Pieria: Perverting Piketty

I have a new post on Piketty on Pieria, pointing out potential problems interpreting his premises and propositions (sorry, it started organically):

I recently wrote about the numerous misconceptions over Thomas Piketty’s use and definition of capital in his book Capital in the 21st Century. Sadly, it seems there are a number of other common, equally important mischaracterisations of Piketty’s model floating around. Here I will consider 5 of the most widespread and show, using direct quotes from Piketty himself, why they are off the mark. The first 3 are simple errors of interpretation with regards to Piketty’s theoretical framework, while the latter 2 are problems with how people have responded to Piketty in general. Although the latter 2 are inevitably more subjective, they are still important for trying to understand and reframe the debate between Piketty and his critics.

Each point gives the common misinterpretation of Piketty’s work, and counters it. For example, one of the most important points (IMO) is this one:

2. ‘Fundamental laws’ of capitalism?

The claim: Piketty’s ‘fundamental laws of capitalism’ are not fundamental at all.

The reality: Although calling them ‘laws’ is misleading, at no point does Piketty claim that his laws are inviolable. They are instead tendencies (with the exception of the first law, which is just an accounting identity) which push capital’s share of income in a certain direction over time, but can be counteracted by any number of things, and only take hold over a long timespan.

Hopefully this will be a useful resource for when people who haven’t read the book (or worse, have read it but clearly either rushed or lack reading comprehension) repeat silly canards about it.



  1. #1 by Rob Rawlings on July 2, 2014 - 3:38 pm

    On your third section on “Savings and accumulation”

    I do not think it is true to say “But Piketty’s definition of savings already includes depreciation”.

    The point that K&S are making is that if you use net savings then the capital/income ratio increases much less as growth falls than if you use gross savings. That is: once you build into your model the fact that you have to pay for depreciation out of savings then a given level of savings will cause total capital to increase less than if you ignore depreciation.

    His model uses net savings to drive his results then almost as an afterthought he mentions that one could alternatively have used gross savings. He never, to my knowledge, mentions that if you do indeed use gross savings that this significantly reduces the impact of his “law” on the capital/income ratio.

    This strikes me as a bit of an intellectual sleight-of-hand.

    • #2 by Unlearningecon on July 3, 2014 - 11:15 am

      Maybe the confusion is mine, but I’m not following your comment (or at least, the third paragraph). Early on you state that: “The point that K&S are making is that if you use net savings then the capital/income ratio increases much less as growth falls than if you use gross savings.” I agree with this, but you then go on to state that “if you do indeed use gross savings that this significantly reduces the impact of his “law” on the capital/income ratio.” Is this just a typo, and you meant net savings, or am I missing something?

      • #3 by Rob Rawlings on July 3, 2014 - 8:08 pm

        Piketty’s model assumes net S stays constant. The text book models assume that gross S stays constant.

        Its clearly an empirical matter which (if either) is closer to reality. K&S claim that net S does not stay constant thru time and sometime goes negative.

        I agree that most people think of savings out of income as net of any depreciation expenses they may incur. But from an economy-wide perspective then it seems more reasonable to see savings as “anything that is not consumed” irrespective of whether that savings goes on replacing worn-out capital or adding to the capital stock. But again its not the definition that matters (as long as its consistently applied) but how well it maps onto reality. My concern is that Piketty has chosen his definition of S based more on the conclusions he is trying to reach than what is more real-world.

      • #4 by Unlearningecon on July 4, 2014 - 6:15 pm

        I’m not sure about the evidence on this, but I don’t think it’s fair to say Piketty is trying to reach certain conclusions. Dynamics, rather than a “balanced growth path”, are the most important aspect of his 2nd law. The main problem with KS is that they completely misinterpret this: as I say in the piece, the 2nd law is supposed to describe the direction in which the capital/income ratio evolves rather than its magnitude. If growth is 0, and net savings are positive, the capital income ratio will increase indefinitely – that’s what β = ∞ should be interpreted as.

        If we choose to put depreciation in the denominator instead, this gives similar results for β. For example, with g = 0, δ = 0.1 and (gross) s = 0.1, we find β = 1, so the the way β will be changing depends on whether it is currently above or below 1. If δ were 0 we would again find indefinite growth of β, since, as above, the savings would not be used up on depreciation. I don’t see how KS’ distinction is especially important for Piketty’s law.

    • #5 by Rob Rawlings on July 3, 2014 - 2:37 pm

      Oops, yes I have muddled my use of net/gross savings and rendered my point very muddled!

      To restate: Piketty uses savings net savings in his model which drives a much higher capital/income ratio than using gross savings. He mentions in passing that one could use gross savings but not (to my knowledge) that this would significantly alter his results.

      Additional point : K&S explain why using gross savings makes more sense. Does Piketty justify his use of net savings anywhere ?

      • #6 by Unlearningecon on July 3, 2014 - 6:42 pm

        Not directly, but this passage hints at why he thinks it should be so:

        This depreciation is substantial, today on the order of 10 percent of GDP in most countries, and it does not correspond to anyone’s income: before wages are distributed to workers or dividends to stockholders, and before genuinely new investments are made, worn-out capital must be replaced or repaired.

        What he’s saying is that peoples’/firms’ actual income is net of depreciation, so it makes sense to model their savings decisions as a function of this rather than of gross savings, as KS do. I think this makes sense but KS object to it simply on the grounds that it is not the same as textbook theory, which to me is no objection at all. What are your thoughts on why people would save their income plus depreciation?