Markup & Capitalization

Post your research in political economy, share ideas and workshop your empirical work.

Moderator: sanha926

Markup & Capitalization

Postby blairfix » Fri Dec 27, 2013 1:58 pm

Here is a chart that I find quite interesting. To make this graph, I ranked the top 4000 Compustat firms by market capitalization. I then "binned" this list into groups of 20 and calculated the average market cap and markup for each group. Note that the markup is a weighted average (total profit for all 20 firms divided by total sales). This helps to get rid of some outliers, as a few firms post incredibly large losses relative to sales.

The result is quite startling. Two trends are notable. First, higher capitalization corresponds with higher markup. Second, if we interpret the range of the markup for a given capitalization as the level of risk, higher capitalization functions to limit risk.

Markup vs Capitalization.png
Markup vs Capitalization.png (141.59 KiB) Viewed 3789 times
blairfix
 
Posts: 104
Joined: Fri Dec 20, 2013 3:45 pm

Re: Markup & Capitalization

Postby josephbaines1 » Fri Dec 27, 2013 2:55 pm

Fascinating chart, Blair.

Equally, it could be that lower risk functions to increase capitalization. It's of course hard to assign a direction of causality here, when risk itself is an integral component of the capitalization formula. Anyhow, I wonder whether you'd see the same heteroscedastic and curved distribution of data for 1950 (the first year included in the Compustat universe)?
josephbaines1
 
Posts: 17
Joined: Mon Nov 25, 2013 3:35 pm

Re: Markup & Capitalization

Postby DT Cochrane » Sat Dec 28, 2013 10:46 am

One of the most important, and perhaps under considered, aspects of CasP, is the role of market participants in constructing capital values. Nitzan and Bichler note that this is an inter-subjective process. I'd also argue that it is a 'quasi-objective' process. Market participants are not just pulling prices out of thin air. They are heavily equipped to monitor a wide array of events and processes, including the process of price construction itself. They watch the market, conversing with each other, they keep an eye on government announcements about unemployment and wholesale inventories,they hold teleconferences with corporate executives, and read analysts reports. They sample the world through an incredible array of mechanisms. This means causality flows in a complicated nonlinear feedback through these inter-subjective, quasi-objective assemblages constructing values. How can we talk about causality without the standard, built-in linearity that it implies?
User avatar
DT Cochrane
 
Posts: 67
Joined: Thu Nov 21, 2013 6:49 am

Re: Markup & Capitalization

Postby jmc » Wed Jan 01, 2014 12:24 pm

Interesting chart, Blair.

I wonder what it would look like if you then broke down markup to see the movements of profits and revenues. Maybe there was a difference between the top firms and the rest. Who knows, but it might be the case that the top firms tend, on the average, not to increase their markup through gains in sales.
jmc
 
Posts: 44
Joined: Mon Oct 19, 2009 2:33 pm

Re: Markup & Capitalization

Postby blairfix » Thu Mar 13, 2014 9:14 pm

Here's a few more charts that demonstrate the robustness of the long-term correlation between markup and capitalization. Large negative values for smaller cap firms seems to begin in 1990. I'm not yet sure if this is is a significant finding or merely due to something weird with compustat data.

Regardless, there is a consistent correlation for large cap firms that scales logarithmically. That is, markup is proportional to the logarithm of market cap.

I think this lends strong support to a power theory of value that builds on the post-Keynesian "cost-plus" theory of firm pricing. The line of causation?

power => capitalization => markup

1950.png
1950.png (121.32 KiB) Viewed 3657 times

1970.png
1970.png (129.19 KiB) Viewed 3657 times

1990.png
1990.png (166.28 KiB) Viewed 3657 times
blairfix
 
Posts: 104
Joined: Fri Dec 20, 2013 3:45 pm

Re: Markup & Capitalization

Postby blairfix » Thu May 08, 2014 1:45 pm

Another chart to show the robust relationship between capitalization and markup.

This time we look at differential markup vs. differential capitalization (per firm) between Quintile 1 and Quintile 2-5 of the Compustat Top 100 (ie: Top 20 Firms / Next 80).

There is a strong relationship through time between differential capitalization and differential markup.

Differential Markup.png
Differential Markup.png (161.5 KiB) Viewed 3616 times
blairfix
 
Posts: 104
Joined: Fri Dec 20, 2013 3:45 pm


Return to Research

Who is online

Users browsing this forum: No registered users and 1 guest

cron