Comment on the global rise of corporate saving by Peter Chen, Loukas Karabarbounis, and Brent Neiman

Abstract

The Chen, Karabarbounis and Neiman (CKN) paper provides interesting global evidence regarding the financing of corporations, and the distribution of aggregate saving between the corporate and household sectors worldwide. To execute their study on global corporate saving, the authors must map the theoretical constructs of corporate profits, investment, issuances, dividends and interest into their empirical counterparts, and, crucially, they must make decisions about what constitutes saving by corporations vs.~households. A key consideration is that, in order for it to be meaningful to distinguish corporate from household saving, and to distinguish internal vs. external financing of the corporate sector's activities, there must be some violation of the Miller and Modigliani Irrelevance Theorems regarding capital structure. That is, if the corporate sector is simply a veil, it is irrelevant whether aggregate saving resides on the balance sheet of households or corporations, and whether corporate activity is financed using internally generated funds, or by raising external finance. Thus, I begin my comment by laying out these definitional choices and highlighting where different assumptions regarding violations of the Miller and Modigliani Irrelevance Theorem (which I will call frictions'') would lead to different measurement (and hence possibly different conclusions). I think this is helpful, because the authors' definition of saving, while reasonable, is very close to a measure of aggregate corporate retained earnings, or corporate net worth. While this is one measure of corporate saving, there are other, also reasonable, definitions. The most appropriate definition depends on the precise question of interest. Potential research questions include: Has the corporate sector changed how it finances total investment, internally or externally? Has the corporate sector changed the composition of investment between productive physical capital, and liquid financial assets (or cash'')? Does whether the corporate sector retains or pays out its profits matter for the distribution of income and/or wealth? These are questions that may actually be better answered with alternative measurement schemes, as I illustrate in my comment.