Life is full of activities that lead to the satisfaction of consumers' needs and hence their welfare, only some of which can be classified as economic. In extreme cases the distinction is easy. Taking a pleasant walk or playing a game of chess with a friend satisfies certain wants, but is not an economic activity; working in a factory or an office is. But what about the household services performed by the housewife and other members of the family? What about cultivating one's own vegetable garden? It has become customary to base the distinction between economic and noneconomic activities on the closeness of ties with the market. Every pursuit whose products are either sold on the market or are largely directed toward it is treated as economic; no others are, though their yield in the way of satisfying wants may be substantial. This solution has a great advantage in that it segregates the sector of life concerned largely with economic activities, and in which measurement is feasible because the yardstick (no matter how it may have to be adjusted) is the market price. In a highly developed economy the disadvantages are reduced by the fact that the majority of the activities intended to produce goods for consumers are market-bound. Even so, the magnitudes omitted are far from minor. For example, the value of housewives' services are roughly estimated at some $2 3 billion in 1929, or more than onefourth of national income. And in countries where the market is less developed than in the United States, the limitation of economic activities to those market-bound leads to a major undercount.
The national income estimator must choose between comprehensive definition—with the consequence that large sectors of the economy either cannot be measured on a continuous basis or cannot be included with more precisely measurable sectors because the errors are so enormous—and a narrower definition that confines economic activities to those market-bound—for which tolerably reliable estimates can be made.
In current national income measurement in this country, the decision is usually in favor of the second alternative. And it finds support in the argument that the activities so segregated for measurement are the ones subject primarily to economic criteria and rationale; whereas those that are not directed at the market are much more a part of life in general. One may and does discharge a housekeeper for inefficiency in managing a household, but by itself this is rarely a ground for divorce.
However justified, this limitation results in omitting a substantial group of activities important in satisfying the needs and wants of the members of society. Moreover, some market-bound activities are omitted largely because they cannot be measured on a continuous basis—taking boarders or lodgers, spare-time jobs, and the like. In coverage, a continuous national income series is thus always on the short side even in terms of market-bound activities, which it tends to omit if they are casual and hence elusive of measurement.
The national income estimator cannot do much about such omissions, since scarcity or lack of data is inherent in the nature of the omitted areas. But in interpreting national income movements in terms of satisfying consumers' wants, the limitation of national income largely to noncasual market-bound activities must be stressed. In this country as in many others where the market is always being extended, the relative importance of the household as a source of consumer goods is declining. Many activities formerly performed by the housewife or other members of the family and not measured (baking, sewing, canning, etc.) have progressively been taken over by business enterprises and gone into market-bound activities; other household functions have vanished without leaving a direct substitute in business activity. Hence, national income totals tend to exaggerate the upward movement in the supply of goods to consumers, if such supply is comprehensively defined as coming from both market-bound and family activities. Likewise, a comparison of the national income of two countries at different stages of the commercialization of family production must take into account the differing importance of the market sphere in the total provision of goods to consumers. The omission of casual activities also imparts an upward bias to the secular trend of national income, since their importance relative to those covered diminishes as more people move to cities and engage in regular, full time, pursuits.
The effect on the interpretation of short term changes in national income is at least as great. During any expansion, whether associated with business cycles or with wars, people move from nonmarket to market areas and from occasional to full time jobs; and in the larger net product the proportion of measurable market-bound activities increases at the expense of nonmarket activities or occasional jobs. As many of us are all too aware, during recent years, when the pressure of war needs for the expansion of market-bound production was especially intense, the number of persons available for family household work decreased materially. Total net production, including production within the household, increased much less than production on farms, • in factories, shops, and offices. During short term contractions, on the contrary, the shrinkage of the market sphere swells the number of persons available for services both within the household and for casual jobs. Being confined to noncasual market-bound activities, national income is thus a more cyclically sensitive index than a more comprehensive total that would include the large productive sector of the household as well as occasional jobs and pursuits. Variations in it therefore exaggerate short term changes in the more comprehensive total.