Critically looking at data on ROC and economic growth over millenia | Macroeconomics | Khan Academy

So we've already talked about the general idea: the thesis that if the return on capital is greater than the growth of an economy, that could lead to inequality. Although we showed a case where, depending on the circumstances with the right numbers, that's not necessarily going to happen.

But what I want to do in this video is to think critically a little bit about some of the other data from the book. Once again, my point right over here is not to support or go against the ideas in the book, but really just to give you tools for thinking a little bit more critically about all of these ideas.

And what's really neat, as I mentioned in previous videos, is Piketty makes all of his data available, all of his charts available at this website right over there. Now, this is neat because this shows us the after-tax rate of return versus the growth rate at the world level. You don't see many charts that start at the year 0 and go over the course of 2000 years.

So you can imagine that the numbers back here are estimates, but let's just go with them for the sake of argument. So what it shows here is that the annual return or rate of growth. This is the pure return of capital after tax and capital losses. So after whatever taxes and any losses in the value of the capital or whatever else, that's here.

And then this is the growth rate of world output. You see, for most of at least the last few thousand years of human history, and the horizontal axis here isn't completely at scale, from here to here gets us almost well, almost 1500 years, while from here to here gets us about 50 years. It's not completely at scale, but you see that for most of human history, the annual rate, the growth of the economy was much smaller than the return on capital.

And once again, especially back here, these are just estimates, especially based on feudal times and whatever else. But on some level, they could be believable. Because in feudal times, when you had kings and all of that, you had a lot of coercive power by the feudal lords or the kings. They could, by force, force people to work on their fields or whatever else.

But if we take these numbers, we then see something interesting happens. As we go into the early 20th century, the after-tax return on capital drops below the rate of economic growth. So we see that right over here. This is over huge swaths of time. Even this data point, this is one data point that represents 37 years right over here.

And then we have another data point that represents the next, what is this, the next 62 years right over that. This is kind of the world that most of us know; this is modern times right over here, where the rate of economic growth, the rate of world output has been larger than the rate of return on capital after taxes.

Now, when you first just look at this chart, you say, "Oh my God, look! They're going to cross again. Maybe that means we're going back to the Gilded Age or feudal times or whatever else." But you have to be very, very, very careful here. Everything after this point is conjecture. This is a model. It might happen or it might not. We could see something very different.

So let me make it very clear. All of, let me just in another color, one that's… so all of this that I'm high—oops, that's not another color—all of this that I'm highlighting in yellow, this over here is all conjecture. You could go the other way. It is possible; it's completely possible that you have a reality... where this continues the trend that it's going on, where the world growth kind of does this actually, and then it goes.

So it goes from here, and maybe it does something like this. So, it's very important to realize that this kind of intersection right over here—this is a projection. You should look at his assumptions; you should decide for yourself on whether you think that this projection makes sense.

Because this is what is saying that, hey, this delta between r and g, that g is greater than r. This makes it look like, hey, it's over. But maybe it's not over. Maybe, when we average over the next—what is this going to be over—if we average over the next 38 years, maybe we get something like this.

And obviously, we're just at the very beginning of this interval so there's not a lot to go on. And that is also assuming, even if this does happen, that is still assuming this relationship that r greater than g leads to inequality, which in extreme forms could eventually lead to some form of a Gilded Age or kind of dynastic wealth, which could hurt innovation or which could even hurt—let me write that down—dynastic wealth, Gilded Age, which maybe hurts meritocracy or even potentially innovation or the economy as a whole if you have less of a middle class and people with purchasing powers.

But there's a lot of arrows here, and you need to decide for yourself which parts of this connection you agree with or disagree with or depending on the circumstances or the evidence you are more inclined to believe or not.