I read through Harvard’s Berkman Center report which was led by Yochai Benkler. Okay, it was 232 pages so I “mostly read through it” to paraphrase a famous movie “The Princess Bride” (see it – there is a great scene where the hero is “only mostly dead”.) I did read the substantive analysis at the beginning (over 100 pages) and skipped only some of the country reports.
Benkler and his team of investigators can be commended for the effort that went into what was no doubt a substantial undertaking. The report, however, seems to be very limited in its analysis despite the FCC’s original charge to “conduct an expert review of existing literature and studies about broadband deployment and usage throughout the world to inform the Commission’s development of a National Broadband Plan.” While the report does look at a range of metrics related to broadband – especially price, adoption and speeds – it really does seem most focused on policy issues related to broadband. To get right to the heart of what it says in summary, it suggests that foreign unbundling and separation regulatory policies are good and the American platform competition model is not. That is a curious tack to take for a number of reasons.
First, I can find little in the way of substantial analysis regarding economic or financial issues relating to deployment and investment. The report focuses on price, adoption levels and speeds and totally ignores in most ways the dynamics regarding the markets in the countries it reviews. So, for example, while it extols France for encouraging competition through unbundling and having some of the fastest speeds available in Europe, France is behind other countries in Europe with regard to fiber deployment, and many of these countries have platform competition Europe as a whole is in fact behind the U. S. in local fiber network deployment.
Further, in criticizing the position of the incumbents against the FCC’s unbundling rules, it totally ignores the TELRIC pricing model (which resulted in prices that were so low they did not allow for a fair return on investment), the fact that unbundling was extreme in terms of how much of the network had to be unbundled (which is far different from the unbundling policies in most countries that do not require unbundling of every facet of the network), and did not even discuss the fact that unbundling policy was open ended (i.e., there was no “end game” to the policy and it very likely could have resulted in an ongoing subsidy to new entrants for years to come). In other words, the study simply assumes unbundling is a good policy, is the right policy to encourage competition and it did not look at the impacts of the policy in terms of investment nor at how the policy was structured - for example, to encourage or discourage network investment. As far as I know, no other country has gone anywhere near as far as the FCC did in requiring unbundling of numerous aspects of the network – from the switch to collocation to the lines. This is a big difference and it is totally ignored from what I can see.
In fact, investment and build out are almost totally ignored in terms of any in depth assessment of these issues. The report does extol the Korean and Japanese models in terms of promoting fiber deployment but does not focus attention, for example, on the fact that the U. S. is ahead of Europe in fiber deployment. It criticizes the use of PON fiber technologies not because they are not good for consumers or don’t produce high capacity networks that benefit consumers but rather because they can’t be physically unbundled. The report ignores whether PON technologies are a good approach from a financial and build out perspective and whether they make economic sense.
The report also provides almost no in depth analysis of the U. S. moves away from the extreme unbundling policies of the late 1990’s and suggests that the changes in unbundling began with the new Administration in 2001-2002. It completely ignores the decision by the Kennard FCC not to impose common carrier regulation on cable broadband networks. This was in fact one of the earliest of the regulatory policy changes that helped promote the investment and competition we have today. It did not start in 2001.
To me, the bottom line with unbundling policy is that price, what is to be unbundled, and how to transition to network based competition are all very difficult problems. Here I can find little analysis in the Harvard report. The reason markets are so important in growth economies is that setting prices is far from an easy task. Competitive markets do it well and in such a way that the prices send the right signals to companies spurring investment and innovation. Regulators simply can’t replicate this process because it is a result of a complex feedback loop among companies, competitors, and consumers. My daughters once asked me when I was selling my house a few years ago “Why can’t you set a higher price or any price you want? Who controls prices?” I said all of us do in a market and it works because I know the prices that prevail in a market and I can take my buying and investment decisions from the pricing signals. But if I want, I can try to set a different price and if I am successful, that may set of some adjustments in the market.
Regulators really can only focus on a few issues in setting prices and the biggest one is cost. But cost alone does not drive prices.
Beyond these concerns, international comparisons do not address demand side problems. The whole section on International Comparisons is interesting, but not particularly helpful in creating a broader understanding. Since it is examining penetration (actual take up), it tends to mix the supply side and demand side arguments. Actually, the study ignores demand side issues pretty much entirely, creating the misimpression that this is all about supply. So for example, it notes the importance of price/income to adoption, and the fact that the U.S. has low rates for lower tier services, but then never explains why the U.S. is in the “middle of the pack.” If we have such low rates, we should have higher penetration. This would lead one to conclude that some other factors (Pew’s emphasis on relevancy, for example) are responsible for the relatively poor uptake numbers in the US.
The paper seems to assume throughout that we should apply the lessons of the past to the future, that we should recreate the copper policies in the fiber world. I think this suffers from two problems. First, the report oversells the lessons of the past - for example, how much of Japan and South Korea’s success can be attributed to direct government financial support through subsidies or loans. It appears quite a bit. Second, the paper seems to admit it’s a lot, but the larger point gets glossed over in favor of “open access works.”
The premise of open access policies was largely based on monopoly providers or providers with market power stifling competition, driving up prices, and so on. We’ve managed to adopt policies and our markets have evolved so that we now have many competitors and new networks. It seems you that the problem has been solved - the premise of the original argument (monopoly provider/network) is gone. But that is not the overall conclusion of the report.
It is also true that while there are general themes worth looking at globally, the reality is that the historical, legal conditions in the US are just different than in other countries. Even if AT&T was a monopoly for years, it’s long gone. Nor do we have the coupling between the government and companies that others had, when the government was the regulator and the monopoly provider.
I will review the comments submitted on the Benkler report with interest but that is my initial take of an extensive report.