Thursday , January 27 2022
Home / Econbrowser - James Hamilton / Per Capita GDP Doing Just Fine, Linearly, since 1947

Per Capita GDP Doing Just Fine, Linearly, since 1947

Summary:
If you don’t believe me, take a look at this time series plot of available US GDP per capita. Figure 1: GDP per capita, in Ch.2012$ SAAR (blue); Linear trend (brown). Source: BEA via FRED, and author’s calculations. Linear regression yields t-stat on time trend of about 60 using robust standard errors. GDP per capita is rising about 40 Ch2012$ per quarter. Or, one might think to plot on a log scale. Figure 2: GDP per capita, in Ch.2012$ SAAR (blue, on log scale). Source: BEA via FRED, and author’s calculations. Which one is misleading? One reader thinks: In general, log scales are to be avoided for all but professional audiences, notably when exponential growth rates are involved. I would almost never use one in a commercial setting or for a general audience, because they are so easy

Topics:
Menzie Chinn considers the following as important:

This could be interesting, too:

Scott Sumner writes A disappointing Powell press conference

Scott Sumner writes What does it mean to say that something is inflationary? (part 2)

Scott Sumner writes What does it mean to say that something is inflationary?

Tyler Cowen writes *Labor Econ Versus the World*

If you don’t believe me, take a look at this time series plot of available US GDP per capita.

Per Capita GDP Doing Just Fine, Linearly, since 1947

Figure 1: GDP per capita, in Ch.2012$ SAAR (blue); Linear trend (brown). Source: BEA via FRED, and author’s calculations.

Linear regression yields t-stat on time trend of about 60 using robust standard errors. GDP per capita is rising about 40 Ch2012$ per quarter.

Or, one might think to plot on a log scale.

Per Capita GDP Doing Just Fine, Linearly, since 1947

Figure 2: GDP per capita, in Ch.2012$ SAAR (blue, on log scale). Source: BEA via FRED, and author’s calculations.

Which one is misleading? One reader thinks:

In general, log scales are to be avoided for all but professional audiences, notably when exponential growth rates are involved. I would almost never use one in a commercial setting or for a general audience, because they are so easy to misinterpret. I strongly prefer simple linear scales when possible.

Well, I think one could better characterize this series as exhibiting exponential growth (with unit root) than unit growth.

Update, 12/21 10:49pm:

Moses Herzog reminds me of how Steven Kopits lauded Jim Hamilton’s February 23, 2014 post on logarithms:

  1. Hugo AndréFebruary 23, 2014 at 5:43 pm
    Lovely blog post.

    I have had to learn to work with logarithms as part of my economics education but I had not (until now) realized just how helpful they can be.

    More of this kind of thing, please. 😉

    1. Steven KopitsFebruary 24, 2014 at 5:38 am
      Let me agree with, Hugo. I encourage you to do teaching snipets like this one. Some of us took econ a long time ago, and a brief refresher on various topics is very welcome.
Apparently, however, if I write something on logs, I’m out of touch.
Menzie Chinn
He is Professor of Public Affairs and Economics at the University of Wisconsin, Madison

Leave a Reply

Your email address will not be published. Required fields are marked *