 1954).

  • Kuhn, Tillo E., Public Enterprises, Project Planning, and Economic Development (Tegucigalpa, Honduras, Banco Central de Honduras, 1963).

    • Search Google Scholar
    • Export Citation
  • Lerner, E.M., and W.T. Carleton, The Integration of Capital Budgeting and Stock Valuation,The American Economic Review (Evanston, Illinois), Vol. LIV (September 1964), pp. 683702.

    • Search Google Scholar
    • Export Citation
  • Lipsey, R.G., and R.K. Lancaster, The General Theory of Second Best,The Review of Economic Studies (Cambridge, England), Vol. XXIV (No. 63, 1956–57), pp. 1132.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lóide Brasileiro, Relatorio (Rio de Janeiro, Brazil), 1960–64.

  • Malayan Railway Administration, Report for the Year (Kuala Lumpur, Malaysia), 1954–62.

  • Marschak, Thomas A., Theory and Policy in the French Nationalized Industries (Santa Monica, California, The Rand Corporation, 1960), 43 pp.

    • Search Google Scholar
    • Export Citation
  • Mauritius Central Electricity Board, Annual Report (Port Louis, Mauritius), 1958–64.

  • The Mexican Light and Power Company, Limited, Annual Report (Mexico, D.F.), 1957–63.

  • Ministero del Tesoro, Relazione generale sulla situazione economica del paese (Rome, Italy), 1956–64.

  • Ministry of Finance, Budget (New Delhi, India), 1949/50–1962/63.

  • Monrovia Port Management, Ltd., Financial Statements and Supplemental Schedules and Commentary for the Year and Accountant’s Opinion (New York, N.Y.), 1954–64.

    • Search Google Scholar
    • Export Citation
  • Morss, Elliott R., J. Eric Fredland, and Saul H. Hymans, Fluctuations in State Expenditures: An Econometric Analysis,The Southern Economic Journal (Chapel Hill, North Carolina), Vol. XXXIII (April 1967), pp. 496517.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • National Coal Board, The National Coal Board Report and Accounts (London, England), 1951–62.

  • Nerlove, M., Distributed Lags and Demand Analysis, Agricultural Handbook, No. 141 (Washington, D.C., 1958).

  • Nigerian Ports Authority, Annual Report (Lagos, Nigeria), 1957/58–1964/65.

  • Nigerian Railway Corporation, Report and Accounts (Lagos, Nigeria), 1954/55–1963/64.

  • Office de Postes et Télécommunications de la République du Dahomey, Résultats Définitifs de l’Exercise (Porto Novo, Dahomey), 1959–64.

    • Search Google Scholar
    • Export Citation
  • Petróleos Mexicanos, Balance General y Estado de Pérdidas y Ganancias (Mexico, D.F.), 1957–64.

  • Port of Callao Authority, Auditors’ Report (Lima, Peru), 1960–64.

  • Prest, A.R., and R. Turvey, Cost-Benefit Analysis: A Survey,The Economic Journal (London, England), Vol. LXXV (December 1965), pp. 683735.

  • Rede Ferroviária Federal, Relatorio (Rio de Janeiro, Brazil), 1960–64.

  • Régie Malgache des Chemins de Fer, Régie d’Acconage et de Manutention du Port de Tamatave, Bilan (Tananarive, Malagasy Republic), 1959 and 1961–63.

    • Search Google Scholar
    • Export Citation
  • Régie Malgache des Chemins de Fer, Reseau Ferré, Bilan (Tananarive, Malagasy Republic),