Early on in the book, Piketty makes a strong claim for the essential interdisciplinarity of the social sciences. Yet there is, in the 21st century, a social phenomenon with the potential to upset all his calculations, which receives only token attention. One can refer to this as "coupled natural and human systems," i.e., situations where natural systems and human systems, each following their own dynamical laws, have a transformative impact on one another. Anthropogenic, carbon-driven climate change is only a special case of this, albeit the one with the greatest destructive potential. (Another example can be seen in the recent algae bloom on Lake Erie.)
None of the demographic or economic projections on which he relies seem to take the effects of climate change into account. The fundamental assumption that runs through his book, that absent either a catastrophe on the order of the World Wars, or a major change in political climate, the rate of return on capital will return to historical norms of 4-5% per year, takes no account of that either. His discussion of the phenomenon is limited to a single two-page subchapter entitled "Climate Change and Public Capital" (567-569).
Even the title of the subchapter shows with what little seriousness he has considered the problem, as climate change poses a threat not only to public capital, but to capital as such. The very thing on which Piketty's entire research program is based is the ability to assign monetary value to the existing capital stock. This, in turn, is based upon the presumption of the predictability of the various natural inputs into the productive process. Farmland--which Piketty shows accounted for a significant share of total national capital in all wealthy countries until the end of the 19th century--is assigned a value based on its ability to produce a certain set of crops in an average year. Urban real estate, whose increased share of the national capital stock in the last 100 years corresponds roughly to the decline of farmland's share, has value to the extent that it can either be rented out as residential real estate or serve as a site for productive economic activity. In either case, the ability to reliably transport sufficient food and potable water into the urban conglomeration is a precondition for that value. Both the value of capital and its rate of return depend conceptually on the ability of markets to approximate long-range average estimates for future productivity.
Climate change (and other coupled natural and human systems) call this fundamentally into question. Not only does the year-to-year volatility of natural inputs increase, but the very physical configuration of the asset threatens to change from decade to decade. Consider Miami: It is only a matter of time before it is swallowed by the sea. Yet the population continues to grow, new buildings continue to be built, and real estate prices continue to rise. Operating in abstraction from natural processes, the social sciences, whether separately or together, can only assign quantitative measure to the frenzy of a collective delusion. It is necessary to measure such delusions, but it is far from sufficient.
Both bourgeois political economy (of which Piketty is a peculiar example--I suspect one could call him neo-Ricardian, but my grasp of the history of political economy is not quite strong enough to defend that as a thesis) and the type of vulgar Marxism exemplified by Soviet planning, treat of nature in a purely extractive manner. Nature is just there, a trove from which human labor can, with a greater or lesser degree of success depending upon the extent of scientific knowledge and technical capacity, extract needs or wants that, even if they derive from a biological substrate, are defined through predominantly social processes. Nature may be fickle but it is unchanging, or at least, functions more or less cyclically according to laws which are gradually uncovered through human ingenuity.
That picture of nature should have gone out the window more than a hundred years ago when Arrhenius first hypothesized the greenhouse effect. But human beings learn slowly, especially when there is money to be made by remaining ignorant.
In effect, Piketty makes a case for fundamental unity of the social sciences at a moment in history when what we should be learning is how to think about the complex articulations between social and natural processes. He makes a necessary case, 100 years too late.
I will let Piketty have the last word, however, because his last words are good ones, that in an understated manner make a claim with which I strongly agree: "Refusing to deal with numbers rarely serves the interests of the least well-off."