Promise, Perils and the Big Switch Ahead for AI and BEAD
Thursday, April 20, 2023
Promise, Perils and the Big Switch Ahead for AI and BEAD
Remarks as Prepared for Delivery
BEAD Success Summit
Telecommunication Industry Association
I want to start with a word of thanks you deserved three years ago but have not yet received, then go back 30 years to discuss the next 20 years, and finally go back 90 years to set the stage for understanding our biggest challenge in the next 5 years.
If you like time travel, this speech is for you.
The thanks are this. Your industry—along with the scientists who so quickly utilized the mRNA technology to develop the vaccine—saved the American economy and society.
It was a collaborative effort involving many players, but the simple truth is that without the technological advancements the companies in this room developed, Covid would have shut the economy down much harder and longer.
Imagine what would have happened if Covid had happened in the era of the dial up internet.
Or even just ten years earlier, when the average download speed was less than 5 Mbps instead of the 80 plus it was when Covid hit.
Here’s one data point to illustrate. From 2019 to 2020 the number of Medicare telehealth visits increased 63-fold.
That data point about healthcare increasingly provided online has many analogues for education, job training, civic engagement, and many other essential functions as well. Those changes were not due to a change in technology. It was due a crisis forcing people to understand the power of the technology you all have created.
Again, you deserve our thanks. I am sure you would rather have it from the President or through an Act of Congress but, hey, take what you can get.
Those advances also illustrate something about the next 20 years. To understand those, let’s go back 30 years to the Federal Communications Commission in late 1993 when the new Chairman, Reed Hundt, arrived.
The biggest issue was what to do about Howard Stern’s language on radio.
Another big issue—for which Hundt received a letter on his first day from the largest number of Senators and House members that he would receive in his four years as Chair—was implementing the 1992 Cable Act, designed to bring competition and lower prices to consumers of multi-channel video. The big brewing battle, however, was how to enable greater competition in voice market by enabling long-distance and local phone providers to compete against each other.
That is what the world, particularly the FCC policy world, saw as central. Reed saw something different.
As he prepared to take over the agency, he had many meetings with experts in the law, economics, and technology of communications.
Perhaps the most memorable was the one with MIT Futurist Nick Negroponte, after which the “Negroponte Switch” was named. He hypothesized that everything we did on wires—voice—would be done wirelessly while everything we did wirelessly—video with broadcast television—would be done on wireless. He told Hundt that data needed to go through fiber optic cable, a vision reinforced by another participant of the meeting, George Gilder, who predicted that fiber would carry an ever-expanding amount of data and therefore would cause vast data creation.
This vision has proven exactly right and created for Reed and his team a North Star of enabling fiber and wireless to wrap around the world that would enable communications to constantly become faster, cheaper, and better.
He brought that understanding to interpreting the August 1993 Congressional action giving the FCC authority to create the digital cellular industry’s structure, its relationship to landline networks, and network operating standards.
He also brought that understanding to how he read a December 1993, New York Times story revealing the Mosaic browser as the first “killer app” of network computing.
These events gave the FCC the opportunity to help jumpstart digital cellular and the Internet (to which Mosaic was the gateway), and American firms to lead in both world changing domains.
Only 14 years later, when Steve Jobs introduced the iPhone in 2007, those domains would be irrevocably fused, again by an American firm.
That history reminds us that strategy consists of selecting among choices and irrevocably committing to the selection. Reed’s FCC chose digital cellular and internet as the essential media, knowing as we did so that we would be furthering the transformation or even potential destruction of all known media and mediums—newspapers, broadcast television, and cable—by enabling superior platforms and their undermining their three sources of revenue: advertising, subscription, and commerce.
Today, government officials have new strategic decisions to make just as momentous as the ones the intersection of policy and technology dumped in our FCC laps back in the early 1990’s.
The biggest one is Artificial Intelligence, AI.
To be clear, the lesson is not that technology eliminates the need for policy. Rather, it’s that when there is a big technology change, what was thought of as the main show may turn out to be less important than what we might see as a side show.
In 1993, the main show was the debate over local entry into long distance, which resulted in the 14-point check list Congress adopted in the Telecommunications Act of 1996 and the FCC implemented. The side show at that time was protecting the internet through Section 230 and requiring telco unbundling to enable dial-up ISP competition; creating wireless competition through auction design and number portability and enabling wireless to wired competition through a reduction of intercarrier compensation.
The side show proved more enduring and important than the main show.
We are about to go through another big switch.
It cannot be stated as simply and eloquently as Negroponte stated his, but it will be every bit as profound. As a result of AI, many things that used to be hard, will now be easy. And many things we never worried about will move to the front burner. So, it is appropriate that we are starting a several decade process of debating, adopting, and revising rules related to AI.
The White House, the Department of Commerce, and Senate are already studying the issue. They are several years behind the EU. And then there’s China.
One big difference between now and the 1990’s. We didn’t have a big geo-political and economic competitor. Now we do.
China’s own regulation and use of AI will continually shape how we think about and use AI.
We should be excited. We should be nervous.
But as we approach the regulatory task, we must be willing to change our views to align with new realities.
No one has said it better than Lincoln: “The dogmas of the quiet past, are inadequate to the stormy present. The occasion is piled high with difficulty, and we must rise—with the occasion. As our case is new, so we must think anew, and act anew. We must disenthrall ourselves, and then we shall save our country.”
For example, we still need to worry about bottlenecks. But we should understand how AI may create some, while at the same time creating opportunities to destroy existing ones.
The same can be said for other concerns, like privacy, safety, transparency, and dangerous misinformation, among other topics. In particular, we need to worry even about cybersecurity.
While Silicon Valley is currently obsessed with capturing the upside of AI, your organization, and policy makers must remain vigilant about protecting against the downside of cyber-attacks. Without secure networks, there will be no AI dividend for our economy and society, something we should all aspire for.
In short, we should hope that policymakers come to the table with the right questions, not answers formed in yesterday’s realities.
We should also anticipate that at every convention you hold for as far as the eye can see, AI, which will travel on the equipment you all build, will become the main event.
As we look out on a future in which more and more of our economic and civic activity involves artificial intelligence and online communications, however, we should not forget there is an urgent and critical task; eliminating the digital divide.
Let’s go back 90 years to when FDR said, “I see one-third of a nation ill-housed...”
Roosevelt did something great to address that problem. He created the Federal Housing Authority, an institution that made it easier for Americans to obtain mortgages. The Authority, however, also did something horrible. It restricted lending in African American neighborhoods. The impact has been devasting and long lasting.
It is one of the most significant reasons that the average wealth of an African American family of a college graduate is less than the average wealth of a white family headed by a high-school graduate.
As the Washington Post reported, “Since housing equity makes up about two-thirds of median household wealth, excluding black Americans from establishing equity during a time of unprecedented rises in home values locked in and exacerbated wealth disparities.”
Let’s project forward.
Wealth creation in this century’s global information economy will require digital access and skills.
So, just as restricting home ownership in the 1930’s still causes negative reverberations for us today, so too, if restrictions on digital access and utilization, whatever their cause, were allowed to continue, it will weaken our economy and society for generations to come.
This has been clear at least since 2010 when the National Broadband Plan found that “The cost of this digital exclusion is large and growing.”
Congress put this understanding into law when, in the 2021 infrastructure bill, it wrote that “a broadband connection and digital literacy are increasingly critical to how individuals participate in the society, economy, and civic institutions of the United States; and access health care and essential services, obtain education, and build careers.”
And it provided funds to build networks everywhere and to assure that all could afford the service.
It is a great thing, and a necessary thing, to understand the divide and commit resources to address it. It is another, however, to effectively execute in solving the problem.
States, who have the principal responsibility, face many pitfalls. For one thing, the many problems of the FCC’s RDOF auction continue to bedevil state offices. For another, as the states design their competitive grant programs, they face a challenge in building the right models and adjusting as the facts change to accurately reflect constantly changing market realities, as well as local needs. States must also thread the needle between assuring the deployments happen while also making sure there are competitive forces that keep costs down and service quality up.
For example, the federal government requires the grantees to provide a match of funds, as well as a letter of credit. This is a reasonable requirement, as it is always important to make sure everyone has skin in the game. But there is uncertainty about the requirements. If interpreted in some ways, it will be difficult for some smaller players, who may be focused on harder to serve markets, to qualify, reducing competition across the state, and thereby increasing the costs to reach every unserved and underserved location.
Further, such requirements could, if not calibrated correctly, add costs without adding benefits to the program.
There is a similar need to thread the needle in drawing the maps of unserved and underserved areas for the potential grantees to bid on. Draw them too large and the state may limit who will bid, driving up the cost. Draw them too small and the state may attract no bidders for some areas, leaving some unconnected.
Another problem is permitting.
America has an unfortunate track record of public projects taking longer and being more expensive than in other countries.
As the National Broadband Plan noted “the cost of deploying a broadband network depends significantly on the costs that service providers incur to access conduits, ducts, poles and rights-of-way on public and private lands” and that permitting requirements and procedures for rights of way, poles, conduits and towers “are key to the efficient and streamlined deployment of broadband,” and that difficulties in such access “often prove to be the greatest impediment to the efficient, cost-effective, and timely deployment of broadband.”
It made multiple recommendations designed to improve deployment.
In that light, I was glad to see a House Subcommittee hearing on the topic yesterday.
However, I am not optimistic that the hearing will result in progress.
For one thing, the hearing, which was titled as “Breaking Barriers: Streamlining Permitting to Expedite Deployment” largely ignored the economic reality that in rural areas, the barriers are overwhelming economic, not administrative.
For another, there was not a single witness representing the institutions—state and local governments—who do the actual permitting.
I personally know most of the witnesses. They are all smart and honorable. I agree with parts of each of their testimony. But none have been involved in financing or constructing a communications network and therefore involved with how permitting affects the costs and process of deployment.
While the witnesses made some excellent points—particularly about the need for transparency and affordability—the hearing too often fell into the trap of blaming bureaucrats—easy to do and generally inaccurate—and didn’t focus on the underlying causes.
- problematic incentives for local capacity building, particularly in dealing with a pig in a python cycle of permitting requests;
- use of outdated technology, a problem throughout all levels of government on many issues;
- bad data and maps; and
- inadequate communications and cooperation between the many government jurisdictions often involved in approving the necessary permits for a project.
Permitting is a critical problem for our country and your industry. To solve it, we need a collaborative process that involves all key stakeholders and asks the right questions.
The hearing did not do that. But perhaps the subcommittee can try again, or others will take up the effort in a more productive manner.
Another pitfall for the states is that there will only be one bite at the apple.
This is not a once in a generation program.
It is a “once” program.
If it succeeds, there will be no need for another. If it fails, there will be no support for another.
But, as will any new program, there will be mistakes. There will need to be adjustments. But it’s hard to adjust when your first shot is your last shot.
The biggest potential pitfalls, however, are at the federal level.
One relates to the Buy America provisions being applied to projects funded by the Infrastructure Act.
In the State of the Union President Biden, after touting the Infrastructure Bill’s commitment to connect all Americans, said “And when we do these projects, we’re going to Buy American…Tonight, I’m also announcing new standards to require all construction materials used in federal infrastructure projects to be made in America.”
Unfortunately, a small but critical number of network components have no American sourcing. To produce the necessary supply here would take many years.
The challenge for the President and his Administration is that our country can close the rural digital divide in the next few years, or it can strictly enforce a Buy America mandate. It cannot do both.
And for reasons I explained in a Brookings blog, the priority should be broadband deployment.
I also think the Administration will come to the same conclusion and show some flexibility.
To be clear, my confidence is not because I think Administration officials will read my blog; it is because your organization, and your allies, have laid out a very compelling policy case for flexibility.
A bigger fear I have is that Congress will fail to refund the Affordable Connectivity Program, known as the ACP, thereby risking disconnecting about 50 million Americans who currently benefit from the program.
That would constitute the biggest step any country has ever taken to widen, rather than close, the digital divide. The damage to those 50 million Americans alone is justification for refunding.
In addition, ACP’s expiration will also create problems for the BEAD program we are discussing here.
The consulting group BCG studied the impact of ACP on that program and concluded that ACP reduces the necessary subsidy to deploy networks in rural areas by 25 percent per household, writing that the $360 per year ACP subsidy “reduces the per-household subsidy required to incentivize ISP investment by $500, generating benefits for the government and increasing the market attractiveness for new entrants and incumbent providers.”
If ACP disappears, the federal dollars will end up funding deployments to significantly fewer unserved and underserved locations.
In normal times, it would not be a big political lift to refund it.
One interesting reason is that it is the most bipartisan program in terms of impact that Congress has ever passed.
Breaking the benefits down by Congressional districts represented by Republicans and Democrats, one finds that in terms of dollars delivered ($2.783 billion v. $2.960 billion), enrolled households (9.975 million v. 10.590 million), and the percent of the district enrolled in the program (11.56 percent v. 12.45 percent), the numbers are almost identical.
Polling numbers also suggest bipartisan support with a January poll showing a “strong bipartisan majority of voters (78 percent) support continuing the ACP, including 64 percent of Republicans, 70 percent of Independents, and 95 percent of Democrats.”
Yet, the current budget politics make refunding an uphill battle.
As an example, earlier I mentioned a Subcommittee hearing that styled itself as breaking barriers to expedite broadband deployment.
It was good that at the hearing, many questions focused on the need for continuing ACP.
That makes sense. If you analyze what would be a bigger stimulus to broadband deployment; permitting reform or ACP reauthorization, the math provides an easy answer—ACP.
But there have been no hearings on refunding ACP.
And to date, neither Democratic nor Republican budgets call for the refunding. That should cause all of us to be concerned. Deeply concerned.
Let me close this way.
The 2010 National Broadband Plan painted a picture of broadband being “essential to opportunity and citizenship,” and that as a result, the lack of broadband “has the potential to exacerbate inequality.”
Twelve years later, that is still the reality we face. Except now, we have the political and financial capital necessary to change that.
Roosevelt, after describing the one-third ill-house, as well as ill-fed and ill-clothed, went on to say “But it is not in despair that I paint you that picture. I paint it for you in hope—because the nation, seeing and understanding the injustice in it, proposes to paint it out. We are determined to make every American citizen the subject of his country’s interest and concern.”
Roosevelt accomplished many great things, and we should share his hope. But we should also understand how his Administration did not live up to the vision of making “every American citizen the subject of his or her country’s interest and concern.”
We must not make the same mistake.
It is time for us to paint out the digital divide.
Blair Levin oversaw the development of the FCC’s 2010 National Broadband Plan. FCC Chairman Tom Wheeler cited Mr. Levin’s work, noting “no one’s done more to advance broadband expansion and competition through the vision of the National Broadband Plan and Gig.U.” Prior to his work on the National Broadband Plan, Mr. Levin worked as an analyst at Legg Mason and Stifel Nicolaus. Barron’s Magazine noted that in his work, he “has always been on top of developing trends and policy shifts in media and telecommunications … and has proved visionary in getting out in front of many of today’s headline-making events.” From 1993-1997, Levin served as Chief of Staff to FCC Chairman Reed Hundt. He is currently a policy advisor to New Street Research, an equity research firm, a Non-Resident Fellow at the Brookings Institute Metropolitan Policy Project, and recently completed a Global Broadband Plan for Refugees funded by the World Bank and the United Nations High Commissioner for Refugees.
The Benton Institute for Broadband & Society is a non-profit organization dedicated to ensuring that all people in the U.S. have access to competitive, High-Performance Broadband regardless of where they live or who they are. We believe communication policy - rooted in the values of access, equity, and diversity - has the power to deliver new opportunities and strengthen communities.
© Benton Institute for Broadband & Society 2023. Redistribution of this email publication - both internally and externally - is encouraged if it includes this copyright statement.
For subscribe/unsubscribe info, please email headlinesATbentonDOTorg