Choice architecture makes its way into anti-trust: Google Ruling
The theory of monopoly has to date been firmly rooted in the analog world. Traditionally, a monopoly occurred when there was a single seller or producer and no close substitutes. With the Justice Department’s ruling that Google operates a monopoly in search, we’re in a whole new realm – one in which cognitive biases are now recognized by law.
.
.
Biased decision-making and choice architecture
As I’ve written about before, important discoveries in the behavioral sciences have shown for years that human decision-making bears little resemblance to the super-rational calculations beloved by economic theory. Economic theory posits that we should always make decisions that maximize our own utility, whatever that may be. But we don’t. Humans are subject to a vast number of cognitive biases, many that we are not even aware of.
There’s a biological reason for this – the world is messy and complex. It’s hard for our brains to sift through the information we are barraged with every second. It takes a lot of effort – and when it comes to brain effort, that means energy and calories. According to one study, the brain consumes around 20% of all our metabolic energy. Further, directing that energy to something that is our focus means that other information is going to be neglected. In a calorie-scarce environment, using too many calories can be fatal. We’ve therefore evolved to require as few calories as possible.
The solution is that humans developed heuristics – short cuts that let us move forward with our lives by selectively taking in information upon which we take action. Eric Johnson’s fabulous book The Elements of Choice shows us how “choice architects” – those who design the decision-making environment – fundamentally influence the decisions we make. Control the choice architecture, he finds, and you can often control the resulting decisions.
So, what does this have to do with the decision on Google?
The power of being the default
Defaults – the option first offered to you in a decision-making situation – are incredibly powerful. Study after study has shown that the default option (particularly in a low-consequences situation) is highly likely to be the one chosen. Which brings us to a fascinating new understanding of monopoly than we have historically used.
With search, there are an incredible number of alternatives to Google. Switching to use one of them is easy – just a few clicks and you’re done. Under traditional conceptions of monopoly, the fact that nearly identical substitutes are available, for free, would make any consideration of monopoly power a non-question. But now, choice architecture has entered the calculus, and the world will never be the same.
As the New York Times reported, “The government argued that by paying billions of dollars to be the automatic search engine on consumer devices, Google had denied its competitors the opportunity to build the scale required to compete with its search engine. Instead, Google collected more data about consumers that it used to make its search engine better and more dominant.” In other words, paying to be the default search engine on so many devices is anti-competitive. Google figured this out long before the government did, reportedly paying Apple $20 billion in 2022 alone to remain the default on Apple iPhones.
More broadly, as Newsweek reports, “Google has systematically secured its position by spending tens of billions of dollars on exclusive contracts with smartphone manufacturers and web browser designers, ensuring that its search engine remains the default option for most users. This strategy has stifled/killed/scared away competition and cemented Google's monopoly, making it nearly impossible for other search engines to even dream of gaining a foothold.”
Default power → Pricing power
The second-order effect of being the dominant default search engine is that Google is the only game in town for many advertisers, giving it unparalleled pricing power. As Paresh Dave writes in Wired, “United States District judge Amit Mehta ruled on Monday that Google has unlawfully maintained its dominance in search by using anticompetitive deals to keep rivals from gaining traction. And without fear of pressure from competitors, Google has been able to charge whatever it wants for search ads, he said.”
What I find particularly interesting is that the “ownership” of the default position in search is now being treated the way physical goods are in the analog world. A sticking point for the anti-trust regulators has always been that since Google’s search product is free, it is hard to make the case that it is engaged in price fixing. Classically, a monopolist can “fix” prices because there is no competition. When the product itself is free and consumers can readily choose an alternative, that’s a tough argument to make.
If, however, control over the default is itself a valuable asset, that opens the door to the argument the government is making. “Notwithstanding the option to switch, the default remains the primary search access point,” Mehta said in his ruling. “Roughly 50% of all general search queries in the United States flow through a search access point covered by one of the challenged contracts.” And nearly one third of queries come from Apple devices on which Google is the default.
For its part, Google argues that it has such a dominant market share because its search engine is the best and that people would use it even if it were not the default option – that consumers would seek it out. As Kent Walker, Google’s president of global affairs, said, “This decision recognizes that Google offers the best search engine, but concludes that we shouldn’t be allowed to make it easily available.”
The beginning of the end of the digital advertising model as we know it?
I’ve argued before that the whole programmatic ad model is looking like a bubble to me.
Bubbles occur when the price of something exceeds its fundamental value by a large margin. The programmatic-ad system that is the dominant part of ad spend tells advertisers that “Programmatic advertising is a perfect realm where precision meets automation, and where your ads reach their perfect audience – almost as if by magic.” Google’s search dominance has long supported its claims that it can finely target prospects by combining huge amounts of data on interests, behaviors, demographics and even time of day. And advertisers, to date, believe! Statista suggests that digital ad spending will be around $300 billion this year in the US alone and if current torrid growth trends persist, could be $452 million by 2028.
What bubbles look like
What if the leaps of faith underlying digital ad spending follow the pattern of a bubble? Economist Hyman P. Minsky, in his 1986 book Stabilizing an Unstable Economy observed that bubbles tend to go through five predictable stages: displacement, boom, euphoria, profit-taking, and panic.
Displacement occurs when investors take note of something new and exciting that promises to make a major change occur. This initiates the “hype” period of a strategic inflection point. In the case of programmatic advertising, a series of advances – from algorithmically-informed ad placements to real-time bidding to advanced analytics, led to deep shifts in advertising spending from primarily manual processes of ad buying. By 2022, ads bought programmatically accounted for 84% of global ad spending, according to one source.
In the Boom stage, more and more participants get involved. Fear of missing out sets in. Those who are invested in the bubble look down their noses at their more conservative acquaintances, claiming that they have a lock on the Next Big Thing. In the case of programmatic ad spend, the first major player was DoubleClick who began offering its services in 1996. The company's main product line was known as DART (Dynamic Advertising, Reporting, and Targeting), which was intended to increase the purchasing efficiency of advertisers and minimize unsold inventory for publishers. An explosion of similar companies entered in the late 90’s. Google entered the ad market with Adwords in 2000 and has steadily been expanding its services since then.
Next comes the phase Hyman calls Euphoria. Caution is thrown to the winds. Asset prices climb ever higher and investors convince themselves that no matter how expensive the asset becomes, there will always be an eager buyer ready to take it off their hands (otherwise known as the Greater Fool Theory). Everybody expects the party to go on. In the case of Internet Advertising, predictions are that the market will grow by 10.5% through 2030.
Then, paying attention to the early warning signs, the smart money starts pulling its resources out. This is the profit taking phase. There is plenty of evidence that this is where we might be with respect to programmatic advertising. For instance, the 2020 book by Tim Hwang, The Sub-Prime Attention Crisis lays out the case that advertisers are not getting what they think they are getting out of ads.
JP Morgan Chase went from placing ads on 400,000 sites to using them on just 5,000. The result? No change in their business. The New York Times blocked all programmatic ad placements in Europe after the GDPR and discovered….. their advertising revenue continued to grow. eBay, in a natural experiment, found that when they stopped spending on programmatic ads against the phrase “eBay” the effect was to …. Change absolutely nothing about the traffic on their site.
The trouble with this period is that it is very hard to estimate the timing of a bubble bursting. As economist John Maynard Keynes put it, "the markets can stay irrational longer than you can stay solvent."
Panic sets in when the supply of ready buyers for the inflated asset dries up, sometimes incredibly quickly. Supply overwhelms demand, the prices and values drop precipitously, and people pull back, way back. We’re clearly not there with programmatic Internet advertising, yet.
But the case involving Google does raise some interesting questions. If one of the remedies is to break up the company to reduce its dominance of internet-based advertising, engaging with its proprietary platforms becomes less compelling. If, as some suggest, the remedy will be years in the making, Google will have the time to figure out what it’s next act is likely to be (and the company has some extraordinarily sharp people to make that happen). If, as might be the case in some scenarios, the bubble pops before Google makes that transition, it will dramatically shift how we finance the wonders of the Internet as we know it.
What I don’t think many people appreciate, yet, is the massive change in the concept of what an asset is that this ruling represents. Rather than a tangible, analog, item, we now are being told that controlling critical elements of choice architecture are understood to control critical parts of our economic lives. That’s a really big deal.