Piketty’s Capital is a stupendous achievement. It lays out a majestic panorama of the rise and fall and rise again of the wealthy over two centuries in several countries. It has a bearing on some unexpected questions. For example, I have a scholarly interest in nineteenth-century French landscape painting. During that period, more than 2,000 paintings on average were exhibited every year in the Paris salon, and similar numbers were rejected. The best of these have become part of humanity’s heritage; millions look at them in museums and reproductions abound. But how did they come into being? Who paid for them? Piketty shows that the French elite were immensely wealthy. The rich were numerous enough to sustain the hopes and expectations of thousands of artisan painters, and the pinnacle achievements of a few of them. Like other enduring assets of humanity, this art was created in response to the tastes and the buying power of the rich.

Piketty’s prime achievement is a rich statistical description of the inequality of income and wealth in France, the US and Britain over two centuries. But something is lost as well. His focus is on the top 10 per cent, 1 per cent and 0.1 per cent, and in his story these are the only ones that count. The rest are written out and denied historical agency. During the past few decades, social invisibility has been lifted by the academic study of history from below. Inadvertently, in pursuit of a quantitative handle on the rich, Piketty has restored a history from above. This exclusive focus on the wealthy also imparts an explanatory bias: the waxing and waning of elite wealth cannot be explained entirely by observing the rich. That, however, is what Piketty attempts to do.

THE MODEL

The core concept is a purported historical regularity, namely that over the long run, r > g, where r is the prevailing business interest rate and g is the rate of economic growth. That in itself, if it were true, would be a simple mechanism for concentrating income and wealth. It does not require anything else. Borrowers would be paying more to service their debt every year than the growth of output as a whole.