The Congressional Research Service (CRS), which is a nonpartisan research arm of the Library of Congress, recently updated a report titled Low Earth Orbit Satellites: Potential to Address the Broadband Digital Divide.

The document seems to be a primer of satellite broadband, which I assume is for folks on Capitol Hill. It’s not a bad primer, and it lays out the latest industry understanding of the LEO satellite industry. One of the challenges CRS will have, which we all share, is keeping the information in this type of paper up to date. It seems like the facts surrounding LEO satellites shift significantly every six months.

The paper is nonpartisan in that it doesn’t take a specific stance on whether satellite broadband solves the rural digital divide. Interestingly, the paper explores this issue by diving into the comparative cost of deploying fiber versus satellite. The paper acknowledges that it’s cheaper to reach remote rural customers with satellite than with fiber. But the paper also acknowledges the high recurring cost of fully replacing satellites every five years. CRS even speculates that if LEO providers don’t meet business plan goals, they might eventually pare back on replacing satellites.

In making the cost comparison between satellite and fiber, the paper makes a common error and assigns a useful life to fiber of 20-25 years. That low expected life was introduced more than twenty years ago when the IRS suggested that fiber can be depreciated over twenty years. That tax ruling has nothing to do with the actual useful life of fiber. It turns out that the life of a specific fiber mostly relates to how often it gets cut over time. While some fiber from the 1980s has clearly deteriorated, there are still plenty of examples of fiber from forty years ago still chugging along. There have been huge improvements in fiber manufacturing and installation methods, which means that fiber deployed properly today might easily still be working 60-75 years from now – we’ll have to wait to find out. Anyway, comparing fiber costs over a more realistic timeline makes fiber look even better as a long-term investment.

The paper makes an interesting statement that fiber is used to deliver broadband with symmetrical speeds between 200 Mbps and 20 Gbps. The only ISP I can find that offers symmetrical 20 Gbps service to residents is GFiber. I’d love to hear if there are ISPs doing this. But those cited speeds miss the bigger picture. ISPs routinely buy sell 100 Gbps and even 400 Gbps connections to large businesses and carriers. There are working field trials of terabit fiber.

The paper raised the question if there will be any need for future broadband grants if LEO satellite is considered as broadband. The paper uncomfortably cites me as the genesis of that idea due to a blog that asked that question. Anybody who has been following my blog knows that I think that there will be many millions of homes missed by BEAD, and I have to wonder if satellite will have the capacity to add the millions of additional customers it will take to fully solve the rural digital divide.

The paper also tries to address the satellite cellular issue and wonders if technology improvements might negate the need for some federal programs that provide support for the build-out of terrestrial cellular networks, such as the 5G Fund for Rural America. The paper does acknowledge that many industry experts question the ability of satellites to handle the volume of traffic handled by cell sites. It’s also going to take some interesting technology to allow people to communicate with satellites from indoors and from moving cars.

Overall, this paper ought to be useful for the folks on Capitol Hill. It’s a daunting task to go to the web to answer even basic questions about the state of LEO and the broadband industry.