How Would You Short The Internet?

Jeff Hulett
Dec 18, 2020
4 min read

Updated: May 26, 2022

The internet advertising world is in a bubble, similar to the environment that led to the Financial and Housing Crisis of 2008 and to the Stock Market Crash of 1929.

Ok, this deserves some context!

Tim Hwang’s book Subprime Attention Crisis focuses on the question of marketing effectiveness when it comes to big company advertising spend. Hwang explores the internet / digital ad world, which is growing very quickly and at the expense of traditional ad media (TV, Newspaper, etc). Please see the “Ad Evolution” graphic.

The premise is straight forward. For a variety of believable reasons, Hwang suggests the internet advertising world is in a bubble, similar to the environment that led to the Financial and Housing Crisis of 2008 and to the Stock Market Crash of 1929. He suggests internet advertising effectiveness is plummeting, with fraud and a lack of transparency hiding this reality. All the while, the cost of ads to company buyers are increasing. The annual internet ad spend is very large, estimated at $300B and almost doubling in the last 4 years. But even more significant is the associated systemic risk, given the internet has become so integrated into almost all facets of society. Naturally, there are some nuances to the comparison, but in the main, Hwang does a commendable job making the argument. Hwang suggests the financial underpinnings of the internet could crash resulting from the lack of confidence in the marketing effectiveness of digital advertising. Because ads are the financial underpinning of the internet, this poses an existential threat to the functioning of the internet itself. Certainly, the threat relates to many companies, like Google and Facebook, whose revenues are almost entirely dependent on digital advertising.

While I buy in to Hwang’s argument, he stops short of the “Big Short” question. Hwang suggests solutions related to government intervention and related ideas for an “orderly deflating” of the bubble. While I hope this is the case, my instincts suggest a “disorderly popping” may also be a realist bubble outcome scenario. Related to a disorderly popping scenario, you may recall from Michael Lewis’ book (and related movie), Michael Burry and others devised a short strategy to trade on the informed belief the US Housing Market was going to crash. They end up making a bunch of money while exposing a massive problem.

So my question is this:

If you were a hedge fund manager, like Dr. Burry, and you had access to significant investment funding, how would you short the internet?

My initial thought was to short or buy put options on bundles of related Social Media / Internet platform companies. The problem is, the timing of the bubble bursting is so hard to know. If you recall from Burry’s experience, he was almost undone by the timing of his short strategy. There was a tremendous inertia, owing to the market makers and institutional players that stood to lose if the short trade was successful.

The trade thesis is that the value of marketing effectiveness is well below what is being paid for the advertising. That is, if one measures how much revenue is being generated from $1 of digital ad spend, the actual revenue is relatively close to zero. Per Hwang, this could be (and has been) measured by Randomized Control Trials ("RCT"), the testing gold standard for causal determination. So the arbitrage thesis is relatively straightforward. But how long will it take for all these ad buyers to perform the RCT based tests, draw the ineffective conclusion, and start a rapid exit from digital advertising? How much “exit inertia“ delays the exit because of the buyer’s dependence? That is, the jobs of the company buyers that manage the advertising budgets depend on marketing effectiveness. Will momentum be created by a “race for the exit?” Will the movement toward an “orderly deflation” get more momentum?

In terms of trading this investment thesis, is there a way to implement a long term strategy that only activates on certain internet advertising market signals? I would love to learn your ideas.