Thread by @TomValletti

Back to the question on vertical mergers (THREAD). I had a little spat with @geoffmanne a short while ago

https://twitter.com/TomValletti/status/1090325132596400128

He sent me a an interesting submission of the Global Antitrust Institute (GAI) of George Mason University to #FTChearings.

1/

They review 13 academic papers from 2009-2018 and conclude “these papers continue to support the conclusions that consumers mostly benefit from vertical integration”. Note, they write *integration*, but we need to know about vertical *mergers*.

https://twitter.com/geoffmanne/status/1090349679303450624

2/

He invited me to have a look at the papers they selected. Which I did. And the devil is in the detail.

First, one paper (Ashenfelter et al.) is about an horizontal merger in the beer industry, discussing the balance between price increases and efficiencies being passed-on.

3/

Second, several papers focus on “transaction cost economics” more than on the interplay foreclosure vs. price effects (reduction in double-marginalization). This is the case for Forman & Gron, or Malik (more on this below), and, partly, for Hanssen, and Gil & Warzynski.

4/

Third, 2 papers are not from Econ: Baker et al. in Health (regression table not shown), and Malik in Management. Former seems to provide a sound analysis, latter unclear.
Many doubts about robustness of Austin (Econ Bulletin ?), aspects of the analysis unclear. I would ignore
5/

Fourth, only 6 papers actually look (partly) at mergers (or simulated mergers), instead of comparing integrated vs. non-integrated firms (with the ex-ante integration decision being endogenous): Suzuki, Taylor et al., Atalay et al., Austin, Koch et al., Crawford et al.

6/

For these (taking Austin out – see point above) the GAI reported effects on welfare are accurate for Suzuki (“mixed effects”), Taylor et al. (“no economic significance”), and Koch et al. (“not addressed”). However:

7/

For Atalay et al. hard to draw welfare-related conclusions because even if productivity goes up with integration, resulting price effects or potential foreclosure effects not assessed and balanced in paper (GAI reports a positive effect on welfare due to vertical mergers)

8/

For Crawford et al., GAI misrepresented the overall findings, which I would qualify more as “mixed” (GAI reports a positive effect on welfare due to vertical mergers).

Here is a more detailed paper-by-paper analysis if you’re interested (sorry for small print).

9/

Yes, there is something we can learn also from other papers. “Horizontal merger in beer CAN BE SEEN as presenting the effects of integration when there are exclusive-territories" or "studies on studio forced divestiture after US Paramount CAN BE SEEN as 'reversed mergers'."

10/

But HOLD ON, this is not how we make our decisions, where the discussion IS about fundamental trade-off between foreclosure incentives and reduction in double marginalisation.

In summary, out of the 13 GAI papers, only 5 talks directly at vertical mergers and seem robust

11/

And out of these 5:
- one does not address welfare (Koch)
- two have mixed results( Suzuki and Crawford)
- one finds no effects (Taylor)
- one is difficult to draw conclusions (Atalay)

Wow. What kind of advice can you give to the @FTC based on this evidence?

12/

Good that @chopraftc and @RKSlaughterFTC started thinking deeper in recent Staples merger.

Am I being bitchy? May be - you judge. Trying to defend good economics. Economics too often falsely reported, especially by non-economists that speak econ jargon from the 80s.

/end