A big advance in national income research was made in this country in 1937 when in his book National Income and Outlay Colin Clark brought together figures not only of total income and its components but also of consumption, saving, capital (or asset) formation, public authority income and outlay and transactions with the rest of the world. Most of what is interesting in this branch of economics is derived precisely from tracing the relationships between such magnitudes as these, and certainly from the practical point of view of the user of such calculations engaged in the task of guiding and interpreting economic policy, little can be done with isolated series of one or other element in the picture. This at least was the point of view adopted by J. E. Meade and myself when we worked together on these problems in the early stages of the war and the rapid development of social (or national) accounting in the statistical work of many governments bears witness to the soundness of this view.

At first the problems of ensuring consistency in the treatment of transactions was tackled in terms of national aggregates, like the national income and expenditure, or of the consolidated transactions of large sectors such as ‘persons’. It came to be realised that conceptually the basic material for such studies is the set of ideal entries in a highly complex accounting system in which every accounting entity is represented. In these terms, the problem becomes—how to set up such an accounting system so as to provide the information sought by the economist and how to combine or consolidate the accounts in the system so as to yield meaningful and consistent aggregates.

I endeavoured to provide some sort of solution to these problems in a memorandum entitled Definition and Measurement of the National Income and Related Totals, which was discussed by the Sub-Committee on National Income Statistics of the League of Nations Committee of Statistical Experts at their meeting in Princeton, N.J., at the end of 1945 and subsequently appeared as an appendix to the Sub-Committee's report Measurement of National Income and the Construction of Social Accounts (1947).