Economist Thomas Philippon’s new book, “The Great Reversal: How America Gave Up on Free Markets,” went on sale this past week, highlighting the United States’ failure to block the country’s largest companies from inhibiting fair competition.
Alongside my review of the book, I spoke with Philippon to dig deeper into why entrepreneurs and venture investors should be concerned by his findings and to hear his response to some of the critiques he’s likely to receive from Silicon Valley.
“The broad picture is that competition is good, but surprisingly fragile,” he said. “In today’s environment, the U.S. is moving from a place where it was at the forefront of having free markets that worked pretty well for most people to being a laggard in many industries.”
Here’s the transcript of our conversation, edited for length and clarity:
Eric Peckham: Does this analysis suggest that VCs should be looking for more opportunities in Europe, because startups have more opportunity there against incumbents, or you’re findings should be read to that level of individual decision making?
Thomas Philippon: I would not go as far as that. The takeaway is that in the U.S. there are some industries where it will be extremely useful to have more competition. Look at the problem of ISPs and how expensive internet is. In New York City, you’re lucky if even have two options, a choice between Verizon FiOS and Time Warner Spectrum. The prices are $79.99, or $79.99. The barriers to entry are too high. It is too costly to go and compete with these guys.
You laid out several options of how to potentially deal with the GAFAM (Google, Apple, Facebook, Amazon, Microsoft) companies, as you framed it. One of which is to restrict them from acquiring so many small startups as a means of taking out potential competition early.
Entrepreneurs and VCs no doubt like that these big tech companies are so active in offering to acquire their startups and provide an attractive exit path aside from an IPO. What’s you case for those worried about inhibiting these big companies from acquiring so many startups?
Well, that’s a specific issue. There is definitely a drop in the number of public offerings in the U.S.. The number of traded companies in the U.S. is half of what it used to be. We’re not talking about 10% move. It happened through mergers and through a lack of IPOs.
As a VC, you clearly like having more options as an exit strategy. So you know that being bought by an existing firm is a good option and has always existed as an option. On the other hand, if all the previous generations of incumbents had just bought each other, then you wouldn’t have the competitive landscape.
One reason this is a tough question to answer is that I think competition is oftentimes a public good. It’s in nobody’s interest to protect it. We all benefit from it but it is in no individual’s interest to protect it.
Could there be other variables here driving the shift to fewer companies going public and more startups exiting through acquisition instead of IPO? Perhaps there’s a greater premium in the market for the strategic value of a company as opposed to its underlying financial value, which means strategic acquirers are willing to pay a lot more than the public market would value a company. Or is there truth to this notion that’s been thrown around a lot in the startup world that public market investors just don’t understand how to value tech companies the right way?
Why would that be more the case today than in the past? Why would there be that difference? There have been technology companies for a long time. The reason firms are going public less is because there is more private financing. There’s more money and more flexibility to get funding without going public. So I think that’s one reason.
And yes, probably some excessive regulation in public markets. That may be true, although the trend started way back before Enron so I don’t think that explains it very well.
Thomas Philippon is the Max L. Heine Professor of Finance at New York University, Stern School of Business
And then, of course you have these big monopolies and the best way to keep their monopoly is to offer extremely high premiums to acquire young competitors. That’s evident in why Facebook made the big check for WhatsApp, and for Instagram as well, to an extent. They pay to prevent their own reign from being threatened. That’s that’s pretty straightforward.
It’s not in the interest of the founders or investors of Instagram or WhatsApp to say “no” to a deal like that, but it is in the public interest to have competition so you need somebody to enforce that.
Regarding the GAFAM — which is a term that hasn’t caught on in the U.S., you guys refer to them as “Big Tech” I suppose — these companies are much less exceptional than people think. I’m all in favor of treating them exactly like we treated IBM before, AT&T before, with the same amount of respect and the same amount of enforcement.
Essentially, everything people say about these companies is wrong. You hear all the time that they are unprecedented. There is absolutely nothing unprecedented about them. If you look at the top five market cap firms in the US since 1980, the top five firms by market cap have always been 10% of the stock market. And today it’s 10%. The names change: GE comes in then drops out, GM comes in then drops out. The only one always in the top 10 is Exxon.
If you look at their profit margins, their profit margins are the same as those of AT&T, IBM, etc. from the top 10 going back 50 years. The top firms have always been very profitable. They are not growing faster now or creating more jobs or creating more productivity than firms in the past.
You point to the GAFAM’s small number of employees and the fact that they build most things in-house so have limited vendor relationships as an argument that they are less integral to the economy than the largest companies of past eras. GM, AT&T, and others employed far more people and had whole economies of other businesses around them dependent on them as customers.
These tech companies act as platforms that so many other companies have built businesses off of in some way, however, right? Like brands who have built community through their Facebook page or through their Instagram following and want to kind of keep everyone in one place.
They are certainly less integral to the economy than the star firms of the past. Being a platform is not the same as being in a long term relationship with suppliers where innovations are produced jointly. There have always been platforms. Platforms are simply marketplaces. The medium has changed but the economics perspective is the same.
One difference is that the time to get critical mass of users is much quicker now than it used to be. So that is true. But I mean, the simplest way to think about it is to just do a thought experiment about take one company in the top 50 years ago. Imagine that company doubled its productivity overnight. Imagine what that would do to the U.S. economy.
Now do the same thing with Facebook. The obvious difference is that today we have literally zero impact. In the past, if you doubled the productivity of GE it would have a big impact on the economy. All the companies upstream and downstream would have higher productivity. An ad targeting business doesn’t do this.
What’s interesting here is the significant difference between Amazon and, say, Facebook. If you do the thought experiment for Amazon then for sure you would have a big impact. But the productivity of Facebook and others doesn’t contribute to the overall productivity growth of the economy. It’s not to say these firms aren’t doing something good — they are. But definitely they are not so precious that we should be worried about doing anything about them from a regulation perspective.
Does this argument that they’re not special translate over to the antitrust framework we have to deal with them? You provided some potential changes to the metrics regulators should use to determine whether an antitrust review in needed, like going from a revenue-based threshold to a valuation based threshold. Generally speaking though, do we need a next generation antitrust framework to oversee these firms effectively or is what we have now fine and we just need to actually use it?
We do not need a new framework. Network effects and fixed costs are nothing new; and increasing economies of scale is nothing new. We have the tools, it’s about the political will to use them.
At the margin, you could make some case that some of the definitions would be updated and some of the restrictions changed. But the tools are there. Honestly, the mandate of antitrust legislation is extremely broad and flexible. The ideology of the judge is much more of an issue in this case.
I am in favor of treating the GAFAM the same way you would treat any company that has too much market power. The argument that we can’t do anything about them because we will break them and it’s going to be a disaster for everyone — it’s an argument I’ve heard in many different ways — is completely bogus.
You hear arguments like if you want Facebook to be able to moderate content then it has to be huge and profitable, otherwise they won’t be able to do it. This is kind of a veiled threat that if you do anything against us, we’re going to start misbehaving and not moderate anything. I think that’s totally bogus, and also that’s a sign that they have too much clout.
It’s a common refrain that GDPR and efforts at protecting IP or moderating content have a disproportionate effect on startups. That it’s actually hindering the potential competitors to the big tech companies instead of challenging them.
That’s classic. That’s exactly what happened in finance too. If you make regulations to protect consumers that are heavy handed, then the paradox is you’re going to favor the incumbents more because they are the only ones who have the budget and the lawyers to understand the regulations and comply. That’s a classic side effect of some regulations.
Europe is establishing itself as a stronger market for software startups and venture capital but it still lags the U.S., without a doubt. If EU regulators are ostensibly doing a better job blocking anti-competitive behavior by the largest companies, what are the other barriers that need to be addressed?
In Europe, the two big impediments are the universities, which are definitely not as good as here, and the lack of a single market because of national regulations and cultural differences, including the language.
It seems like there are some sectors where we actually do see more activity from startups in Europe than in the U.S. though, like in fintech in particular with these challenger banks that are creating new banks from scratch. Startups in the U.S. have just built a new interface for an existing bank because it’s impossible to start a new bank in the U.S.
Absolutely true. But I’m not sure it is going to have a huge impact on market shares: it is going to push the banks to get their act together.
Take the telecom industry. The U.S. is the worst market in the world to have a cell phone. It’s ridiculous. You pay more than twice what you would pay in the UK or Germany or France for worse service.
In Europe, we encourage entry. In France before 2011, we had an oligopoly with three firms, all legacy carriers who were making sure there was no competition. People would pay €40-60 per month. And then because of a new license to operate, Free Mobile came in at €20 per month for the same contract. Within six months the incumbents matched the price. If you look at market shares now, Free has maybe 15% of the market. You might say “that doesn’t sound right” because they had such a better price, but the true impact was massive in consumer savings and increased competition because they forced the incumbents to get their act together.
I don’t think the big banks are going to lose that much. But they’re going to get their act together and provide better services for cheaper.
At a time when it seems like both Republicans and Democrats are talking about taking more regulatory action against some big tech companies, what is the risk that the net result of this would just be weakening U.S. technology companies relative to Chinese technology companies like Tencent who gain more penetration in the U.S. and emerging markets?
There’s no question we should have a united front across Europe and the U.S. against spying and stealing technology or other IP. But the claims people make are 95% self-serving. The classic one in the U.S. is “you need to let us have monopoly power, because otherwise, we won’t be able to roll out the 5G network.” It’s so bogus. All they need to do is reduce a tiny bit their payout to their payholders like they would do if they had competition.
The other thing that’s also bogus is the idea that by having fat monopolies at home you create firms that are able to compete globally. The evidence is strongly the other way. It’s like the U.S. selection for the Olympics. It’s very hard to qualify. Many good people don’t make the U.S. qualifier. But the result is the U.S. has by far the largest number of gold medals in the Olympics. The argument that the athletes shouldn’t have to compete too hard at home because they’re going to get tired when they get to the Olympics doesn’t work. In economics, it doesn’t work either. The data is very clear that the companies that do the best globally are the ones that had really tough competition at home which forced them to be good and productive.
What is your prediction for U.S. regulation here? Politicians seem to be coming around to many of the critiques you make in the book. Both the Trump administration and the leading Democratic candidates are speaking out against big tech companies over antitrust issues and other regulatory intervention.
It is good, but I’m also suspicious. This momentum is useful. On the other hand, it’s all politics and they are targeting the wrong firms. If the goal is to revive the economy and help the middle class, this is not going to do it. The reason politicians in both parties are going after Google and Facebook is because they couldn’t agree on anything else.
Even if you successfully go after Google and Facebook, you would have extremely small impact on purchasing power of the middle class or lower middle class. People are angry because prices are too high and wages are too low. You have to target the big ticket items: transport, energy, telecom, and healthcare. These are markets where the monopoly rents are outrageous and have a huge impact on people.
If I look at the evolution of prices over the past 20 years — Europe versus the U.S. — I estimate that if the markup did not increase in the U.S., the media household would save $300 each month from lower prices on cell phones, internet providers, transportation, and stuff like that.