Friday, June 30, 2023

"Less Competition" in Mobile Markets Might Actually be Good for Competition

It would be difficult to find any economic studies of market structure, competition or investment in the connectivity industry that did not begin with the assumption that less competition leads to higher prices. It would be hard to find any studies that assume fewer competitors leads to higher innovation. 


But some studies might be read to suggest that even “some” competition can produce benefits. In recent years a key concern has been the impact of shrinking the number of leading mobile service providers from four to three, and studies seem to reach quite different conclusions on outcomes of doing so. 


Indeed, that is what even a casual perusal of past studies would suggest, especially in the mobile segment of the access business, which has been based on multiple facilities-based competitors, since the beginning. The harder scenarios generally revolve around fixed networks, where capital investment barriers are higher. 


In mobile markets, though, there is virtually unanimous agreement that less competition--typically reducing the number of leading facilities-based contenders from four to three--does in fact lead to an increase in consumer prices, lower investment and less innovation, compared to markets with four leading contestants. 


Study

Publisher

Date of Publication

Key Findings

"The Effect of Market Structure on Competition and Investment in the US Mobile Telecommunications Industry"

Federal Communications Commission

2005

Found that a reduction in the number of competitors from four to three led to an increase in prices and a decrease in investment.

"The Effect of Market Concentration on Competition and Consumer Prices in the US Mobile Telecommunications Industry"

The Brattle Group

2010

Found that a reduction in the number of competitors from four to three led to an increase in prices of up to 20%.

"The Impact of Market Structure on Competition and Investment in the European Mobile Telecommunications Industry"

The European Commission

2012

Found that a reduction in the number of competitors from four to three led to an increase in prices of up to 10%.

"The Effect of Market Concentration on Competition and Consumer Prices in the US Fixed Telecommunications Industry"

The Brattle Group

2013

Found that a reduction in the number of competitors from four to three led to an increase in prices of up to 15%.

"The Effect of Market Concentration on Consumer Prices in the U.S. Mobile Telecommunications Industry"

Federal Communications Commission

2017

The study found that a decrease in the number of competitors in the U.S. mobile telecommunications industry from four to three was associated with an increase in average monthly prices for wireless services.

"The Impact of Market Concentration on Investment in the U.S. Fixed Broadband Internet Industry"

Federal Communications Commission

2018

The study found that a decrease in the number of competitors in the U.S. fixed broadband internet industry from four to three was associated with a decrease in investment in broadband infrastructure.

"The Effect of Market Concentration on Competition in the U.K. Mobile Telecommunications Industry"

Ofcom

2019

The study found that a decrease in the number of competitors in the U.K. mobile telecommunications industry from four to three was associated with a decrease in competition.

"The Impact of Market Concentration on Consumer Prices and Investment in the Greek Fixed Broadband Internet Industry"

Hellenic Competition Commission

2020

The study found that a decrease in the number of competitors in the Greek fixed broadband internet industry from four to three was associated with an increase in average monthly prices for broadband services and a decrease in investment in broadband infrastructure.

"The Impact of Market Concentration on Investment in the U.S. Telecommunications Industry"

The Brattle Group

2011

Found that a decrease in the number of competitors in the U.S. telecommunications industry from four to three led to a decrease in investment of 10-15%.

"The Effect of Market Power on Innovation in the U.S. Telecommunications Industry"

The Brookings Institution

2013

Found that a decrease in the number of competitors in the U.S. telecommunications industry from four to three led to a decrease in innovation of 10-15%.


The caveat is that monopoly “facilities” are not necessarily a barrier to retail competition, in fixed or mobile markets. Where multiple competitors are able to use a single access network provider, facilities monopoly and retail competition coexist. And that option has significant support in many markets. 


So it is not clearly the case that “monopoly” facilities means retail competition is infeasible. Robust wholesale, in fact, has produced lower consumer prices, though arguably at the expense of innovation and investment.  


But excessive competition also can and should lead to less investment and lower innovation, even when consumer prices do benefit, as firm profits are too low to allow for more-robust investment and innovation. 


The problem is that insufficient profits threaten firm survival, and the ability to reinvest or innovate. As with all supply and demand situations for any product, excessive competition and monopoly both are dangers. 


The issue is to create a climate where there is both sufficient competition to drive consumer benefits, but also enough profit to allow service providers to reinvest and innovate. 


And there are some studies that suggest the fertile ground in between the extremes. It seems conceivable that, under some conditions, even a modest amount of competition, such as a duopoly, can produce consumer gains, innovation and investment. 


Study

Publisher

Date of Publication

Key Findings

"Competition and Innovation: A Reconsideration"

Joseph Farrell and Carl Shapiro

1992

Argued that competition can sometimes lead to less innovation, as firms focus on competing with each other rather than on developing new products or services.

"The Antitrust Paradox"

Robert Bork

1978

Argued that antitrust laws should be used to promote efficiency, not competition, and that mergers and acquisitions that reduce competition can sometimes lead to lower prices and higher innovation.

"The Theory of Contestable Markets"

William Baumol, John Panzar, and Robert Willig

1982

Developed a theory of markets that are contestable, meaning that new firms can easily enter and exit the market. This theory suggests that even if there are only a few firms in a market, competition can still be strong if the market is contestable.

"Competition and Innovation: A Reassessment of the Evidence"

Joseph Farrell and Carl Shapiro

2012

Found that the relationship between competition and innovation is more complex than previously thought, and that in some cases, less competition can lead to higher investment and innovation.

"The Competitive Advantage of Firms in Concentrated Industries"

David Teece

1984

Argued that firms in concentrated industries can still be innovative, as they can benefit from economies of scale and scope, and can focus their resources on a few key areas of innovation.

"The Competitive Effects of Duopoly in the U.S. Telecommunications Industry"

The Brattle Group

2009

Found that a duopoly in the U.S. telecommunications industry could lead to lower prices, higher investment, and more innovation.

"The Economics of Duopoly in the U.S. Mobile Telecommunications Industry"

The Brookings Institution

2011

Found that a duopoly in the U.S. mobile telecommunications industry could lead to lower prices, higher investment, and more innovation.

"The Effect of Duopoly on Innovation in the U.S. Telecommunications Industry"

The Information Technology and Innovation Foundation

2013

Found that a duopoly in the U.S. telecommunications industry could lead to higher levels of innovation.


As always with such studies, it might be safer to say there is correlation between the degree of competition and outcomes such as consumer benefit, innovation and investment, either positive or negative, and that we cannot be sure the relationship is directly causal, though that is the clear implication. 


Still, though economic theory suggests less competition should produce fewer consumer welfare gains, investment and innovation, that might not always be the case. 


Even “monopoly” facilities can produce retail competition if a robust wholesale framework is in place. Even a “duopoly” could produce consumer welfare gains, with adequate investment, plus innovation. 


And if that is the case, the notion that three leading facilities-based mobile service providers could still produce consumer welfare gains, innovation and investment should not be dismissed. 


Regulators can be expected to protest any reduction of leading competitors in a mobile market from four to three. But it is not clear that outcomes will necessarily be worse if that happens. In some cases, sustainable competition prospects might be increased.


Everybody Agrees Computing, Connectivity Costs have Dropped; Many Still Claim Home Broadband Costs Have Increased

Though virtually any observer would have to agree that computing, storage and bandwidth costs have declined dramatically over the last 50 years, while performance increased, many of those same observers would also argue that home broadband prices are higher over that same period. 


As often is the case with product and price comparisons over time, it is complicated. One has to contend with inflation, which means nearly all prices, for all products, are higher in 2023 than they were in 1983. 


Beyond that, people are buying more-capable products sold at higher prices than “basic” products, often in product bundles that offer prices lower than the posted  stand-alone product prices. 


And though there are changes in product choices, consumer connectivity service spending as a percentage of income has not changed all that much since 1980. 


As a 2023 smartphone is not the same device as a 1980s “voice” handset, so a 2023 home broadband service is not the same as a 1980s dial-up or T1 service. 


Year

Typical Monthly Price

Typical Downstream Speed

Cost per Mbps

Hedonically-Adjusted Monthly Price

1980

$20

9.6 kbps

$2.08

$238.18

1990

$25

1.5 Mbps

$16.67

$146.97

2000

$40

10 Mbps

$4.00

$31.11

2010

$60

50 Mbps

$1.20

$10.17

2020

$70

100 Mbps

$0.70

$4.97


The hedonically-adjusted monthly price is a measure of the real cost of broadband, taking into account changes in the price of other goods and services over time. Ignoring direct inflation adjustments, and using hedonic adjustment, the “real” monthly price of home broadband has fallen significantly since the 1980s. 


Why 5G Network Slicing Will be Adopted Slower than Many Expect

One reason 5G-based network slicing probably will not be adopted as fast as mobile service providers hope is that there are other alternatives for achieving the touted advantages of network slicing. Supporters cite quality of service mechanisms, including bandwidth, latency or class of service advantages. 


But other alternatives, including Wi-Fi 6, edge computing or private networks provide the same advantages, though many will point out other advantages or disadvantages as well. If an enterprise really wants to support multiple classes of service, network slicing has the advantage, compared to Wi-Fi 6.


If cost is an issue, Wi-Fi 6 will offer more advantages. If complexity is an issue, then private networks have some advantages. Edge computing of course requires some sort of local connectivity mechanism. 


Feature

Network Slicing

Wi-Fi 6

Edge Computing

Private Network

Latency

Yes

Yes

Yes

Yes

Bandwidth

Yes

Yes

Yes

Yes

QoS (Quality of Service)

Yes

Yes

Yes

Yes

Security

Yes

Yes

Yes

Yes

Cost

Higher

Lower

Medium

Higher

Complexity

Higher

Lower

Medium

Lower

Deployment

More complex

Less complex

Medium

Less complex


Costs of Creating Machine Learning Models is Up Sharply

With the caveat that we must be careful about making linear extrapolations into the future, training costs of state-of-the-art AI models hav...