A Roadmap to Better Broadband Grants

I’ve been thinking about the effectiveness of federal broadband grant programs. We’ve had three recent major sets of federal grant awards – the stimulus grants of 2007, the first CAF II grants in 2015 and the recently awarded CAF II reverse auctions. We also have an upcoming e-Connectivity grant program for $600 million. I think there are lessons to be learned from studying the difference in the results between these grants. These lessons apply to State grant programs as well as any new federal programs.

Don’t Reward Slow Broadband Speeds. Probably the most bone-headed decision made by the FCC in my memory was handing out billions in CAF II to upgrade rural copper to 10/1 Mbps. This wasn’t considered decent broadband at the time of this decision and yet these upgrades continue to be funded today. The FCC could still take back the remaining CAF II money and redirect these funds to a reverse auction, which we just saw produced much faster speeds in areas with far less density than the CAF II footprint.

Keep Politics Out of It. The CAF II decision to give all of the funding to the big telcos was purely political and resulted in a huge waste of money that could have created many real broadband solutions. The FCC is supposed to be an independent agency, and it’s shameful that lobbyists were able to kill the reverse auction originally planned for CAF II. We are seeing politics back on the table with the e-Connectivity grants where Congress created a feel-good grant program, but then saddled it with a restriction that no more than 10% of homes in a study area can have existing 10/1 Mbps speeds. The reason for this provision was not even hidden, with the big telcos saying they didn’t want federal grant money to be used to compete against them.

Don’t Fund Inadequate Technologies. AT&T is using LTE cellular broadband to satisfy CAF II. This technology will never provide adequate broadband. In the recent reverse auction we saw money going to high-altitude satellite companies. Regardless of speeds that can be delivered with these satellites, the latency is so poor that it limits the ability to use the broadband for important activities like working at home or taking on-line classes.

Don’t Stress Anchor Institutions over People. The stimulus grants required middle mile providers to pop off of highways to build expensive last mile fiber to a handful of anchor institutions – schools, libraries, etc. While these anchor institutions need good broadband, so do the neighborhoods around them. This requirement added a lot of cost to the middle-mile projects as well as made it harder for anybody else to build a last mile network since the biggest bandwidth users in a community already have fiber.

Build to Industry Practices. The stimulus grants required that fiber builders conduct expensive environmental studies and historic preservation studies. That was the first time I ever saw those requirements in my forty years in the industry. Since telecom infrastructure is built almost entirely in existing public right-of-way these restrictions added a lot of cost but zero value to the projects.

Penalize Companies that Cheat. There needs to be repercussions for companies that cheat on grant applications to win the funding. The biggest area of cheating is claiming speeds that the technology can’t deliver. The FCC follows up grants with a decent speed-test program, but the worst repercussion in failing these tests is to not get funding going forward. A carrier that badly fails the speed tests should have to return the original grant funding. I’m also hearing rumors that the many rural households covered by CAF II will not get the promised upgrades – and if so, the big telcos should be forced to return a proportionate amount of that funding for homes that don’t get the promised upgrades.

Building Fiber to Anchor Institutions

The Schools, Health & Libraries Broadband Coalition (SHLB) announced a strategy to bring broadband to every anchor institution in the continental US. They estimate this would cost between $13 and $19 billion. They believe this would act as a first step to bring broadband to unserved and underserved rural communities.

While this sounds like a reasonable idea, we’ve tried this before and it largely hasn’t worked. Recall that the BTOP program in 2009 and 2010 funded a lot of middle mile fiber projects that brought broadband deeper into parts of the country that didn’t have enough fiber. That program required the BTOP middle mile fiber providers to serve all anchor institutions along the path of their networks and was a smaller version of this same proposal.

We’re approaching a decade later and a lot of the communities connected by BTOP middle mile grants still don’t have a last mile broadband network. There are some success stories, so I don’t want to say that middle mile fiber has no value – but for the most part nobody is making that last mile investment in rural areas just because the BTOP middle mile fiber was built.

BTOP isn’t the only program that has built fiber to anchor institutions. There are a number of states and counties that have built fiber networks for the express purposes of serving anchor institutions. There are also numerous fiber networks that have been built by school systems to support the schools.

In many cases I’ve seen these various anchor institution networks actually hurt potential last mile fiber investment. Anybody that is going to build rural fiber needs as many ‘large’ customers as it can get to help offset building expensive rural fiber. I’ve had clients who were thinking about building fiber to a small rural town only to find out that the school, city hall and other government locations already had inexpensive broadband on an existing fiber network. Taking those revenues out of the equation can be enough to sink a potential business plan.

At least BTOP fiber required that the network owners make it easy for last mile providers to get reasonably priced backbone access on their networks. Many of the state and school board networks are prohibited from allowing any commercial use of their network. I’ve never understood these prohibitions against sharing spare pairs of government fiber with others, but they are fairly common. Most come from State edicts that are likely prompted by the lobbyists for the big carriers.

I’m sure I’ll take some flak for my position, but I’ve seen the negative results of this idea too many times in the real world. Communities get frustrated when they see a gigabit connection at a school or City Hall when nobody else in the area has decent broadband. I’ve even seen government staff and officials who have fast broadband in their offices turn a deaf ear to the rest of the community that has poor or no broadband.

To make matters worse, many of the BTOP networks have run into economic difficulties. The companies that invested in BTOP bought into the hype that the middle mile fiber networks would attract last mile fiber investments, and they counted on those extra revenues for long-term viability. But a significant portion of the BTOP middle mile networks ended up being fiber to nowhere. Companies funded by BTOP needed to bring matching capital, and a number of the BTOP providers have had to sell their networks at a huge discount and walk away from their unpaid debt since the revenues to cover debt payments never materialized.

This also raises the question of who is going to maintain the enormous miles of fiber that would be built by this proposal. Somebody has to pay the electric bill to keep the fiber lit. Somebody needs to do routine maintenance as well as fix fiber cuts and storm damage. And somebody has to pay to periodically replace the electronics on the network, which have an average economic life of around ten years.

I feel certain I will get an inbox full of comments about this blog. I’m bound to get stories telling me about some of the great success stories from the BTOP networks – and they do exist. There are cases where the middle mile fiber made it easier for some ISP to build last mile fiber to a rural community. And certainly a lot of extremely rural schools, libraries and other anchor institutions have benefitted from the BTOP requirement to serve them. But I believe there are more stories of failure that offset the success stories.

I seriously doubt that this FCC and administration would release this much money for any kind of rural broadband. But this is the kind of idea that can catch the interest of Congress and that could somehow get funded. There is no politician in DC who will take a stance against schools and libraries.

I can think of much better ways to spend that much money in ways that would bring broadband solutions many whole rural communities, not just to the anchor institutions. That’s not enough money to fix all of our rural broadband issues, but it would be a great start, particularly if distributed in a grant program for last mile projects that requires matching private investment.

Checking in on Internet2

internet2I recently checked in with Internet2 to see what they are up to these days. For those of you not familiar with Internet2, it’s a high-speed data network operator by and for the benefit of many colleges and universities.

Internet2 came about due to the fact that universities were unable to buy the commercial bandwidth they needed. In the early 90s universities were among the largest data users in the country. Universities often collaborate on science research and they wanted to be able to transmit huge data files for research on things like particle physics and medical imaging. There were also supercomputers at some campuses that other schools wanted to use.

And so the universities started looking for solutions to get fast data pipes between campuses. The first attempt of this was funded by the National Science Foundation in 1995 and was called vBNS. In 1997 the effort was picked up by EDUCOM, a consortium that was organized by the non-profit University Corporation for Advanced Internet Development. This corporation changed their name to Internet2 and worked with Qwest to build and assemble a 10 Gbps network in 1998. You have to put that speed into perspective. Today there are amazingly a handful of residential customers in the country with 10 Gbps connections. But in 1998 that speed was cutting edge. Internet2 subsequently began working with Level3 and they upgraded the network in 2007 to 100 Gbps.

The organization has grown over the years. The original project was funded by 34 universities. The network has now expanded and is comprised today of 252 universities, 82 corporations that collaborate in research, 68 large government affiliate members such as federal agencies, 41 regional and state education networks and 65 other research and educational networking partners around the world.

Internet2 has also reached out through these many connections into communities all over the country. The network (through the affiliated regional partners) now connects to over 60,000 anchor institutions that include primary and secondary schools, community colleges and universities, public libraries, museums and health care organizations.

The organization has also taken on other roles. In the early years the corporation was focused on assembling the network needed to connect to institutions. The network has been assembled through long-term IRU agreements for access to fiber. The second phase of the roll-out was to form the many alliances with state and regional government networks to extend the reach of the Internet2 bandwidth.

But now the organization is focusing on other ways to benefit members. For example, it has been negotiating cheap pricing for cloud services since many of the member institutions are moving functions into the cloud. The Internet2 staff also has begun negotiating bulk pricing for other telecom services like VoIP, commonly used software, etc.

Internet2 is probably the biggest example I know of what I call a service organization. This has always been a popular model in rural America and there are numerous service organizations that operate or assist rural electric companies or rural cooperatives of all types. There are also a number of statewide fiber networks owned by independent telephone companies that are smaller clones of Internet2. The one thing I’ve always seen is that the roles of service organizations always grow over time as members find more and more things that they can do more affordably in as a group. So I would expect the Internet2 crew to be asked to tackle even more future tasks as new technologies emerge that are needed by its university members.

There is one area where Internet2 has not yet extended its cheap bandwidth – which is to rural broadband networks. One of the biggest hurdles that rural America still faces is getting access to affordable bandwidth. Often, the only place to buy bandwidth in rural areas is through premium prices paid to the telco or cable company in the area. But the Internet2 network already extends to many of these rural towns and counties. I routinely find when I visit a rural community that the only cheap bandwidth in town is at the library or at city hall. But this ‘government’ bandwidth is never made available to serve the residents of these communities.

There would be a huge social benefit if Internet2 and its affiliates would allow rural ISPs to become anchor institutions on the extensive Internet2 network. The folks at Internet2 tell me that this is something they have considered, and I hope they give it more thought. They have an opportunity to help rural America in the same way they have benefitted our schools and universities – and rural America sure could use the help.