E-mail: Young@ilo.org
Fax: +41 22 799 6957
Phone: +41 22 799 7152

This paper is placed on the following websites: www.stat.gov.pl                                                                        www.econ.nyu.edu/dept/iariw

INTRODUCTION

The depth and intensity of economic and social transformations are a usual motive of debate. Without a historic perspective it is normal that some see events as superficial changes within a rather constant structure while others believe that the structure is changing. Therefore, in this paper there is, of course, no aim to dismiss the topic. Nonetheless, for those measuring income and wealth, and more concretely household income, changes occurred or under evolution today, constitute conceptual and methodological challenges. As discussed in the other papers of this Session, both those working at the macro and micro level have to deal with consequences of these changes on the area of household income.

The criteria used here to select a few, within all those challenges, is twofold: that they are of quantitative and qualitative importance and up to now the Canberra Group and countries have not found an empirical solution to measure them, or they are still not important enough. In some cases, these are issues which practitioners are increasingly being confronted with, and which they should be aware of, even though there are as yet no concrete recommendations on their measurements. Therefore, the methodology to measure them is not included in the other papers of this Session. They will probably be in the agenda of future Canberra Groups. There are, of course, many other important effects of economic, demographic and social transformations that are not directly linked to the measurement of household income and, therefore, will not be discussed here.

Four have been selected: the changes in the role of public and private sector, especially in the area of social services consumed by households; the increasing importance in economic activities of the transnational enterprises, especially in developed countries; the fundamental role that micro-enterprises and self employed play in the labor market of developing economies; and the new financial instruments associated to the present stage of capitalism. Figures will be presented that show that a good measurement of income distribution will depend on the possibility of handling theoretical and operational problems that they pose.

1. THE CHANGING ROLE OF PUBLIC AND PRIVATE SECTOR

The present patterns of development have introduced important changes in the roles and in the functioning of the public and private sector. Among these changes, the public sector is progressively concentrating in social public expenditure and getting rid of activities related to the direct production of goods and services. In this last aspect, more emphasis is being placed in its regulatory role especially in markets that are naturally monopolistic or oligopolistic.

In relation to social public expenditure, some very important changes have been introduced in its operation. Probably, one of the most important is the fact that in social security matters the principle of solidarity has loose ground in favor of the principle of individual accounts. Something similar has occurred with health, education and housing. To materialize these principles, the private sector is playing an increasing role in all of these areas. Furthermore, another policy orientation is being proposed and has relation to the subsidy of demand. It pursues to make monetary transfers to individuals or families to which public produced services were rendered by the government. The proposal pursues replacing the in kind service by a monetary transfer with the condition that it would be used to select a similar private produced service. Here, it is believed that efficiency would improve and that the beneficiary would benefit from the election of the producer of the service.

In paper 2 of this Session the topic of public services in kind, social insurance, social security and social assistance has been discussed. Up to now, in most countries, in relation to the pension system the role of the state and the solidarity principle seems to prevail. Nonetheless, as mentioned there, the so-called excluded schemes such as non-employee pension scheme need a different treatment. Whenever an individual financial fund is established, and when the value of that fund is related to funds with variable value and interest, many accounting problems arise and furthermore, it is not always easy for the owner of the fund to have a clear picture of its current financial state at any moment.

In relation to social transfers in kind, at present the possibility of distributing these services between households and estimating its costs constitute a theoretical and empirical difficult exercise. If an increasing trend towards private provision of services prevails and if the financing of the cost of those services changes from present patterns, this will constitute a challenge of important magnitude to the measurement of household income and consumption. At present, the private insurances for health are being widely used in some countries. Private education is having an increasing importance and a great effect in the distribution of human capital, especially in countries where its quality is, in most cases, superior to public education. Probably, the old dilemma of valuing public sector at their cost and private ones including benefit should introduce changes in the relative importance of social services in extended private consumption. It is also true that if individually financed households services systems prevail, the distribution and cost of these services could become easier to measure at least at the micro-level.

Therefore, changes in trends that increase the importance of the private sector in social services will transfer importance from the system of accounting of public services to the accounting of social services provided by the private sector which in many ways constitute a quite different financing and property system that will require further study of items that exist today, but that are empirically not very important for households as a whole.

2. TRANSNATIONAL ENTERPRISES 

During the eighties and the nineties there has been an increasing trend towards the creation of huge enterprises on the one end and a proliferation of micro-enterprises and self-employment on the other end. The first phenomenon is more intense in developed countries, while the second in developing countries.

An effort has been made to describe the increasing importance in the economic activities of this type of enterprises and of the different institutional characteristics of their function. The aim is to illustrate the growing challenges for measuring their activities and especially the need of cooperation among countries for being able to capture correctly their activities. It is also worthwhile to remember that due to their size they are becoming increasingly important in many components of household income. Therefore, a failure in progress in this field of statistics will weakened the quality of income distribution measurements.

The increase in the size of enterprises is both a national and international trend. Among the explanations, the importance of the scale of production and the fact that there is room for increasing returns of scale seem to play a very important role. Most of the theoretical and econometric models, until recently, have used the assumption of constant or decreasing returns of scale. In the last years, and especially in relation to the growth productivity of the United States economy, models that allow for increasing returns of scale are being used (See Klein). The increasing internationalization of economies that has followed the liberalization reforms that are being implemented throughout the world has extended the phenomenon of growth of enterprises to the international level. The increasing importance and number of transnational enterprises is nowadays one of the most important features of the international economy.

Associated to the expansion of transnational enterprises, a survey conducted in 1996 with responses for more than 3.100 transnational corporations and international experts shows emerging trends. Within them, the following are pertinent to the objective of this paper:

A rapid rise in the proportion of sales generated abroad.
A likely increase of production carried out abroad.
Market access remains the decisive motive for the choice of location.
The drive to internationalization is more marked for sales and production functions than for research and administrative activities.
An increase reliance on mergers, acquisitions, joint ventures and alliances is a vehicle for international expansion.
Increased emphasis on developing countries, reflecting a massive increase in local production capacities, while major restructuring will take place in developed countries.

These trends highlight that the transnational enterprise activities are carried out in different countries and that the relations between headquarters and subsidiaries of an enterprise are essential for them. The accounting system of such type of enterprise is obviously complex and the need to have a policy that considers maximization of profits, minimization of costs is not at all independent of transfer prices and of the countries where taxes will be paid.

Some figures illustrate the increasing importance of these trends. Mergers and acquisitions involving operations where enterprises of more than one country are involved (value of cross-border mergers and acquisitions) increased their value from 160 billion dollars in 1990 to 544 billion dollars in 1998. Of this total, deals where the purchasing company acquires more than 50% of the equity share rises from 116 billion dollars in 1990 to 411 billion dollars in 1998 (see tables 1 and 2).

In the United States, in the 1980=s there were generally fewer than 2.000 mergers in acquisitions per year. In the 1990=s they jump from the value of the 1980=s to values as high as 8,000. In terms of valuation the increase has been even more striking.

Table 1
Cross-border mergers and acquisitions purchases
(Billions of dollars)

Table 2
Cross-border mergers and acquisitions sales
(Billions of dollars)

Business alliances such as Disney-Capital Cities, Time Warner-Turner Broadcasting, and NBC Microsoft are examples of such big deals that are sweeping across industries. Nearly 20.000 alliances were form world wide in the last 2 years, and of the over 6 thousand US alliances formed, more than half involved foreign partners. Outside the United States, 75% involved more than one country. The reasons for this growing trend are to share risk, enhance competitiveness by accessing core capabilities to enter new markets, leverage complementarity asset position, and boost market shares. A survey of 5 thousand US companies between 1987 and 1995 revealed that the average return on investment from strategic alliances was over 16 percent, which is higher than the industry average. The fact that sharing information is a central element that does not correspond to a classical transaction is one of the features that calls for a special treatment of this new institution.

Furthermore, mergers and especially acquisitions have bring light to the fact that intangibles related, for example, to the technological process underway within the enterprise, and the potential share and access to markets, give a potential value to an equity that is normally quite different and much bigger that the one that could be deduced from traditional accounting or the present value of shares in the market.

Both, the fact of operating in different countries and of uncertainty in the value of shares constitute a major challenge for the measurement of income and wealth. In effect, the consistency and meaning of different balances in different countries are already Statistical National Offices studying a problem. There are several initiatives in connection with this topic. The United States conducts periodically a very large and detailed survey of the worldwide structure of American enterprises. Canada and France conducted a limited experiment designed to compare reporting structures of French and Canadian transnationals. Nonetheless, there is a critical need for better understanding the way in which these enterprises operates is extremely important. It is recognized as a very complicated problem and as a big challenge for the accounting system. Each of these balances includes a type of balance of payment of the company with subsidiaries, customers and providers all around the world. Furthermore, one of the sectors where the transnational phenomenon is very important is financial services, in itself a very complicated one before the transnational phenomenon. Therefore, when looking at data from the macro side (National Accounts) to deduce income payment to households will be a much more difficult exercise than at present.

In paper 2 of this Session, the topic of capital gains is discussed. In the example included, real capital gains (after deducing inflation) are relatively low. The nineties are full of important examples where real capital gains are, in fact, very important. For example, the expansion of the United States economy finds one of its explanations in the wealth effect associated to the very significant real increase of equities. If one of the most important shareholders in the economies are households and the value of the transnational enterprises is variable, difficult to estimate, and volatile, the challenge to include this concept in household income estimates will be extremely great.

3. INFORMAL SECTOR

Employment in micro-enterprises, especially in the informal sector, has increased at high rates in developing countries in the last twenty years. The relative importance of this type of employment can be illustrated by the fact that, according to ILO estimates, employment in the informal sector for the nine Latin-American countries that make up most of the population reached, in 1999, 58% of total employment. Also Charmes gives figures of 74%, 43% and 62% respectively for Sub-Saharan Africa, North Africa and Asia for the share of informal sector employment in non-agricultural employment in the 1990s. Furthermore, during 1999, there was a fall in employment in the private modern sector and practically all-new net employment was informal.

Contribution to GDP has been estimated as between 20.0% and 37.0% in some African countries in the early nineties (See Table 3). Figures available for a few countries indicate that income from the informal sector represented 17.0% and 36.0% of total income in Venezuela (1987) and Zambia (1986) respectively with the corresponding percentages of female informal sector earnings to total female earnings given as 12.7% and 62.4% in these countries. These figures reflect in part the low income derived from informal sector enterprises compared to income from other sources within the economy as a whole. Income from informal sector activity however represents a very large percentage of the total income of those households engaged in the sector. When this is taken in conjunction with the high employment figures in the sector, it is seen that the accurate measurement of informal sector income is of importance in the evaluation of the distribution of household income, especially at the bottom end of the distribution. Due to the nature of informal sector enterprises, this measurement has however posed difficult unresolved problems and so remains a major challenge in the assessment of the distribution of income.

* excluding subsistence agriculture.

Source: Charmes, J.: Contribution of Women Working in the Informal Sector in Africa: A Case study, Report of the 2^nd Meeting of Expert Group on Informal Sector Statistics (Delhi Group), April 1998, Tables 1 and 2

Firstly there is need for clarity in the identification of the informal sector, the definition of which has always been a subject of debate. Should it be based on informal employment, that is in terms of work arrangements of those engaged in both formal and informal enterprises such as low productivity, dangerous or precarious work? Or should it be defined in terms of the ‘informal’ characteristics of the enterprise (with both formal and informal workers)? Should ‘informal’ be distinguished from the concept of ‘unregistered’, ‘unrecorded’, ‘unobserved’? The latter, representing a gap in the statistical information system of a country, is of particular interest for national accounts and would include enterprises engaged in the production of illegal goods and services or avoiding tax, labour laws, etc. This would however make the concept of the informal sector depend on the level of sophistication of the statistical system of the country.

In February 1993, the Fifteenth International Conference of Labor Statisticians (ICLS), convened in Geneva by the Governing Body of the International Labor Office adopted a resolution on statistics of employment in the informal sector. The Conference decided to base the concept of the informal sector on the informal characteristics of the enterprise and to limit its coverage to those production units which Atypically operate at a low level of organization, with little or no division between labor and capital as factors of production and on a small scale.@ The resolution also notes that Athe concept of informal sector activities should be distinguished from the concept of activities of the hidden or underground economy@. Employment in the informal sector is then determined as that labour directly linked with informal enterprises.

For statistical purposes, the resolution regards the informal sector as a group of production units which form part of the household sector as household enterprises or equivalently, unincorporated enterprises owned by households, according to the definition of the United Nations System of National Accounts (SNA 1993).

These enterprises are production units not formally nor legally incorporated into companies, working on a small-scale, with rudimentary organization and with little or no distinction between capital and labour. They are thus not legally separate from their owners and do not keep a complete set of accounts. In addition, they operate on low capital, low productivity, with limited access to organized markets, credit, training, modern technology, public services and amenities and frequently are without fixed location or not very visible and not regulated by government. They exist mainly to generate income and employment for the survival of their owners and mostly in developing countries.

The informal sector comprises (i) Informal own-account enterprises and (ii) Enterprises of informal employers. The former are household enterprises owned and operated by own-account workers and contributing family workers without employing hired workers, while enterprises of informal employers are those which employ one or more employees on a continuous basis. These latter enterprises may be defined in terms of their size (number of employees) or whether or not they are registered in accordance with national legislation while the former may be defined in terms of the registration criterion or taken in its entirety. It should be understood that these are not sufficient conditions. There may be enterprises that satisfy these criteria, e.g. size, that are definitely not informal in the sense of having little or no organization in their operations and not keeping a complete set of accounts. For example, among its recommendations, the resolution states that units engaged in professional or technical activities should be included in the informal sector only if they fulfil the requirements of items (i) and (ii) above. With regard to domestic workers, these should be included or excluded from the informal sector depending upon national circumstances. Out-workers were also to be included within the sector. Finally, it is noted that for practical reasons the scope of the informal sector may be limited to enterprises engaged in non-agricultural activities.

The ICLS definition was adopted for use in SNA 1993. It is now widely accepted, especially in the Delhi Group, as the best basis for defining the informal sector although some regard it as not sufficiently tight for operational uses. It proposes that information should be collected directly from enterprise owners not only on the estimate of employment but also on the production function, institutional relations and other characteristics that are useful for analyses and policy. Within these characteristics, income is one of the most important. It recommends a mixed household-enterprise survey as the best source. In this, enterprises are identified through their owners using household surveys and then the enterprises are canvassed directly, preferably at their location.

For various reasons, some countries collect data only on persons engaged in the informal sector using labour force surveys or population censuses. This is done by many Latin-American countries, for example. To identify such persons, ECLAC uses the concept of low productivity employment that consists of non-professional and non-technician working on enterprises of five or less persons or self-employed. It is however difficult to get reliable details about the enterprises themselves from this source. Some other countries use household budget surveys to identify and collect statistics from household enterprises. The usually small sample sizes of such surveys are not likely to yield an adequate number of such enterprises. Yet others have tried establishment-type surveys, economic censuses etc. Ambulant and house-based activities tend to be omitted from the sampling frame of such surveys.

The reports of the meetings of the Delhi Group on Informal Sector Statistics, formed within the framework of the UN Statistical Commission, capture well difficulties and methods oriented towards solving the problems of identification of informal sector enterprises. Although the issue of measuring enterprise income has been mentioned at some of these meetings, substantial discussion of it is not expected until the 4^th meeting in 2000. Early estimates of the value of production were mainly indirectly derived using the residue technique and not directly through surveys.

The difficulties in direct measurement through surveys are many. One is that these enterprises are heterogeneous and vary from those requiring little or no capital and skills and with marginal returns, such as street trading or shoe-shining, to others such as car repairs, food processing and manufacturing. They are not very visible, have very short lifetimes, usually mobile in location and widely dispersed, seasonal in their operations and small in size. Mostly poor, they range from those that are destitute to those with income much higher than the minimum formal sector wage. For example, Charmes estimated that street vendors= incomes in the cities of Benin in 1992 ranged from 1.1 to 4.5 times the minimum wage. The pluri-activity of their owners and shared ownership also create some difficulties.

Clearly, as a result of this heterogeneity, assessing the income accruing to owners of these enterprise for use as their self-employment income cannot be done uniformly. Some authors are trying to develop a classification of these enterprises based on degree of informality which could be used to split them up into more homogeneous sub-groups. Group-specific approaches could then be developed to measure income. For example, Charmes has put forward a possible framework for collecting income data from street vendors.

The concept of mixed income, usually accepted as the way to measure the return to labour from self-employment, could be used for the few of such enterprises at the top end of this system,. For those with limited capital and using little or no inputs, we need to explore the use of receipts while some of the others could be assessed using >drawings=. Asking for prices and quantities of goods/services and the percentage of total sales these represent over the reference period instead of the direct question on the value of production, is another possibility.

A problem arises with the reference period that should be used in collecting the data. Given the absence of records, bank accounts, and so on and taking into consideration the nature of the enterprises, we mostly have to use a very short reference period such as a day or a week to minimize recall errors. It would then be necessary to assess the likely period for which the enterprise could have existed over a 12 month period to annualize the data. Alternatively we could allow the reporting unit to use whatever reference period it finds convenient and then use supplementary information such as frequency of purchases of intermediate goods and of other key expenses to convert this value to that for a standard reference period.

Some of the difficulties in measuring income from informal sector enterprises no doubt also apply to other household enterprises belonging to the hidden or unrecorded economy. To the extent that these other enterprises are hidden out of deliberate choice, these problems become even more difficult.

The above proposals for measuring this type of income have so far not been adequately developed and tested. They need to be experimented as part of the broader challenge on how best to measure this type of income. It has become evident that unless a significant effort is made, data on the income originating in the informal sector will not be available and, therefore, income distribution estimates at least for developing countries will lack a very important component.

4. FINANCIAL INSTRUMENTS

During the nineties, the value of international financial transactions grew at a high rate. In fact, its quantity surpasses in many times the value of international trade transactions. These last ones added to direct foreign investment represent approximately a 7 or 8% of the 1.3 to 1.5 billion of working day transactions in the exchange rate market. The instruments that are being used in these transactions are rather new, at least in its relative importance. One of the most important is derivatives. According to the Bank for International Settlements the stock of financial derivatives added to near 400 billion at the end of 1998. The topic is conceptually and operatively difficult enough as to have been one where a modification of SCN 1993 was recently approved in the Statistical Commission, not without difficulties to come to a consensus. These instruments and increasing financial liberalization of the economies have had influence in recent international financial crisis. On the other hand, the massive use of Internet and the creation of enterprises associated to its operation and possibilities have influenced the equity markets. Changes in the long-term rate of interest have also played a role in the diminishing of the USA budget-deficit. All these changes mean a constant change in the value assets and therefore capital gross and losses are frequent and qualitatively important. Furthermore, tax on capital gains could constitute an important source of incomes for the governments and the financial situation could change if gains are replaced by losses.

These and other changes in the financial market do not seem to have at first view a clear relation with household income. Nonetheless, at least part of the savings of households such as those belonging to their investment in mutual or pension funds are dependent of the value of equities and of other type of investments included in the portfolio of those funds. If the present trends towards strengthening of the financial aspects of capitalism continue to prevail, capital gains and losses are not independent of all these new forms of financial instruments and of their volatility.

Therefore, when the aim is to study income distribution and when financial instruments represent such high values in relation to real economy, their influence could be extremely significant. At present, in terms of wealth and of income distribution the eighties and the nineties have shown concentration in many countries. It is not at all clear how well have capital gains been included in those figures of income distribution. It is also possible that at present an increasing percentage of financial investment of households is done offshore and that the development of financial institutions and instruments of international character make it difficult to obtain figures in practice of these trends. All these trends illustrate that as in the case of transnational enterprises it would become increasingly difficult to make a follow-up of the value and distribution of financial capital and, therefore, new approaches and methods to measure it and the flows they originate are also increasingly necessary.

--------------------------------------------------