Karl Smith and Matt Yglesias break down a graph from Thomas Piketty’s new book Capital in the 21st Century (published in French, but available in English next month). The graph shows the value of French capital as a percent of its GDP and shows that it fell in the middle of the 20th century and has increased in the last 40 years.
Or does it? Both Smith and Yglesias point out that what the graph truly shows is the changing nature of capital in the French economy. What we traditionally think of as capital (factories, stores, machines, etc.) has been fairly constant, at between 150-200% of GDP. But the value of agricultural land has fallen from almost 500% of GDP in 1700 to a negligible amount now (despite and because of the amazing increases in agricultural productivity). What has replaced it has been the value of French housing, which we can think of as the value of land, probably mostly in and around Paris. Eyeballing the numbers, it looks like it has increased from less than 100% of GDP to almost 400% of GDP.
So let’s think about housing for a minute. On the one hand, housing is clearly capital. It is a structure that provides housing services to whoever is living there. If it is not owner occupied then it provides a stream of income to its owner, just like other forms of capital. There is little difference, economically, between an 18th century rentier who owns large tracts of agricultural land and receives the rent coming from the product of the land and a 21st century rentier who owns many apartment buildings and receives the rent from the housing services they provide.
But what of the owner occupied housing that makes up most of the housing capital stock in the U.S. (I’m not sure about France)? It is clearly capital, in that it provides housing services and is an asset owned by the homeowner, but it is not clearly increasing the level of income inequality in the same way that agricultural land and other capital does. That is, capital in general will only increase income inequality as much as that capital in unequally owned. That was the case with agricultural land and is almost always the case with traditional capital as well.
But housing ownership in many developed countries has, I believe, been more equally distributed. And because owner-occupied does not directly provide income to its owner, it does not directly increase income inequality. To try to be a little clearer, imagine a world in which owner-occupied housing was the only capital in the economy. Housing and the land on which its sits, depending on its location, size, and desirability, would have different values. People with higher incomes would then purchase the housing that they liked the most and that fit within their budget. But the fact that someone who earned a million dollars a year might be able to purchase a Newport mansion while someone earning $30,000 a year had to make due with a two bedroom ranch would not further exacerbate the level of income inequality.
As my French has steadily gotten worse over the last 15 years, I’m waiting for the English version of Piketty’s book. But based on the reviews I’ve read, it certainly seems possible that the increase in the value of housing may be more related to economic growth, land regulations, and agglomeration effects. And the inequality of the value of housing and land may be more of a consequence of income inequality than a cause. In this case it may be that plus ça change, plus ce n’est pas la même chose.