In this paper we consider the question of measuring aggregate economic growth and its sources. We derive a theoretically justified solution for aggregating (across firms, industries, countries, etc.) growth rates and their sources within the framework of Solow?s (1957) growth accounting method. The resulting aggregation scheme turns out to be quite intuitive and, in fact, the one that is sometimes, in an ad hoc way, used in practice, and so the main value of our work is that our formal derivations clearly show under what conditions this scheme is theoretically justified. We also provide a small empirical illustration of our method on the real data set and show how different the conclusions can be depending on the aggregation scheme used.