Archives For markets

Last week the editorial board of the Washington Post penned an excellent editorial responding to the European Commission’s announcement of its decision in its Google Shopping investigation. Here’s the key language from the editorial:

Whether the demise of any of [the complaining comparison shopping sites] is specifically traceable to Google, however, is not so clear. Also unclear is the aggregate harm from Google’s practices to consumers, as opposed to the unlucky companies. Birkenstock-seekers may well prefer to see a Google-generated list of vendors first, instead of clicking around to other sites…. Those who aren’t happy anyway have other options. Indeed, the rise of comparison shopping on giants such as Amazon and eBay makes concerns that Google might exercise untrammeled power over e-commerce seem, well, a bit dated…. Who knows? In a few years we might be talking about how Facebook leveraged its 2 billion users to disrupt the whole space.

That’s actually a pretty thorough, if succinct, summary of the basic problems with the Commission’s case (based on its PR and Factsheet, at least; it hasn’t released the full decision yet).

I’ll have more to say on the decision in due course, but for now I want to elaborate on two of the points raised by the WaPo editorial board, both in service of its crucial rejoinder to the Commission that “Also unclear is the aggregate harm from Google’s practices to consumers, as opposed to the unlucky companies.”

First, the WaPo editorial board points out that:

Birkenstock-seekers may well prefer to see a Google-generated list of vendors first, instead of clicking around to other sites.

It is undoubtedly true that users “may well prefer to see a Google-generated list of vendors first.” It’s also crucial to understanding the changes in Google’s search results page that have given rise to the current raft of complaints.

As I noted in a Wall Street Journal op-ed two years ago:

It’s a mistake to consider “general search” and “comparison shopping” or “product search” to be distinct markets.

From the moment it was technologically feasible to do so, Google has been adapting its traditional search results—that familiar but long since vanished page of 10 blue links—to offer more specialized answers to users’ queries. Product search, which is what is at issue in the EU complaint, is the next iteration in this trend.

Internet users today seek information from myriad sources: Informational sites (Wikipedia and the Internet Movie Database); review sites (Yelp and TripAdvisor); retail sites (Amazon and eBay); and social-media sites (Facebook and Twitter). What do these sites have in common? They prioritize certain types of data over others to improve the relevance of the information they provide.

“Prioritization” of Google’s own shopping results, however, is the core problem for the Commission:

Google has systematically given prominent placement to its own comparison shopping service: when a consumer enters a query into the Google search engine in relation to which Google’s comparison shopping service wants to show results, these are displayed at or near the top of the search results. (Emphasis in original).

But this sort of prioritization is the norm for all search, social media, e-commerce and similar platforms. And this shouldn’t be a surprise: The value of these platforms to the user is dependent upon their ability to sort the wheat from the chaff of the now immense amount of information coursing about the Web.

As my colleagues and I noted in a paper responding to a methodologically questionable report by Tim Wu and Yelp leveling analogous “search bias” charges in the context of local search results:

Google is a vertically integrated company that offers general search, but also a host of other products…. With its well-developed algorithm and wide range of products, it is hardly surprising that Google can provide not only direct answers to factual questions, but also a wide range of its own products and services that meet users’ needs. If consumers choose Google not randomly, but precisely because they seek to take advantage of the direct answers and other options that Google can provide, then removing the sort of “bias” alleged by [complainants] would affirmatively hurt, not help, these users. (Emphasis added).

And as Josh Wright noted in an earlier paper responding to yet another set of such “search bias” charges (in that case leveled in a similarly methodologically questionable report by Benjamin Edelman and Benjamin Lockwood):

[I]t is critical to recognize that bias alone is not evidence of competitive harm and it must be evaluated in the appropriate antitrust economic context of competition and consumers, rather individual competitors and websites. Edelman & Lockwood´s analysis provides a useful starting point for describing how search engines differ in their referrals to their own content. However, it is not useful from an antitrust policy perspective because it erroneously—and contrary to economic theory and evidence—presumes natural and procompetitive product differentiation in search rankings to be inherently harmful. (Emphasis added).

We’ll have to see what kind of analysis the Commission relies upon in its decision to reach its conclusion that prioritization is an antitrust problem, but there is reason to be skeptical that it will turn out to be compelling. The Commission states in its PR that:

The evidence shows that consumers click far more often on results that are more visible, i.e. the results appearing higher up in Google’s search results. Even on a desktop, the ten highest-ranking generic search results on page 1 together generally receive approximately 95% of all clicks on generic search results (with the top result receiving about 35% of all the clicks). The first result on page 2 of Google’s generic search results receives only about 1% of all clicks. This cannot just be explained by the fact that the first result is more relevant, because evidence also shows that moving the first result to the third rank leads to a reduction in the number of clicks by about 50%. The effects on mobile devices are even more pronounced given the much smaller screen size.

This means that by giving prominent placement only to its own comparison shopping service and by demoting competitors, Google has given its own comparison shopping service a significant advantage compared to rivals. (Emphasis added).

Whatever truth there is in the characterization that placement is more important than relevance in influencing user behavior, the evidence cited by the Commission to demonstrate that doesn’t seem applicable to what’s happening on Google’s search results page now.

Most crucially, the evidence offered by the Commission refers only to how placement affects clicks on “generic search results” and glosses over the fact that the “prominent placement” of Google’s “results” is not only a difference in position but also in the type of result offered.

Google Shopping results (like many of its other “vertical results” and direct answers) are very different than the 10 blue links of old. These “universal search” results are, for one thing, actual answers rather than merely links to other sites. They are also more visually rich and attractively and clearly displayed.

Ironically, Tim Wu and Yelp use the claim that users click less often on Google’s universal search results to support their contention that increased relevance doesn’t explain Google’s prioritization of its own content. Yet, as we note in our response to their study:

[I]f a consumer is using a search engine in order to find a direct answer to a query rather than a link to another site to answer it, click-through would actually represent a decrease in consumer welfare, not an increase.

In fact, the study fails to incorporate this dynamic even though it is precisely what the authors claim the study is measuring.

Further, as the WaPo editorial intimates, these universal search results (including Google Shopping results) are quite plausibly more valuable to users. As even Tim Wu and Yelp note:

No one truly disagrees that universal search, in concept, can be an important innovation that can serve consumers.

Google sees it exactly this way, of course. Here’s Tim Wu and Yelp again:

According to Google, a principal difference between the earlier cases and its current conduct is that universal search represents a pro-competitive, user-serving innovation. By deploying universal search, Google argues, it has made search better. As Eric Schmidt argues, “if we know the answer it is better for us to answer that question so [the user] doesn’t have to click anywhere, and in that sense we… use data sources that are our own because we can’t engineer it any other way.”

Of course, in this case, one would expect fewer clicks to correlate with higher value to users — precisely the opposite of the claim made by Tim Wu and Yelp, which is the surest sign that their study is faulty.

But the Commission, at least according to the evidence cited in its PR, doesn’t even seem to measure the relative value of the very different presentations of information at all, instead resting on assertions rooted in the irrelevant difference in user propensity to click on generic (10 blue links) search results depending on placement.

Add to this Pinar Akman’s important point that Google Shopping “results” aren’t necessarily search results at all, but paid advertising:

[O]nce one appreciates the fact that Google’s shopping results are simply ads for products and Google treats all ads with the same ad-relevant algorithm and all organic results with the same organic-relevant algorithm, the Commission’s order becomes impossible to comprehend. Is the Commission imposing on Google a duty to treat non-sponsored results in the same way that it treats sponsored results? If so, does this not provide an unfair advantage to comparison shopping sites over, for example, Google’s advertising partners as well as over Amazon, eBay, various retailers, etc…?

Randy Picker also picks up on this point:

But those Google shopping boxes are ads, Picker told me. “I can’t imagine what they’re thinking,” he said. “Google is in the advertising business. That’s how it makes its money. It has no obligation to put other people’s ads on its website.”

The bottom line here is that the WaPo editorial board does a better job characterizing the actual, relevant market dynamics in a single sentence than the Commission seems to have done in its lengthy releases summarizing its decision following seven full years of investigation.

The second point made by the WaPo editorial board to which I want to draw attention is equally important:

Those who aren’t happy anyway have other options. Indeed, the rise of comparison shopping on giants such as Amazon and eBay makes concerns that Google might exercise untrammeled power over e-commerce seem, well, a bit dated…. Who knows? In a few years we might be talking about how Facebook leveraged its 2 billion users to disrupt the whole space.

The Commission dismisses this argument in its Factsheet:

The Commission Decision concerns the effect of Google’s practices on comparison shopping markets. These offer a different service to merchant platforms, such as Amazon and eBay. Comparison shopping services offer a tool for consumers to compare products and prices online and find deals from online retailers of all types. By contrast, they do not offer the possibility for products to be bought on their site, which is precisely the aim of merchant platforms. Google’s own commercial behaviour reflects these differences – merchant platforms are eligible to appear in Google Shopping whereas rival comparison shopping services are not.

But the reality is that “comparison shopping,” just like “general search,” is just one technology among many for serving information and ads to consumers online. Defining the relevant market or limiting the definition of competition in terms of the particular mechanism that Google (or Foundem, or Amazon, or Facebook…) happens to use doesn’t reflect the extent of substitutability between these different mechanisms.

Properly defined, the market in which Google competes online is not search, but something more like online “matchmaking” between advertisers, retailers and consumers. And this market is enormously competitive. The same goes for comparison shopping.

And the fact that Amazon and eBay “offer the possibility for products to be bought on their site” doesn’t take away from the fact that they also “offer a tool for consumers to compare products and prices online and find deals from online retailers of all types.” Not only do these sites contain enormous amounts of valuable (and well-presented) information about products, including product comparisons and consumer reviews, but they also actually offer comparisons among retailers. In fact, Fifty percent of the items sold through Amazon’s platform, for example, are sold by third-party retailers — the same sort of retailers that might also show up on a comparison shopping site.

More importantly, though, as the WaPo editorial rightly notes, “[t]hose who aren’t happy anyway have other options.” Google just isn’t the indispensable gateway to the Internet (and definitely not to shopping on the Internet) that the Commission seems to think.

Today over half of product searches in the US start on Amazon. The majority of web page referrals come from Facebook. Yelp’s most engaged users now access it via its app (which has seen more than 3x growth in the past five years). And a staggering 40 percent of mobile browsing on both Android and iOS now takes place inside the Facebook app.

Then there are “closed” platforms like the iTunes store and innumerable other apps that handle copious search traffic (including shopping-related traffic) but also don’t figure in the Commission’s analysis, apparently.

In fact, billions of users reach millions of companies every day through direct browser navigation, social media, apps, email links, review sites, blogs, and countless other means — all without once touching Google.com. So-called “dark social” interactions (email, text messages, and IMs) drive huge amounts of some of the most valuable traffic on the Internet, in fact.

All of this, in turn, has led to a competitive scramble to roll out completely new technologies to meet consumers’ informational (and merchants’ advertising) needs. The already-arriving swarm of VR, chatbots, digital assistants, smart-home devices, and more will offer even more interfaces besides Google through which consumers can reach their favorite online destinations.

The point is this: Google’s competitors complaining that the world is evolving around them don’t need to rely on Google. That they may choose to do so does not saddle Google with an obligation to ensure that they can always do so.

Antitrust laws — in Europe, no less than in the US — don’t require Google or any other firm to make life easier for competitors. That’s especially true when doing so would come at the cost of consumer-welfare-enhancing innovations. The Commission doesn’t seem to have grasped this fundamental point, however.

The WaPo editorial board gets it, though:

The immense size and power of all Internet giants are a legitimate focus for the antitrust authorities on both sides of the Atlantic. Brussels vs. Google, however, seems to be a case of punishment without crime.

On July 1, the minimum wage will spike in several cities and states across the country. Portland, Oregon’s minimum wage will rise by $1.50 to $11.25 an hour. Los Angeles will also hike its minimum wage by $1.50 to $12 an hour. Recent research shows that these hikes will make low wage workers poorer.

A study supported and funded in part by the Seattle city government, was released this week, along with an NBER paper evaluating Seattle’s minimum wage increase to $13 an hour. The papers find that the increase to $13 an hour had significant negative impacts on employment and led to lower incomes for minimum wage workers.

The study is the first study of a very high minimum wage for a city. During the study period, Seattle’s minimum wage increased from what had been the nation’s highest state minimum wage to an even higher level. It is also unique in its use of administrative data that has much more detail than is usually available to economics researchers.

Conclusions from the research focusing on Seattle’s increase to $13 an hour are clear: The policy harms those it was designed to help.

  • A loss of more than 5,000 jobs and a 9 percent reduction in hours worked by those who retained their jobs.
  • Low-wage workers lost an average of $125 per month. The minimum wage has always been a terrible way to reduce poverty. In 2015 and 2016, I presented analysis to the Oregon Legislature indicating that incomes would decline with a steep increase in the minimum wage. The Seattle study provides evidence backing up that forecast.
  • Minimum wage supporters point to research from the 1990s that made headlines with its claims that minimum wage increases had no impact on restaurant employment. The authors of the Seattle study were able to replicate the results of these papers by using their own data and imposing the same limitations that the earlier researchers had faced. The Seattle study shows that those earlier papers’ findings were likely driven by their approach and data limitations. This is a big deal, and a novel research approach that gives strength to the Seattle study’s results.

Some inside baseball.

The Seattle Minimum Wage Study was supported and funded in part by the Seattle city government. It’s rare that policy makers go through any effort to measure the effectiveness of their policies, so Seattle should get some points for transparency.

Or not so transparent: The mayor of Seattle commissioned another study, by an advocacy group at Berkeley whose previous work on the minimum wage is uniformly in favor of hiking the minimum wage (they testified before the Oregon Legislature to cheerlead the state’s minimum wage increase). It should come as no surprise that the Berkeley group released its report several days before the city’s “official” study came out.

You might think to yourself, “OK, that’s Seattle. Seattle is different.”

But, maybe Seattle is not that different. In fact, maybe the negative impacts of high minimum wages are universal, as seen in another study that came out this week, this time from Denmark.

In Denmark the minimum wage jumps up by 40 percent when a worker turns 18. The Danish researchers found that this steep increase was associated with employment dropping by one-third, as seen in the chart below from the paper.

3564_KREINER-Fig1

Let’s look at what’s going to happen in Oregon. The state’s employment department estimates that about 301,000 jobs will be affected by the rate increase. With employment of almost 1.8 million, that means one in six workers will be affected by the steep hikes going into effect on July 1. That’s a big piece of the work force. By way of comparison, in the past when the minimum wage would increase by five or ten cents a year, only about six percent of the workforce was affected.

This is going to disproportionately affect youth employment. As noted in my testimony to the legislature, unemployment for Oregonians age 16 to 19 is 8.5 percentage points higher than the national average. This was not always the case. In the early 1990s, Oregon’s youth had roughly the same rate of unemployment as the U.S. as a whole. Then, as Oregon’s minimum wage rose relative to the federal minimum wage, Oregon’s youth unemployment worsened. Just this week, Multnomah County made a desperate plea for businesses to hire more youth as summer interns.

It has been suggested Oregon youth have traded education for work experience—in essence, they have opted to stay in high school or enroll in higher education instead of entering the workforce. The figure below shows, however, that youth unemployment has increased for both those enrolled in school and those who are not enrolled in school. The figure debunks the notion that education and employment are substitutes. In fact, the large number of students seeking work demonstrates many youth want employment while they further their education.

OregonYouthUnemployment

None of these results should be surprising. Minimum wage research is more than a hundred years old. Aside from the “mans bites dog” research from the 1990s, economists were broadly in agreement that higher minimum wages would be associated with reduced employment, especially among youth. The research published this week is groundbreaking in its data and methodology. At the same time, the results are unsurprising to anyone with any understanding of economics or experience running a business.

Regardless of the merits and soundness (or lack thereof) of this week’s European Commission Decision in the Google Shopping case — one cannot assess this until we have the text of the decision — two comments really struck me during the press conference.

First, it was said that Google’s conduct had essentially reduced innovation. If I heard correctly, this is a formidable statement. In 2016, another official EU service published stats that described Alphabet as increasing its R&D by 22% and ranked it as the world’s 4th top R&D investor. Sure it can always be better. And sure this does not excuse everything. But still. The press conference language on incentives to innovate was a bit of an oversell, to say the least.

Second, the Commission views this decision as a “precedent” or as a “framework” that will inform the way dominant Internet platforms should display, intermediate and market their services and those of their competitors. This may fuel additional complaints by other vertical search rivals against (i) Google in relation to other product lines, but also against (ii) other large platform players.

Beyond this, the Commission’s approach raises a gazillion questions of law and economics. Pending the disclosure of the economic evidence in the published decision, let me share some thoughts on a few (arbitrarily) selected legal issues.

First, the Commission has drawn the lesson of the Microsoft remedy quagmire. The Commission refrains from using a trustee to ensure compliance with the decision. This had been a bone of contention in the 2007 Microsoft appeal. Readers will recall that the Commission had imposed on Microsoft to appoint a monitoring trustee, who was supposed to advise on possible infringements in the implementation of the decision. On appeal, the Court eventually held that the Commission was solely responsible for this, and could not delegate those powers. Sure, the Commission could “retai[n] its own external expert to provide advice when it investigates the implementation of the remedies.” But no more than that.

Second, we learn that the Commission is no longer in the business of software design. Recall the failed untying of WMP and Windows — Windows Naked sold only 11,787 copies, likely bought by tech bootleggers willing to acquire the first piece of software ever designed by antitrust officials — or the browser “Choice Screen” compliance saga which eventually culminated with a €561 million fine. Nothing of this can be found here. The Commission leaves remedial design to the abstract concept of “equal treatment”.[1] This, certainly, is a (relatively) commendable approach, and one that could inspire remedies in other unilateral conduct cases, in particular, exploitative conduct ones where pricing remedies are both costly, impractical, and consequentially inefficient.

On the other hand, readers will also not fail to see the corollary implication of “equal treatment”: search neutrality could actually cut both ways, and lead to a lawful degradation in consumer welfare if Google were ever to decide to abandon rich format displays for both its own shopping services and those of rivals.

Third, neither big data nor algorithmic design is directly vilified in the case (“The Commission Decision does not object to the design of Google’s generic search algorithms or to demotions as such, nor to the way that Google displays or organises its search results pages”). In fact, the Commission objects to the selective application of Google’s generic search algorithms to its own products. This is an interesting, and subtle, clarification given all the coverage that this topic has attracted in recent antitrust literature. We are in fact very close to a run of the mill claim of disguised market manipulation, not causally related to data or algorithmic technology.

Fourth, Google said it contemplated a possible appeal of the decision. Now, here’s a challenging question: can an antitrust defendant effectively exercise its right to judicial review of an administrative agency (and more generally its rights of defense), when it operates under the threat of antitrust sanctions in ongoing parallel cases investigated by the same agency (i.e., the antitrust inquiries related to Android and Ads)? This question cuts further than the Google Shopping case. Say firm A contemplates a merger with firm B in market X, while it is at the same time subject to antitrust investigations in market Z. And assume that X and Z are neither substitutes nor complements so there is little competitive relationship between both products. Can the Commission leverage ongoing antitrust investigations in market Z to extract merger concessions in market X? Perhaps more to the point, can the firm interact with the Commission as if the investigations are completely distinct, or does it have to play a more nuanced game and consider the ramifications of its interactions with the Commission in both markets?

Fifth, as to the odds of a possible appeal, I don’t believe that arguments on the economic evidence or legal theory of liability will ever be successful before the General Court of the EU. The law and doctrine in unilateral conduct cases are disturbingly — and almost irrationally — severe. As I have noted elsewhere, the bottom line in the EU case-law on unilateral conduct is to consider the genuine requirement of “harm to competition” as a rhetorical question, not an empirical one. In EU unilateral conduct law, exclusion of every and any firm is a per se concern, regardless of evidence of efficiency, entry or rivalry.

In turn, I tend to opine that Google has a stronger game from a procedural standpoint, having been left with (i) the expectation of a settlement (it played ball three times by making proposals); (ii) a corollary expectation of the absence of a fine (settlement discussions are not appropriate for cases that could end with fines); and (iii) a full seven long years of an investigatory cloud. We know from the past that EU judges like procedural issues, but like comparably less to debate the substance of the law in unilateral conduct cases. This case could thus be a test case in terms of setting boundaries on how freely the Commission can U-turn a case (the Commissioner said “take the case forward in a different way”).

Today, the Senate Committee on Health, Education, Labor, and Pensions (HELP) enters the drug pricing debate with a hearing on “The Cost of Prescription Drugs: How the Drug Delivery System Affects What Patients Pay.”  By questioning the role of the drug delivery system in pricing, the hearing goes beyond the more narrow focus of recent hearings that have explored how drug companies set prices.  Instead, today’s hearing will explore how pharmacy benefit managers, insurers, providers, and others influence the amounts that patients pay.

In 2016, net U.S. drug spending increased by 4.8% to $323 billion (after adjusting for rebates and off-invoice discounts).  This rate of growth slowed to less than half the rates of 2014 and 2015, when net drug spending grew at rates of 10% and 8.9% respectively.  Yet despite the slowing in drug spending, the public outcry over the cost of prescription drugs continues.

In today’s hearing, there will be testimony both on the various causes of drug spending increases and on various proposals that could reduce the cost of drugs.  Several of the proposals will focus on ways to increase competition in the pharmaceutical industry, and in turn, reduce drug prices.  I have previously explained several ways that the government could reduce prices through enhanced competition, including reducing the backlog of generic drugs awaiting FDA approval and expediting the approval and acceptance of biosimilars.  Other proposals today will likely call for regulatory reforms to enable innovative contractual arrangements that allow for outcome- or indication-based pricing and other novel reimbursement designs.

However, some proposals will undoubtedly return to the familiar call for more government negotiation of drug prices, especially drugs covered under Medicare Part D.  As I’ve discussed in a previous post, in order for government negotiation to significantly lower drug prices, the government must be able to put pressure on drug makers to secure price concessions. This could be achieved if the government could set prices administratively, penalize manufacturers that don’t offer price reductions, or establish a formulary.  Setting prices or penalizing drug makers that don’t reduce prices would produce the same disastrous effects as price controls: drug shortages in certain markets, increased prices for non-Medicare patients, and reduced incentives for innovation. A government formulary for Medicare Part D coverage would provide leverage to obtain discounts from manufacturers, but it would mean that many patients could no longer access some of their optimal drugs.

As lawmakers seriously consider changes that would produce these negative consequences, industry would do well to voluntarily constrain prices.  Indeed, in the last year, many drug makers have pledged to limit price increases to keep drug spending under control.  Allergan was first, with its “social contract” introduced last September that promised to keep price increases below 10 percent. Since then, Novo Nordisk, AbbVie, and Takeda, have also voluntarily committed to single-digit price increases.

So far, the evidence shows the drug makers are sticking to their promises. Allergan has raised the price of U.S. branded products by an average of 6.7% in 2017, and no drug’s list price has increased by more than single digits.  In contrast, Pfizer, who has made no pricing commitment, has raised the price of many of its drugs by 20%.

If more drug makers brought about meaningful change by committing to voluntary pricing restraints, the industry could prevent the market-distorting consequences of government intervention while helping patients afford the drugs they need.   Moreover, avoiding intrusive government mandates and price controls would preserve drug innovation that has brought life-saving and life-enhancing drugs to millions of Americans.

 

 

 

It’s fitting that FCC Chairman Ajit Pai recently compared his predecessor’s jettisoning of the FCC’s light touch framework for Internet access regulation without hard evidence to the Oklahoma City Thunder’s James Harden trade. That infamous deal broke up a young nucleus of three of the best players in the NBA in 2012 because keeping all three might someday create salary cap concerns. What few saw coming was a new TV deal in 2015 that sent the salary cap soaring.

If it’s hard to predict how the market will evolve in the closed world of professional basketball, predictions about the path of Internet innovation are an order of magnitude harder — especially for those making crucial decisions with a lot of money at stake.

The FCC’s answer for what it considered to be the dangerous unpredictability of Internet innovation was to write itself a blank check of authority to regulate ISPs in the 2015 Open Internet Order (OIO), embodied in what is referred to as the “Internet conduct standard.” This standard expanded the scope of Internet access regulation well beyond the core principle of preserving openness (i.e., ensuring that any legal content can be accessed by all users) by granting the FCC the unbounded, discretionary authority to define and address “new and novel threats to the Internet.”

When asked about what the standard meant (not long after writing it), former Chairman Tom Wheeler replied,

We don’t really know. We don’t know where things will go next. We have created a playing field where there are known rules, and the FCC will sit there as a referee and will throw the flag.

Somehow, former Chairman Wheeler would have us believe that an amorphous standard that means whatever the agency (or its Enforcement Bureau) says it means created a playing field with “known rules.” But claiming such broad authority is hardly the light-touch approach marketed to the public. Instead, this ill-conceived standard allows the FCC to wade as deeply as it chooses into how an ISP organizes its business and how it manages its network traffic.

Such an approach is destined to undermine, rather than further, the objectives of Internet openness, as embodied in Chairman Powell’s 2005 Internet Policy Statement:

To foster creation, adoption and use of Internet broadband content, applications, services and attachments, and to ensure consumers benefit from the innovation that comes from competition.

Instead, the Internet conduct standard is emblematic of how an off-the-rails quest to heavily regulate one specific component of the complex Internet ecosystem results in arbitrary regulatory imbalances — e.g., between ISPs and over-the-top (OTT) or edge providers that offer similar services such as video streaming or voice calling.

As Boston College Law Professor, Dan Lyons, puts it:

While many might assume that, in theory, what’s good for Netflix is good for consumers, the reality is more complex. To protect innovation at the edge of the Internet ecosystem, the Commission’s sweeping rules reduce the opportunity for consumer-friendly innovation elsewhere, namely by facilities-based broadband providers.

This is no recipe for innovation, nor does it coherently distinguish between practices that might impede competition and innovation on the Internet and those that are merely politically disfavored, for any reason or no reason at all.

Free data madness

The Internet conduct standard’s unholy combination of unfettered discretion and the impulse to micromanage can (and will) be deployed without credible justification to the detriment of consumers and innovation. Nowhere has this been more evident than in the confusion surrounding the regulation of “free data.”

Free data, like T-Mobile’s Binge On program, is data consumed by a user that has been subsidized by a mobile operator or a content provider. The vertical arrangements between operators and content providers creating the free data offerings provide many benefits to consumers, including enabling subscribers to consume more data (or, for low-income users, to consume data in the first place), facilitating product differentiation by mobile operators that offer a variety of free data plans (including allowing smaller operators the chance to get a leg up on competitors by assembling a market-share-winning plan), increasing the overall consumption of content, and reducing users’ cost of obtaining information. It’s also fundamentally about experimentation. As the International Center for Law & Economics (ICLE) recently explained:

Offering some services at subsidized or zero prices frees up resources (and, where applicable, data under a user’s data cap) enabling users to experiment with new, less-familiar alternatives. Where a user might not find it worthwhile to spend his marginal dollar on an unfamiliar or less-preferred service, differentiated pricing loosens the user’s budget constraint, and may make him more, not less, likely to use alternative services.

In December 2015 then-Chairman Tom Wheeler used his newfound discretion to launch a 13-month “inquiry” into free data practices before preliminarily finding some to be in violation of the standard. Without identifying any actual harm, Wheeler concluded that free data plans “may raise” economic and public policy issues that “may harm consumers and competition.”

After assuming the reins at the FCC, Chairman Pai swiftly put an end to that nonsense, saying that the Commission had better things to do (like removing barriers to broadband deployment) than denying free data plans that expand Internet access and are immensely popular, especially among low-income Americans.

The global morass of free data regulation

But as long as the Internet conduct standard remains on the books, it implicitly grants the US’s imprimatur to harmful policies and regulatory capriciousness in other countries that look to the US for persuasive authority. While Chairman Pai’s decisive intervention resolved the free data debate in the US (at least for now), other countries are still grappling with whether to prohibit the practice, allow it, or allow it with various restrictions.

In Europe, the 2016 EC guidelines left the decision of whether to allow the practice in the hands of national regulators. Consequently, some regulators — in Hungary, Sweden, and the Netherlands (although there the ban was recently overturned in court) — have banned free data practices  while others — in Denmark, Germany, Spain, Poland, the United Kingdom, and Ukraine — have not. And whether or not they allow the practice, regulators (e.g., Norway’s Nkom and the UK’s Ofcom) have lamented the lack of regulatory certainty surrounding free data programs, a state of affairs that is compounded by a lack of data on the consequences of various approaches to their regulation.

In Canada this year, the CRTC issued a decision adopting restrictive criteria under which to evaluate free data plans. The criteria include assessing the degree to which the treatment of data is agnostic, whether the free data offer is exclusive to certain customers or certain content providers, the impact on Internet openness and innovation, and whether there is financial compensation involved. The standard is open-ended, and free data plans as they are offered in the US would “likely raise concerns.”

Other regulators are contributing to the confusion through ambiguously framed rules, such as that of the Chilean regulator, Subtel. In a 2014 decision, it found that a free data offer of specific social network apps was in breach of Chile’s Internet rules. In contrast to what is commonly reported, however, Subtel did not ban free data. Instead, it required mobile operators to change how they promote such services, requiring them to state that access to Facebook, Twitter and WhatsApp were offered “without discounting the user’s balance” instead of “at no cost.” It also required them to disclose the amount of time the offer would be available, but imposed no mandatory limit.

In addition to this confusing regulatory make-work governing how operators market free data plans, the Chilean measures also require that mobile operators offer free data to subscribers who pay for a data plan, in order to ensure free data isn’t the only option users have to access the Internet.

The result is that in Chile today free data plans are widely offered by Movistar, Claro, and Entel and include access to apps such as Facebook, WhatsApp, Twitter, Instagram, Pokemon Go, Waze, Snapchat, Apple Music, Spotify, Netflix or YouTube — even though Subtel has nominally declared such plans to be in violation of Chile’s net neutrality rules.

Other regulators are searching for palatable alternatives to both flex their regulatory muscle to govern Internet access, while simultaneously making free data work. The Indian regulator, TRAI, famously banned free data in February 2016. But the story doesn’t end there. After seeing the potential value of free data in unserved and underserved, low-income areas, TRAI proposed implementing government-sanctioned free data. The proposed scheme would provide rural subscribers with 100 MB of free data per month, funded through the country’s universal service fund. To ensure that there would be no vertical agreements between content providers and mobile operators, TRAI recommended introducing third parties, referred to as “aggregators,” that would facilitate mobile-operator-agnostic arrangements.

The result is a nonsensical, if vaguely well-intentioned, threading of the needle between the perceived need to (over-)regulate access providers and the determination to expand access. Notwithstanding the Indian government’s awareness that free data will help to close the digital divide and enhance Internet access, in other words, it nonetheless banned private markets from employing private capital to achieve that very result, preferring instead non-market processes which are unlikely to be nearly as nimble or as effective — and yet still ultimately offer “non-neutral” options for consumers.

Thinking globally, acting locally (by ditching the Internet conduct standard)

Where it is permitted, free data is undergoing explosive adoption among mobile operators. Currently in the US, for example, all major mobile operators offer some form of free data or unlimited plan to subscribers. And, as a result, free data is proving itself as a business model for users’ early stage experimentation and adoption of augmented reality, virtual reality and other cutting-edge technologies that represent the Internet’s next wave — but that also use vast amounts of data. Were the US to cut off free data at the legs under the OIO absent hard evidence of harm, it would substantially undermine this innovation.

The application of the nebulous Internet conduct standard to free data is a microcosm of the current incoherence: It is a rule rife with a parade of uncertainties and only theoretical problems, needlessly saddling companies with enforcement risk, all in the name of preserving and promoting innovation and openness. As even some of the staunchest proponents of net neutrality have recognized, only companies that can afford years of litigation can be expected to thrive in such an environment.

In the face of confusion and uncertainty globally, the US is now poised to provide leadership grounded in sound policy that promotes innovation. As ICLE noted last month, Chairman Pai took a crucial step toward re-imposing economic rigor and the rule of law at the FCC by questioning the unprecedented and ill-supported expansion of FCC authority that undergirds the OIO in general and the Internet conduct standard in particular. Today the agency will take the next step by voting on Chairman Pai’s proposed rulemaking. Wherever the new proceeding leads, it’s a welcome opportunity to analyze the issues with a degree of rigor that has thus far been appallingly absent.

And we should not forget that there’s a direct solution to these ambiguities that would avoid the undulations of subsequent FCC policy fights: Congress could (and should) pass legislation implementing a regulatory framework grounded in sound economics and empirical evidence that allows for consumers to benefit from the vast number of procompetitive vertical agreements (such as free data plans), while still facilitating a means for policing conduct that may actually harm consumers.

The Golden State Warriors are the heavy odds-on favorite to win another NBA Championship this summer, led by former OKC player Kevin Durant. And James Harden is a contender for league MVP. We can’t always turn back the clock on a terrible decision, hastily made before enough evidence has been gathered, but Chairman Pai’s efforts present a rare opportunity to do so.

Today the International Center for Law & Economics (ICLE) Antitrust and Consumer Protection Research Program released a new white paper by Geoffrey A. Manne and Allen Gibby entitled:

A Brief Assessment of the Procompetitive Effects of Organizational Restructuring in the Ag-Biotech Industry

Over the past two decades, rapid technological innovation has transformed the industrial organization of the ag-biotech industry. These developments have contributed to an impressive increase in crop yields, a dramatic reduction in chemical pesticide use, and a substantial increase in farm profitability.

One of the most striking characteristics of this organizational shift has been a steady increase in consolidation. The recent announcements of mergers between Dow and DuPont, ChemChina and Syngenta, and Bayer and Monsanto suggest that these trends are continuing in response to new market conditions and a marked uptick in scientific and technological advances.

Regulators and industry watchers are often concerned that increased consolidation will lead to reduced innovation, and a greater incentive and ability for the largest firms to foreclose competition and raise prices. But ICLE’s examination of the underlying competitive dynamics in the ag-biotech industry suggests that such concerns are likely unfounded.

In fact, R&D spending within the seeds and traits industry increased nearly 773% between 1995 and 2015 (from roughly $507 million to $4.4 billion), while the combined market share of the six largest companies in the segment increased by more than 550% (from about 10% to over 65%) during the same period.

Firms today are consolidating in order to innovate and remain competitive in an industry replete with new entrants and rapidly evolving technological and scientific developments.

According to ICLE’s analysis, critics have unduly focused on the potential harms from increased integration, without properly accounting for the potential procompetitive effects. Our brief white paper highlights these benefits and suggests that a more nuanced and restrained approach to enforcement is warranted.

Our analysis suggests that, as in past periods of consolidation, the industry is well positioned to see an increase in innovation as these new firms unite complementary expertise to pursue more efficient and effective research and development. They should also be better able to help finance, integrate, and coordinate development of the latest scientific and technological developments — particularly in rapidly growing, data-driven “digital farming” —  throughout the industry.

Download the paper here.

And for more on the topic, revisit TOTM’s recent blog symposium, “Agricultural and Biotech Mergers: Implications for Antitrust Law and Economics in Innovative Industries,” here.

On Thursday, March 30, Friday March 31, and Monday April 3, Truth on the Market and the International Center for Law and Economics presented a blog symposium — Agricultural and Biotech Mergers: Implications for Antitrust Law and Economics in Innovative Industries — discussing three proposed agricultural/biotech industry mergers awaiting judgment by antitrust authorities around the globe. These proposed mergers — Bayer/Monsanto, Dow/DuPont and ChemChina/Syngenta — present a host of fascinating issues, many of which go to the core of merger enforcement in innovative industries — and antitrust law and economics more broadly.

The big issue for the symposium participants was innovation (as it was for the European Commission, which cleared the Dow/DuPont merger last week, subject to conditions, one of which related to the firms’ R&D activities).

Critics of the mergers, as currently proposed, asserted that the increased concentration arising from the “Big 6” Ag-biotech firms consolidating into the Big 4 could reduce innovation competition by (1) eliminating parallel paths of research and development (Moss); (2) creating highly integrated technology/traits/seeds/chemicals platforms that erect barriers to new entry platforms (Moss); (3) exploiting eventual network effects that may result from the shift towards data-driven agriculture to block new entry in input markets (Lianos); or (4) increasing incentives to refuse to license, impose discriminatory restrictions in technology licensing agreements, or tacitly “agree” not to compete (Moss).

Rather than fixating on horizontal market share, proponents of the mergers argued that innovative industries are often marked by disruptions and that investment in innovation is an important signal of competition (Manne). An evaluation of the overall level of innovation should include not only the additional economies of scale and scope of the merged firms, but also advancements made by more nimble, less risk-averse biotech companies and smaller firms, whose innovations the larger firms can incentivize through licensing or M&A (Shepherd). In fact, increased efficiency created by economies of scale and scope can make funds available to source innovation outside of the large firms (Shepherd).

In addition, innovation analysis must also account for the intricately interwoven nature of agricultural technology across seeds and traits, crop protection, and, now, digital farming (Sykuta). Combined product portfolios generate more data to analyze, resulting in increased data-driven value for farmers and more efficiently targeted R&D resources (Sykuta).

While critics voiced concerns over such platforms erecting barriers to entry, markets are contestable to the extent that incumbents are incentivized to compete (Russell). It is worth noting that certain industries with high barriers to entry or exit, significant sunk costs, and significant costs disadvantages for new entrants (including automobiles, wireless service, and cable networks) have seen their prices decrease substantially relative to inflation over the last 20 years — even as concentration has increased (Russell). Not coincidentally, product innovation in these industries, as in ag-biotech, has been high.

Ultimately, assessing the likely effects of each merger using static measures of market structure is arguably unreliable or irrelevant in dynamic markets with high levels of innovation (Manne).

Regarding patents, critics were skeptical that combining the patent portfolios of the merging companies would offer benefits beyond those arising from cross-licensing, and would serve to raise rivals’ costs (Ghosh). While this may be true in some cases, IP rights are probabilistic, especially in dynamic markets, as Nicolas Petit noted:

There is no certainty that R&D investments will lead to commercially successful applications; (ii) no guarantee that IP rights will resist to invalidity proceedings in court; (iii) little safety to competition by other product applications which do not practice the IP but provide substitute functionality; and (iv) no inevitability that the environmental, toxicological and regulatory authorization rights that (often) accompany IP rights will not be cancelled when legal requirements change.

In spite of these uncertainties, deals such as the pending ag-biotech mergers provide managers the opportunity to evaluate and reorganize assets to maximize innovation and return on investment in such a way that would not be possible absent a merger (Sykuta). Neither party would fully place its IP and innovation pipeline on the table otherwise.

For a complete rundown of the arguments both for and against, the full archive of symposium posts from our outstanding and diverse group of scholars, practitioners and other experts is available at this link, and individual posts can be easily accessed by clicking on the authors’ names below.

We’d like to thank all of the participants for their excellent contributions!

In a recent long-form article in the New York Times, reporter Noam Scheiber set out to detail some of the ways Uber (and similar companies, but mainly Uber) are engaged in “an extraordinary experiment in behavioral science to subtly entice an independent work force to maximize its growth.”

That characterization seems innocuous enough, but it is apparent early on that Scheiber’s aim is not only to inform but also, if not primarily, to deride these efforts. The title of the piece, in fact, sets the tone:

How Uber Uses Psychological Tricks to Push Its Drivers’ Buttons

Uber and its relationship with its drivers are variously described by Scheiber in the piece as secretive, coercive, manipulative, dominating, and exploitative, among other things. As Schreiber describes his article, it sets out to reveal how

even as Uber talks up its determination to treat drivers more humanely, it is engaged in an extraordinary behind-the-scenes experiment in behavioral science to manipulate them in the service of its corporate growth — an effort whose dimensions became evident in interviews with several dozen current and former Uber officials, drivers and social scientists, as well as a review of behavioral research.

What’s so galling about the piece is that, if you strip away the biased and frequently misguided framing, it presents a truly engaging picture of some of the ways that Uber sets about solving a massively complex optimization problem, abetted by significant agency costs.

So I did. Strip away the detritus, add essential (but omitted) context, and edit the article to fix the anti-Uber bias, the one-sided presentation, the mischaracterizations, and the fundamentally non-economic presentation of what is, at its core, a fascinating illustration of some basic problems (and solutions) from industrial organization economics. (For what it’s worth, Scheiber should know better. After all, “He holds a master’s degree in economics from the University of Oxford, where he was a Rhodes Scholar, and undergraduate degrees in math and economics from Tulane University.”)

In my retelling, the title becomes:

How Uber Uses Innovative Management Tactics to Incentivize Its Drivers

My transformed version of the piece, with critical commentary in the form of tracked changes to the original, is here (pdf).

It’s a long (and, as I said, fundamentally interesting) piece, with cool interactive graphics, well worth the read (well, at least in my retelling, IMHO). Below is just a taste of the edits and commentary I added.

For example, where Scheiber writes:

Uber exists in a kind of legal and ethical purgatory, however. Because its drivers are independent contractors, they lack most of the protections associated with employment. By mastering their workers’ mental circuitry, Uber and the like may be taking the economy back toward a pre-New Deal era when businesses had enormous power over workers and few checks on their ability to exploit it.

With my commentary (here integrated into final form rather than tracked), that paragraph becomes:

Uber operates under a different set of legal constraints, however, also duly enacted and under which millions of workers have profitably worked for decades. Because its drivers are independent contractors, they receive their compensation largely in dollars rather than government-mandated “benefits” that remove some of the voluntariness from employer/worker relationships. And, in the case of overtime pay, for example, the Uber business model that is built in part on offering flexible incentives to match supply and demand using prices and compensation, would be next to impossible. It is precisely through appealing to drivers’ self-interest that Uber and the like may be moving the economy forward to a new era when businesses and workers have more flexibility, much to the benefit of all.

Elsewhere, Scheiber’s bias is a bit more subtle, but no less real. Thus, he writes:

As he tried to log off at 7:13 a.m. on New Year’s Day last year, Josh Streeter, then an Uber driver in the Tampa, Fla., area, received a message on the company’s driver app with the headline “Make it to $330.” The text then explained: “You’re $10 away from making $330 in net earnings. Are you sure you want to go offline?” Below were two prompts: “Go offline” and “Keep driving.” The latter was already highlighted.

With my edits and commentary, that paragraph becomes:

As he started the process of logging off at 7:13 a.m. on New Year’s Day last year, Josh Streeter, then an Uber driver in the Tampa, Fla., area, received a message on the company’s driver app with the headline “Make it to $330.” The text then explained: “You’re $10 away from making $330 in net earnings. Are you sure you want to go offline?” Below were two prompts: “Go offline” and “Keep driving.” The latter was already highlighted, but the former was listed first. It’s anyone’s guess whether either characteristic — placement or coloring — had any effect on drivers’ likelihood of clicking one button or the other.

And one last example. Scheiber writes:

Consider an algorithm called forward dispatch — Lyft has a similar one — that dispatches a new ride to a driver before the current one ends. Forward dispatch shortens waiting times for passengers, who may no longer have to wait for a driver 10 minutes away when a second driver is dropping off a passenger two minutes away.

Perhaps no less important, forward dispatch causes drivers to stay on the road substantially longer during busy periods — a key goal for both companies.

Uber and Lyft explain this in essentially the same way. “Drivers keep telling us the worst thing is when they’re idle for a long time,” said Kevin Fan, the director of product at Lyft. “If it’s slow, they’re going to go sign off. We want to make sure they’re constantly busy.”

While this is unquestionably true, there is another way to think of the logic of forward dispatch: It overrides self-control.

* * *

Uber officials say the feature initially produced so many rides at times that drivers began to experience a chronic Netflix ailment — the inability to stop for a bathroom break. Amid the uproar, Uber introduced a pause button.

“Drivers were saying: ‘I can never go offline. I’m on just continuous trips. This is a problem.’ So we redesigned it,” said Maya Choksi, a senior Uber official in charge of building products that help drivers. “In the middle of the trip, you can say, ‘Stop giving me requests.’ So you can have more control over when you want to stop driving.”

It is true that drivers can pause the services’ automatic queuing feature if they need to refill their tanks, or empty them, as the case may be. Yet once they log back in and accept their next ride, the feature kicks in again. To disable it, they would have to pause it every time they picked up a new passenger. By contrast, even Netflix allows users to permanently turn off its automatic queuing feature, known as Post-Play.

This pre-emptive hard-wiring can have a huge influence on behavior, said David Laibson, the chairman of the economics department at Harvard and a leading behavioral economist. Perhaps most notably, as Ms. Rosenblat and Luke Stark observed in an influential paper on these practices, Uber’s app does not let drivers see where a passenger is going before accepting the ride, making it hard to judge how profitable a trip will be.

Here’s how I would recast that, and add some much-needed economics:

Consider an algorithm called forward dispatch — Lyft has a similar one — that dispatches a new ride to a driver before the current one ends. Forward dispatch shortens waiting times for passengers, who may no longer have to wait for a driver 10 minutes away when a second driver is dropping off a passenger two minutes away.

Perhaps no less important, forward dispatch causes drivers to stay on the road substantially longer during busy periods — a key goal for both companies — by giving them more income-earning opportunities.

Uber and Lyft explain this in essentially the same way. “Drivers keep telling us the worst thing is when they’re idle for a long time,” said Kevin Fan, the director of product at Lyft. “If it’s slow, they’re going to go sign off. We want to make sure they’re constantly busy.”

While this is unquestionably true, and seems like another win-win, some critics have tried to paint even this means of satisfying both driver and consumer preferences in a negative light by claiming that the forward dispatch algorithm overrides self-control.

* * *

Uber officials say the feature initially produced so many rides at times that drivers began to experience a chronic Netflix ailment — the inability to stop for a bathroom break. Amid the uproar, Uber introduced a pause button.

“Drivers were saying: ‘I can never go offline. I’m on just continuous trips. This is a problem.’ So we redesigned it,” said Maya Choksi, a senior Uber official in charge of building products that help drivers. “In the middle of the trip, you can say, ‘Stop giving me requests.’ So you can have more control over when you want to stop driving.”

Tweaks like these put paid to the arguments that Uber is simply trying to abuse its drivers. And yet, critics continue to make such claims:

It is true that drivers can pause the services’ automatic queuing feature if they need to refill their tanks, or empty them, as the case may be. Yet once they log back in and accept their next ride, the feature kicks in again. To disable it, they would have to pause it every time they picked up a new passenger. By contrast, even Netflix allows users to permanently turn off its automatic queuing feature, known as Post-Play.

It’s difficult to take seriously claims that Uber “abuses” drivers by setting a default that drivers almost certainly prefer; surely drivers seek out another fare following the last fare more often than they seek out another bathroom break. In any case, the difference between one default and the other is a small change in the number of times drivers might have to push a single button; hardly a huge impediment.

But such claims persist, nevertheless. Setting a trivially different default can have a huge influence on behavior, claims David Laibson, the chairman of the economics department at Harvard and a leading behavioral economist. Perhaps most notably — and to change the subject — as Ms. Rosenblat and Luke Stark observed in an influential paper on these practices, Uber’s app does not let drivers see where a passenger is going before accepting the ride, making it hard to judge how profitable a trip will be. But there are any number of defenses of this practice, from both a driver- and consumer-welfare standpoint. Not least, such disclosure could well create isolated scarcity for a huge range of individual ride requests (as opposed to the general scarcity during a “surge”), leading to longer wait times, the need to adjust prices for consumers on the basis of individual rides, and more intense competition among drivers for the most profitable rides. Given these and other explanations, it is extremely unlikely that the practice is actually aimed at “abusing” drivers.

As they say, read the whole thing!

Geoffrey A. Manne is Executive Director of the International Center for Law & Economics

Dynamic versus static competition

Ever since David Teece and coauthors began writing about antitrust and innovation in high-tech industries in the 1980s, we’ve understood that traditional, price-based antitrust analysis is not intrinsically well-suited for assessing merger policy in these markets.

For high-tech industries, performance, not price, is paramount — which means that innovation is key:

Competition in some markets may take the form of Schumpeterian rivalry in which a succession of temporary monopolists displace one another through innovation. At any one time, there is little or no head-to-head price competition but there is significant ongoing innovation competition.

Innovative industries are often marked by frequent disruptions or “paradigm shifts” rather than horizontal market share contests, and investment in innovation is an important signal of competition. And competition comes from the continual threat of new entry down the road — often from competitors who, though they may start with relatively small market shares, or may arise in different markets entirely, can rapidly and unexpectedly overtake incumbents.

Which, of course, doesn’t mean that current competition and ease of entry are irrelevant. Rather, because, as Joanna Shepherd noted, innovation should be assessed across the entire industry and not solely within merging firms, conduct that might impede new, disruptive, innovative entry is indeed relevant.

But it is also important to remember that innovation comes from within incumbent firms, as well, and, often, that the overall level of innovation in an industry may be increased by the presence of large firms with economies of scope and scale.

In sum, and to paraphrase Olympia Dukakis’ character in Moonstruck: “what [we] don’t know about [the relationship between innovation and market structure] is a lot.”

What we do know, however, is that superficial, concentration-based approaches to antitrust analysis will likely overweight presumed foreclosure effects and underweight innovation effects.

We shouldn’t fetishize entry, or access, or head-to-head competition over innovation, especially where consumer welfare may be significantly improved by a reduction in the former in order to get more of the latter.

As Katz and Shelanski note:

To assess fully the impact of a merger on market performance, merger authorities and courts must examine how a proposed transaction changes market participants’ incentives and abilities to undertake investments in innovation.

At the same time, they point out that

Innovation can dramatically affect the relationship between the pre-merger marketplace and what is likely to happen if the proposed merger is consummated…. [This requires consideration of] how innovation will affect the evolution of market structure and competition. Innovation is a force that could make static measures of market structure unreliable or irrelevant, and the effects of innovation may be highly relevant to whether a merger should be challenged and to the kind of remedy antitrust authorities choose to adopt. (Emphasis added).

Dynamic competition in the ag-biotech industry

These dynamics seem to be playing out in the ag-biotech industry. (For a detailed look at how the specific characteristics of innovation in the ag-biotech industry have shaped industry structure, see, e.g., here (pdf)).  

One inconvenient truth for the “concentration reduces innovation” crowd is that, as the industry has experienced more consolidation, it has also become more, not less, productive and innovative. Between 1995 and 2015, for example, the market share of the largest seed producers and crop protection firms increased substantially. And yet, over the same period, annual industry R&D spending went up nearly 750 percent. Meanwhile, the resulting innovations have increased crop yields by 22%, reduced chemical pesticide use by 37%, and increased farmer profits by 68%.

In her discussion of the importance of considering the “innovation ecosystem” in assessing the innovation effects of mergers in R&D-intensive industries, Joanna Shepherd noted that

In many consolidated firms, increases in efficiency and streamlining of operations free up money and resources to source external innovation. To improve their future revenue streams and market share, consolidated firms can be expected to use at least some of the extra resources to acquire external innovation. This increase in demand for externally-sourced innovation increases the prices paid for external assets, which, in turn, incentivizes more early-stage innovation in small firms and biotech companies. Aggregate innovation increases in the process!

The same dynamic seems to play out in the ag-biotech industry, as well:

The seed-biotechnology industry has been reliant on small and medium-sized enterprises (SMEs) as sources of new innovation. New SME startups (often spinoffs from university research) tend to specialize in commercial development of a new research tool, genetic trait, or both. Significant entry by SMEs into the seed-biotechnology sector began in the late 1970s and early 1980s, with a second wave of new entrants in the late 1990s and early 2000s. In recent years, exits have outnumbered entrants, and by 2008 just over 30 SMEs specializing in crop biotechnology were still active. The majority of the exits from the industry were the result of acquisition by larger firms. Of 27 crop biotechnology SMEs that were acquired between 1985 and 2009, 20 were acquired either directly by one of the Big 6 or by a company that itself was eventually acquired by a Big 6 company.

While there is more than one way to interpret these statistics (and they are often used by merger opponents, in fact, to lament increasing concentration), they are actually at least as consistent with an increase in innovation through collaboration (and acquisition) as with a decrease.

For what it’s worth, this is exactly how the startup community views the innovation ecosystem in the ag-biotech industry, as well. As the latest AgFunder AgTech Investing Report states:

The large agribusinesses understand that new innovation is key to their future, but the lack of M&A [by the largest agribusiness firms in 2016] highlighted their uncertainty about how to approach it. They will need to make more acquisitions to ensure entrepreneurs keep innovating and VCs keep investing.

It’s also true, as Diana Moss notes, that

Competition maximizes the potential for numerous collaborations. It also minimizes incentives to refuse to license, to impose discriminatory restrictions in technology licensing agreements, or to tacitly “agree” not to compete…. All of this points to the importance of maintaining multiple, parallel R&D pipelines, a notion that was central to the EU’s decision in Dow-DuPont.

And yet collaboration and licensing have long been prevalent in this industry. Examples are legion, but here are just a few significant ones:

  • Monsanto’s “global licensing agreement for the use of the CRISPR-Cas genome-editing technology in agriculture with the Broad Institute of MIT and Harvard.”
  • Dow and Arcadia Biosciences’ “strategic collaboration to develop and commercialize new breakthrough yield traits and trait stacks in corn.”
  • Monsanto and the University of Nebraska-Lincoln’s “licensing agreement to develop crops tolerant to the broadleaf herbicide dicamba. This agreement is based on discoveries by UNL plant scientists.”

Both large and small firms in the ag-biotech industry continually enter into new agreements like these. See, e.g., here and here for a (surely incomplete) list of deals in 2016 alone.

At the same time, across the industry, new entry has been rampant despite increased M&A activity among the largest firms. Recent years have seen venture financing in AgTech skyrocket — from $400 million in 2010 to almost $5 billion in 2015 — and hundreds of startups now enter the industry annually.

The pending mergers

Today’s pending mergers are consistent with this characterization of a dynamic market in which structure is being driven by incentives to innovate, rather than monopolize. As Michael Sykuta points out,

The US agriculture sector has been experiencing consolidation at all levels for decades, even as the global ag economy has been growing and becoming more diverse. Much of this consolidation has been driven by technological changes that created economies of scale, both at the farm level and beyond.

These deals aren’t fundamentally about growing production capacity, expanding geographic reach, or otherwise enhancing market share; rather, each is a fundamental restructuring of the way the companies do business, reflecting today’s shifting agricultural markets, and the advanced technology needed to respond to them.

Technological innovation is unpredictable, often serendipitous, and frequently transformative of the ways firms organize and conduct their businesses. A company formed to grow and sell hybrid seeds in the 1920s, for example, would either have had to evolve or fold by the end of the century. Firms today will need to develop (or purchase) new capabilities and adapt to changing technology, scientific knowledge, consumer demand, and socio-political forces. The pending mergers seemingly fit exactly this mold.

As Allen Gibby notes, these mergers are essentially vertical combinations of disparate, specialized pieces of an integrated whole. Take the proposed Bayer/Monsanto merger, for example. Bayer is primarily a chemicals company, developing advanced chemicals to protect crops and enhance crop growth. Monsanto, on the other hand, primarily develops seeds and “seed traits” — advanced characteristics that ensure the heartiness of the seeds, give them resistance to herbicides and pesticides, and speed their fertilization and growth. In order to translate the individual advances of each into higher yields, it is important that these two functions work successfully together. Doing so enhances crop growth and protection far beyond what, say, spreading manure can accomplish — or either firm could accomplish working on its own.

The key is that integrated knowledge is essential to making this process function. Developing seed traits to work well with (i.e., to withstand) certain pesticides requires deep knowledge of the pesticide’s chemical characteristics, and vice-versa. Processing huge amounts of data to determine when to apply chemical treatments or to predict a disease requires not only that the right information is collected, at the right time, but also that it is analyzed in light of the unique characteristics of the seeds and chemicals. Increased communications and data-sharing between manufacturers increases the likelihood that farmers will use the best products available in the right quantity and at the right time in each field.

Vertical integration solves bargaining and long-term planning problems by unifying the interests (and the management) of these functions. Instead of arm’s length negotiation, a merged Bayer/Monsanto, for example, may better maximize R&D of complicated Ag/chem products through fully integrated departments and merged areas of expertise. A merged company can also coordinate investment decisions (instead of waiting up to 10 years to see what the other company produces), avoid duplication of research, adapt to changing conditions (and the unanticipated course of research), pool intellectual property, and bolster internal scientific capability more efficiently. All told, the merged company projects spending about $16 billion on R&D over the next six years. Such coordinated investment will likely garner far more than either company could from separately spending even the same amount to develop new products. 

Controlling an entire R&D process and pipeline of traits for resistance, chemical treatments, seeds, and digital complements would enable the merged firm to better ensure that each of these products works together to maximize crop yields, at the lowest cost, and at greater speed. Consider the advantages that Apple’s tightly-knit ecosystem of software and hardware provides to computer and device users. Such tight integration isn’t the only way to compete (think Android), but it has frequently proven to be a successful model, facilitating some functions (e.g., handoff between Macs and iPhones) that are difficult if not impossible in less-integrated systems. And, it bears noting, important elements of Apple’s innovation have come through acquisition….

Conclusion

As LaFontaine and Slade have made clear, theoretical concerns about the anticompetitive consequences of vertical integrations are belied by the virtual absence of empirical support:

Under most circumstances, profit–maximizing vertical–integration and merger decisions are efficient, not just from the firms’ but also from the consumers’ points of view.

Other antitrust scholars are skeptical of vertical-integration fears because firms normally have strong incentives to deal with providers of complementary products. Bayer and Monsanto, for example, might benefit enormously from integration, but if competing seed producers seek out Bayer’s chemicals to develop competing products, there’s little reason for the merged firm to withhold them: Even if the new seeds out-compete Monsanto’s, Bayer/Monsanto can still profit from providing the crucial input. Its incentive doesn’t necessarily change if the merger goes through, and whatever “power” Bayer has as an input is a function of its scientific know-how, not its merger with Monsanto.

In other words, while some competitors could find a less hospitable business environment, consumers will likely suffer no apparent ill effects, and continue to receive the benefits of enhanced product development and increased productivity.

That’s what we’d expect from innovation-driven integration, and antitrust enforcers should be extremely careful before thwarting or circumscribing these mergers lest they end up thwarting, rather than promoting, consumer welfare.

Nicolas Petit is Professor of Law at the University of Liege (Belgium) and Research Professor at the University of South Australia (UniSA)

This symposium offers a good opportunity to look again into the complex relation between concentration and innovation in antitrust policy. Whilst the details of the EC decision in Dow/Dupont remain unknown, the press release suggests that the issue of “incentives to innovate” was central to the review. Contrary to what had leaked in the antitrust press, the decision has apparently backed off from the introduction of a new “model”, and instead followed a more cautious approach. After a quick reminder of the conventional “appropriability v cannibalizationframework that drives merger analysis in innovation markets (1), I make two sets of hopefully innovative remarks on appropriability and IP rights (2) and on cannibalization in the ag-biotech sector (3).

Appropriability versus cannibalization

Antitrust economics 101 teach that mergers affect innovation incentives in two polar ways. A merger may increase innovation incentives. This occurs when the increment in power over price or output achieved through merger enhances the appropriability of the social returns to R&D. The appropriability effect of mergers is often tied to Joseph Schumpeter, who observed that the use of “protecting devices” for past investments like patent protection or trade secrecy constituted a “normal elemen[t] of rational management”. The appropriability effect can in principle be observed at firm – specific incentives – and industry – general incentives – levels, because actual or potential competitors can also use the M&A market to appropriate the payoffs of R&D investments.

But a merger may decrease innovation incentives. This happens when the increased industry position achieved through merger discourages the introduction of new products, processes or services. This is because an invention will cannibalize the merged entity profits in proportions larger as would be the case in a more competitive market structure. This idea is often tied to Kenneth Arrow who famously observed that a “preinvention monopoly power acts as a strong disincentive to further innovation”.

Schumpeter’s appropriability hypothesis and Arrow’s cannibalization theory continue to drive much of the discussion on concentration and innovation in antitrust economics. True, many efforts have been made to overcome, reconcile or bypass both views of the world. Recent studies by Carl Shapiro or Jon Baker are worth mentioning. But Schumpeter and Arrow remain sticky references in any discussion of the issue. Perhaps more than anything, the persistence of their ideas denotes that both touched a bottom point when they made their seminal contribution, laying down two systems of belief on the workings of innovation-driven markets.

Now beyond the theory, the appropriability v cannibalization gravitational models provide from the outset an appealing framework for the examination of mergers in R&D driven industries in general. From an operational perspective, the antitrust agency will attempt to understand if the transaction increases appropriability – which leans in favour of clearance – or cannibalization – which leans in favour of remediation. At the same time, however, the downside of the appropriability v cannibalization framework (and of any framework more generally) may be to oversimplify our understanding of complex phenomena. This, in turn, prompts two important observations on each branch of the framework.

Appropriability and IP rights

Any antitrust agency committed to promoting competition and innovation should consider mergers in light of the degree of appropriability afforded by existing protecting devices (essentially contracts and entitlements). This is where Intellectual Property (“IP”) rights become relevant to the discussion. In an industry with strong IP rights, the merging parties (and its rivals) may be able to appropriate the social returns to R&D without further corporate concentration. Put differently, the stronger the IP rights, the lower the incremental contribution of a merger transaction to innovation, and the higher the case for remediation.

This latter proposition, however, rests on a heavy assumption: that IP rights confer perfect appropriability. The point is, however, far from obvious. Most of us know that – and our antitrust agencies’ misgivings with other sectors confirm it – IP rights are probabilistic in nature. There is (i) no certainty that R&D investments will lead to commercially successful applications; (ii) no guarantee that IP rights will resist to invalidity proceedings in court; (iii) little safety to competition by other product applications which do not practice the IP but provide substitute functionality; and (iv) no inevitability that the environmental, toxicological and regulatory authorization rights that (often) accompany IP rights will not be cancelled when legal requirements change. Arrow himself called for caution, noting that “Patent laws would have to be unimaginably complex and subtle to permit [such] appropriation on a large scale”. A thorough inquiry into the specific industry-strength of IP rights that goes beyond patent data and statistics thus constitutes a necessary step in merger review.

But it is not a sufficient one. The proposition that strong IP rights provide appropriability is essentially valid if the observed pre-merger market situation is one where several IP owners compete on differentiated products and as a result wield a degree of market power. In contrast, the proposition is essentially invalid if the observed pre-merger market situation leans more towards the competitive equilibrium and IP owners compete at prices closer to costs. In both variants, the agency should thus look carefully at the level and evolution of prices and costs, including R&D ones, in the pre-merger industry. Moreover, in the second variant, the agency ought to consider as a favourable appropriability factor any increase of the merging entity’s power over price, but also any improvement of its power over cost. By this, I have in mind efficiency benefits, which can arise as the result of economies of scale (in manufacturing but also in R&D), but also when the transaction combines complementary technological and marketing assets. In Dow/Dupont, no efficiency argument has apparently been made by the parties, so it is difficult to understand if and how such issues have played a role in the Commission’s assessment.

Cannibalization, technological change, and drastic innovation

Arrow’s cannibalization theory – namely that a pre-invention monopoly acts as a strong disincentive to further innovation – fails to capture that successful inventions create new technology frontiers, and with them entirely novel needs that even a monopolist has an incentive to serve. This can be understood with an example taken from the ag-biotech field. It is undisputed that progress in crop protection science has led to an expanding range of resistant insects, weeds, and pathogens. This, in turn, is one (if not the main) key drivers of ag-tech research. In a 2017 paper published in Pest Management Science, Sparks and Lorsbach observe that:

resistance to agrochemicals is an ongoing driver for the development of new chemical control options, along with an increased emphasis on resistance management and how these new tools can fit into resistance management programs. Because resistance is such a key driver for the development of new agrochemicals, a highly prized attribute for a new agrochemical is a new MoA [method of action] that is ideally a new molecular target either in an existing target site (e.g., an unexploited binding site in the voltage-gated sodium channel), or new/under-utilized target site such as calcium channels.

This, and other factors, leads them to conclude that:

even with fewer companies overall involved in agrochemical discovery, innovation continues, as demonstrated by the continued introduction of new classes of agrochemicals with new MoAs.

Sparks, Hahn, and Garizi make a similar point. They stress in particular that the discovery of natural products (NPs) which are the “output of nature’s chemical laboratory” is today a main driver of crop protection research. According to them:

NPs provide very significant value in identifying new MoAs, with 60% of all agrochemical MoAs being, or could have been, defined by a NP. This information again points to the importance of NPs in agrochemical discovery, since new MoAs remain a top priority for new agrochemicals.

More generally, the point is not that Arrow’s cannibalization theory is wrong. Arrow’s work convincingly explains monopolists’ low incentives to invest in substitute invention. Instead, the point is that Arrow’s cannibalization theory is narrower than often assumed in the antitrust policy literature. Admittedly, Arrow’s cannibalization theory is relevant in industries primarily driven by a process of cumulative innovation. But it is much less helpful to understand the incentives of a monopolist in industries subject to technological change. As a result of this, the first question that should guide an antitrust agency investigation is empirical in nature: is the industry under consideration one driven by cumulative innovation, or one where technology disruption, shocks, and serendipity incentivize drastic innovation?

Note that exogenous factors beyond technological frontiers also promote drastic innovation. This point ought not to be overlooked. A sizeable amount of the specialist scientific literature stresses the powerful innovation incentives created by changing dietary habits, new diseases (e.g. the Zika virus), global population growth, and environmental challenges like climate change and weather extremes. In 2015, Jeschke noted:

In spite of the significant consolidation of the agrochemical companies, modern agricultural chemistry is vital and will have the opportunity to shape the future of agriculture by continuing to deliver further innovative integrated solutions. 

Words of wisdom caution for antitrust agencies tasked with the complex mission of reviewing mergers in the ag-biotech industry?