Capital in the 21st Century, the zeitgeist book by Thomas Piketty, the winner of Prospect's 2015 World Thinkers poll, has become a totem for the ubiquitous interest in income inequality, regardless of whether people have read or understood the book, or agree with its conclusions. It is fitting then that we should ask Matthew Rognlie to take a bow.
Matthew who, you ask? Rognlie is a graduate student at the Massachusetts Institute of Technology, and at the tender age of 26, he has just published probably the most robust riposte to Piketty yet. Starting off as a blog, Rognlie’s work has morphed into a 50-page academic paper, published at the Brookings Papers on Economic Activity. Entitled modestly, "Deciphering the fall and rise in the net capital share," Rognlie’s paper is not easily accessible and is dense with algebra. Yet, this apprentice says some important things that sometimes contradict Piketty, while offering conclusions that are by no means incompatible with those of his sorcerer.
Piketty’s central theme, by way of reminders, is that because the rate of return on capital normally exceeds the rate of economic growth, capital accumulates, the rich who own capital get richer and pass it to their progeny, and income inequality intensifies. Observing these trends over long swathes of history and the shift in income composition from labour to capital, Piketty asserts that the future is pre-ordained, and that we can only address the problem by globally co-ordinated wealth taxation.
Rognlie isn’t totally comfortable with all of Piketty’s insights. He disputes that the rate of return on capital is high and reasonably stable over time. He thinks it is prone to decline from time to time as diminishing returns to the accumulation of capital set in. We might think, for example, of the speedy obsolescence of new products and processes, courtesy of rapid changes in technology. This means that owners of capital have to keep writing off and re-investing their returns instead of living off them.
He thinks the recent behaviour of income shares is misunderstood. He concurs with Piketty that the capital share of income fell sharply in the first half of the 20th century, but says that instead of recovering steadily in the wake of the second world war, it traces a U-shape, falling until about the 1970s, and only then recovering. He also thinks the future behaviour of income shares may not be as Piketty has insisted. The main reason is that capital accumulation will only happen if there is continuous substitution of capital for labour, with income flowing to the former. It is possible that digital technologies and robotics will validate Piketty’s view, but Rognlie and others wonder whether we are all assuming too much here, and that there will be limits to substitution of machines for men and women, in which case labour returns might benefit in due course.
Rognlie, however, also goes to places Piketty didn’t visit. In particular, he breaks capital up into housing and non-housing and argues that the behaviour and long-term increase in the capital share of income are almost entirely due to gains originating in the housing sector, and that the non-housing share of income has been unremarkable. This is an important distinction.
In G7 countries, Rognlie shows that the non-housing capital income share only made a partial recovery after the 1970s and remains below what it was in the 1950s. In the US, for example, the non-housing share dropped from 22 per cent in the 1950s to 18.6 per cent in the 1970s before recovering to 19.4 per cent in the 2000s. In the UK, it fell from 27.2 per cent to 18.3 per cent before rising to 23.4 per cent. But the housing share of capital income share in G7 countries rose over three times from about 3 per cent in 1950 to over 10 per cent today. In the US, it rose from 5.3 per cent to 8.2 per cent, and in the UK from 1.2 per cent to 7.3 per cent. These figures, of course, do not take account of the collapse in housing asset values during the financial crisis or what has happened subsequently.
Nevertheless, if Rognlie is right about this, the implications are significant. Home ownership is more broadly distributed than capital ownership, per se, and so standard income distribution arguments stand on weaker ground. If homeowners are the new rentiers, we will have to re-frame the debate from the politics of capitalism, the role of organised labour, and the income distribution implications of rapid technological change, to the effects of QE on asset prices, policies affecting home ownership and house-building, and planning regulations and other polices that affect the supply of building land.
In the light of the forthcoming election in the UK, Rognlie’s work throws up some interesting issues about the way we tax housing capital. If we think housing wealth inequality merits attention, inefficient tools such as business rates, stamp duty, and the mooted mansion tax should be binned in favour of a revamp of Council Tax, still based on 1991 house price valuations, and a new system of land value taxation—in effect, a charge on the rental value of all land. The likelihood of these proposals seeing the light of day in party manifestos is small, but Matthew Rognlie’s work should give us all food for a lot more thought.
Matthew who, you ask? Rognlie is a graduate student at the Massachusetts Institute of Technology, and at the tender age of 26, he has just published probably the most robust riposte to Piketty yet. Starting off as a blog, Rognlie’s work has morphed into a 50-page academic paper, published at the Brookings Papers on Economic Activity. Entitled modestly, "Deciphering the fall and rise in the net capital share," Rognlie’s paper is not easily accessible and is dense with algebra. Yet, this apprentice says some important things that sometimes contradict Piketty, while offering conclusions that are by no means incompatible with those of his sorcerer.
Piketty’s central theme, by way of reminders, is that because the rate of return on capital normally exceeds the rate of economic growth, capital accumulates, the rich who own capital get richer and pass it to their progeny, and income inequality intensifies. Observing these trends over long swathes of history and the shift in income composition from labour to capital, Piketty asserts that the future is pre-ordained, and that we can only address the problem by globally co-ordinated wealth taxation.
Rognlie isn’t totally comfortable with all of Piketty’s insights. He disputes that the rate of return on capital is high and reasonably stable over time. He thinks it is prone to decline from time to time as diminishing returns to the accumulation of capital set in. We might think, for example, of the speedy obsolescence of new products and processes, courtesy of rapid changes in technology. This means that owners of capital have to keep writing off and re-investing their returns instead of living off them.
He thinks the recent behaviour of income shares is misunderstood. He concurs with Piketty that the capital share of income fell sharply in the first half of the 20th century, but says that instead of recovering steadily in the wake of the second world war, it traces a U-shape, falling until about the 1970s, and only then recovering. He also thinks the future behaviour of income shares may not be as Piketty has insisted. The main reason is that capital accumulation will only happen if there is continuous substitution of capital for labour, with income flowing to the former. It is possible that digital technologies and robotics will validate Piketty’s view, but Rognlie and others wonder whether we are all assuming too much here, and that there will be limits to substitution of machines for men and women, in which case labour returns might benefit in due course.
Rognlie, however, also goes to places Piketty didn’t visit. In particular, he breaks capital up into housing and non-housing and argues that the behaviour and long-term increase in the capital share of income are almost entirely due to gains originating in the housing sector, and that the non-housing share of income has been unremarkable. This is an important distinction.
In G7 countries, Rognlie shows that the non-housing capital income share only made a partial recovery after the 1970s and remains below what it was in the 1950s. In the US, for example, the non-housing share dropped from 22 per cent in the 1950s to 18.6 per cent in the 1970s before recovering to 19.4 per cent in the 2000s. In the UK, it fell from 27.2 per cent to 18.3 per cent before rising to 23.4 per cent. But the housing share of capital income share in G7 countries rose over three times from about 3 per cent in 1950 to over 10 per cent today. In the US, it rose from 5.3 per cent to 8.2 per cent, and in the UK from 1.2 per cent to 7.3 per cent. These figures, of course, do not take account of the collapse in housing asset values during the financial crisis or what has happened subsequently.
Nevertheless, if Rognlie is right about this, the implications are significant. Home ownership is more broadly distributed than capital ownership, per se, and so standard income distribution arguments stand on weaker ground. If homeowners are the new rentiers, we will have to re-frame the debate from the politics of capitalism, the role of organised labour, and the income distribution implications of rapid technological change, to the effects of QE on asset prices, policies affecting home ownership and house-building, and planning regulations and other polices that affect the supply of building land.
In the light of the forthcoming election in the UK, Rognlie’s work throws up some interesting issues about the way we tax housing capital. If we think housing wealth inequality merits attention, inefficient tools such as business rates, stamp duty, and the mooted mansion tax should be binned in favour of a revamp of Council Tax, still based on 1991 house price valuations, and a new system of land value taxation—in effect, a charge on the rental value of all land. The likelihood of these proposals seeing the light of day in party manifestos is small, but Matthew Rognlie’s work should give us all food for a lot more thought.
