There’s this @ecipe study (funded by Google) making the rounds, alleging that the European Commission's DSA will cost 2 million jobs and €85 billion per year in lost GDP! That’s a lot.

So I looked into it.

1/
Dubious methodologies and identification, but let’s ignore them. The core of the study is this: they have a bunch of country-sector EU data from 2010 to 2017. They estimate some TFP and see how TFP is affected by an “ex ante regulation” dummy that covers the period 2015-2017.

2/
What happened 2015? Explanation buried in footnote 3: “Regulation 2015/2020 shifts the general antitrust regulatory approach from ex post to a universal ex ante obligation in the EU telecom sector.”

3/
I’ve studied telecoms my whole professional life and I can tell you: *nothing* changed in EU telecoms regulation in 2015 that brought regulation from ex post to ex ante.

This study is just ridiculous. Can journalists please avoid even mentioning it? Much appreciated.

Fin/Ciao

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Tommaso Valletti

Tommaso Valletti Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @TomValletti

31 Oct
Thought-provoking piece – yet again! – from Germán Gutiérrez and @ThomasPHI2.

Common wisdom about dominant firms in the digital economy at odds with productivity paradox. Tremendous innovations by large “superstar” alongside lacklustre aggregate productivity growth. True?

1/ Image
1. Market share of dominant firms has not increased (Panel A: domestic sales to domestic GDP; Panel B: consolidated global sales to global GDP).

2. Relative productivity of dominant firms has not increased since 90s (Panel A: top US firms; Panel B: global stars).

2/ ImageImage
3. Contribution of top firms to labor productivity has *decreased* by about 40%

Conclusion: not a paradox after all? Top firms now are neither larger nor more productive than past top firms. In fact their contribution to overall growth has declined -> paradox explained

3/ Image
Read 4 tweets
13 Sep
Thread on research in economics & its linkage to policy.

Recent top publications on hot competition policy topics, eg
@jan_eeckhout & al on increasing margins
@joseazar @martincschmalz & al on common ownership
@cunningham_lbs @florianederer @ProfSongMa on killer acquisitions

1/
These papers freaked out corporate interests. Big Pharma, Big Tech, ‘Passive’ Investors (& good chunk of the legal community who’s behind them) reacted, as expected.

As they don’t know what academic research in economics is, they use shills under semi-academic camouflage.

2/
They pay for their own self-serving studies. Funding is opaque, to say the least. In some cases, they attack and harass.

These lobbying (sorry, research) centres don’t publish in any academic meaningful sense. And never will.

3/
Read 13 tweets
30 May
Important THREAD on the etymology of CAPPUCCINO.

First the spelling. It’s not capacino, capucino, capucinno, etc.
It's CAPPUCCINO (singular), 2P’s, 2 C’s, 1N. CAPPUCCINI (plural), still 2P’s 2C’s, 1N.

Second the etiquette. 1/
You don’t drink it after noon. You don’t order it after lunch, let alone after dinner. And absolutely terrible to drink it DURING a meal (I feel bad just at the thought of it). You don’t drink it in a plastic cup. You don’t drink it on the go. You don’t drink it with a straw. 2/
(Personal note: I split up with a British girlfriend when I tried to explain all this, observing in shock that the table next to ours had just ordered Hawaiian pizza (the horror!) and cappuccino at 10pm. Obviously the relationship was not going very well.) 3/
Read 8 tweets
26 Apr
I have been asked why, if in normal times we think competition is a feature of market-based economies that produces the best outcomes, we now hear that with corona "putting aside any qualms about competition" should be the best way forward for society (thanks @martincschmalz). 1/
Let me use simple steps using IO. As a starter, I’d say a plausible scenario could be one where there are uninternalized positive externalities, e.g. knowledge spillovers (or coordination problems in the supply chain), so that a competitive equilibrium is not ‘good enough’. 2/
But we already have e.g. a system of R&D exemptions for that… So one has to further ask why a) externalities are now bigger than before, b) a relaxation of comp rules would actually fix those externalities & c) the current system is not good to provide exemptions (too slow?). 3/
Read 6 tweets
10 Dec 19
Little THREAD. Remember Zuck's speech at Georgetown? He argued that FB would not remove of political ads because they make so little money (1-2% revenues) from them, so let's keep 'em as it’s good for democracy to expose lies on FB. This is fallacious on 2 grounds possibly 3. 1/
1) No one really has an idea about how much money FB makes directly from political ads, since their accounts are opaque to say the least. But let me imagine 1-2% is the true figure (still, quite a bit of money since 2019 revenues will be around $90bn). 2/
2) Business model of FB is to keep users engaged as long as possible (ie spend time), so that they can micro target with ads. Politics, as well as sex and violence, keep people online. It’s the indirect effect that matters: political ads drive business. This is a LOT of money. 3/
Read 5 tweets
4 Nov 19
Phenomenal paper by @DrDaronAcemoglu et al.
It responds to “but people don’t care about privacy”, “I have nothing to hide”, “it’s all about revealed preferences”, “we get a barter, it’s all good”. Too fast.
There is in fact a massive negative externality. Here’s how it goes 1/
With fully rational consumers (a big if), individuals who don’t care about privacy release their data. This compromises those who do care and whose information is correlated. (Cambridge Analytica scandal:data shared by 1/4m FB users but allowed to infer info about 50m users!). 2/
A data market? Once information leaks, less incentive to protect privacy. And less to gain from giving data away, as the platform already knows about you: market price of data is low. This gives the “impression” that users do not value privacy, but it’s due to externality. 3/
Read 4 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!