The monetary equivalent of national product – national income – can be measured in various ways. One is to measure it as the «value added» by economic activity in agriculture, manufacturing, mining, and so on. (Value added is calculated by summing output at producers’ prices and deducting the cost of the fuel and raw materials used to produce the output.) Another way is to measure it as the aggregate value of the final products of the economy. Still another is to total the incomes accruing to persons supplying different productive factors (such as wages and salaries, profits, rents). Each of these approaches yields the same total, provided a consistent scheme of valuation is used. The component detail of each, however, illuminates different facets of the process of production, distribution, and consumption of the nation’s output, and each serves a different use.

Changes in national income may be measured either in current prices – the prices that prevailed during the year in which the economic activity took place – or in constant prices – the prices of a given year, for example, those of 1929, which then serve as a base. In a study of financial developments or market trends the former is often preferable. But if the purpose is to analyze change in consumer levels of living or national productivity, the latter is more appropriate. For purposes of studying economic growth, therefore, it is constant price measurement that is desirable.

There are two additional requirements for the measurement of economic growth if the purpose is to calculate change in material welfare. A nation’s rate of growth must be divided by the size of its population in order to find the rate per capita; if an increased number of people is required to produce an increase in the amount of goods and services produced, no one is better off than before. On the other hand, high levels of both population and output growth, even without corresponding growth in per capita output, bespeak an economy’s ability to sustain large increases in population, and this is of interest to students of the sources of national influence and power. A final point: the increase in output should not be a temporary one, such as might follow a year of unusually good harvests. Nor should it merely represent an upward movement in the business cycle. Economic growth is sustained growth, secular in duration rather than cyclical.

In the output data of various countries scholars have found growth cycles (often called «long swings») of varying lengths, some of them 10 years long, others 60 years, and still others even 100 years. In the data of American history the most common long swing, named the «Kuznets cycle» after its discoverer, the Nobel Prize-winning economist Simon Kuznets, ranges between 10 and 20 years. A swing is a change in the rate of growth. During a long swing there occurs an expansion phase, followed by a period of continued growth at a retarded rate, culminating in depression. In the 124-year period between 1814 and 1938, nine long swings have been found, averaging 14 years in duration. In the expansion phase of these swings gnp grew at an average rate of about 6 percent, followed by retardation averaging 2 percent. During the depression phase, the rate of growth was extremely low or, ceasing altogether, negative.

Except for agriculture, the pace of growth of nearly every kind of economic activity registered advances during the expansion phase. Long swings occurred in the growth of population, labor force, immigration, transport development, internal migration, geographical settlement, urbanization, residential construction, the prices of common stocks, railroad bond yields, the money supply, commodity prices, and still other economic variables. Long swings, it should be emphasized, took place not in the