The 470 to 512 MHz band carries police, fire and EMS radio systems in a small number of the country’s largest metropolitan areas, and for nearly nine years federal law required those agencies to leave it. Congress repealed that requirement in December 2020, by which time the FCC had frozen new applications, several major cities had spent a decade defending their spectrum instead of replacing their systems, and the vendors who built that equipment had moved on. This piece covers what the band is, what the statute actually said, what the repeal fixed and did not fix, and how to write a capital plan that survives a policy reversal.

What the T-Band is and where it exists

The T-Band is a set of land mobile channels overlaid on UHF television channels 14 through 20, which together occupy 470 to 512 MHz. It was never a clean public safety allocation with its own national channel plan, because the television service was already there and stayed there. A repeater and a co-channel television station cannot occupy the same geography, so the FCC’s Part 90 rules make these channels available only inside a table of named urbanized areas, and every land mobile assignment has to sit at a specified distance from co-channel and adjacent-channel television facilities. Repeater pairs in the band are conventionally offset by 3 MHz, which is why a T-Band system’s input and output frequencies look so far apart compared with a 450 to 470 MHz system.

The markets usually named in public safety documents are Boston, Chicago, Dallas and Fort Worth, Houston, Los Angeles, Miami, New York and northeastern New Jersey, Philadelphia, Pittsburgh, San Francisco and Oakland, and the Washington, DC region, which is the eleven-market count that appeared throughout the relocation debate. The underlying rule table in Part 90 lists urbanized areas for land mobile use in this band, and depending on which document you read the count comes out at eleven or thirteen, because not every listed area carried public safety systems and some carried only industrial and business licensees. If the exact list matters to your project, read the current table in the FCC rules rather than relying on any secondary source, including this one.

What makes the band consequential out of proportion to its size is who lives in it. These are the densest radio environments in the United States, with fleets in the tens of thousands of subscribers in some cities, and in several of those markets the T-Band channels are not a supplement to a 700 or 800 MHz system but the primary dispatch and tactical spectrum. That concentration is exactly why a national statute written to raise auction revenue landed so hard on a handful of agencies.

How public safety ended up in the television band

Land mobile radio in the big cities ran out of room early. By the late 1960s the VHF high band and the 450 to 470 MHz channels were saturated in the largest markets, and the UHF television band had allotments that were unused in many of those same markets because of the spacing rules broadcasting required. Around 1970 the FCC decided to let land mobile share the lower UHF television channels in the most congested urban areas, subject to protecting the television stations that were on the air. The Commission’s own records are the place to read that history in detail, and I would rather point you there than recite a docket number from memory.

Agencies took the channels because the channels existed. Over the following four decades those systems grew into trunked networks, then into narrowband digital networks, with site grids, microwave backhaul, dispatch consoles and subscriber fleets all engineered around 470 to 512 MHz propagation and around the specific frequencies the licensee held. A radio system is not a portfolio of frequencies that can be swapped out; it is a set of antennas, filters, combiners, duplexers, site leases and coverage predictions built for a particular slice of spectrum, and the further you move from that slice the less of the investment travels with you.

The physics matter for any relocation discussion. Moving a system up in frequency, from the low 500s to 700 or 800 MHz, generally costs you in-building penetration and increases path loss, which in practical terms means more sites to hold the same in-building coverage over the same footprint. That is before you ask whether the destination band even has channels available in a market like New York or Los Angeles, where the 700 and 800 MHz inventory was already heavily assigned.

Section 6103 and the nine-year clock

The Middle Class Tax Relief and Job Creation Act of 2012 became Public Law 112-96 when it was signed on February 22, 2012. Its spectrum title is best known for creating the First Responder Network Authority and providing $7 billion toward a nationwide public safety broadband network, funded in part by anticipated spectrum auction proceeds. Section 6103 of that act dealt with the T-Band, and it directed the FCC to reallocate the 470 to 512 MHz spectrum then used by public safety eligibles and to begin competitive bidding for it no later than nine years after enactment, which put the auction deadline in February 2021. Public safety users were then to relocate within two years after the proceeds became available, with relocation costs to be covered from the auction receipts deposited in the Public Safety Trust Fund.

The structure of that mandate is the part worth understanding, because it explains why the affected agencies could not simply start planning a move. The statute did not appropriate relocation money up front and it did not identify replacement spectrum. It assumed an auction would raise enough to pay for the move, and it made the funding contingent on receipts nobody could size in advance. An agency in Boston or Philadelphia asking a straightforward question in 2013, which was where its radios would operate in 2023 and who would pay for the change, had no authority it could ask for an answer.

The FCC’s Public Safety and Homeland Security Bureau responded in 2012 by freezing acceptance of applications for new or expanded T-Band facilities, which is the ordinary regulatory move when a band is slated for reallocation and the Commission does not want new encumbrances appearing in it. For the licensees, the freeze converted a distant statutory deadline into an immediate operational constraint, because adding a site to fix a coverage hole, or modifying a license to reflect a system change, became a matter of asking for relief rather than filing a routine application.

The bill for a move nobody funded

The National Public Safety Telecommunications Council studied the problem and published its T-Band report in 2013. NPSTC put the national cost of relocating the affected public safety systems at roughly $5.9 billion, and that figure became the anchor for the entire policy argument that followed. The second finding in that report mattered as much as the price. NPSTC concluded that sufficient alternative spectrum did not exist in most of the affected markets, and in some of them there was effectively none, which meant the money question was not the only obstacle. A funded move requires somewhere to move to.

Treat the $5.9 billion figure as what it is, which is a 2013 estimate produced by a public safety organization advocating against relocation, built on assumptions about site counts, subscriber replacement and destination bands that were reasonable at the time and are now more than a decade old. I have not seen a competing government estimate of comparable scope, and I am not going to manufacture a range where I cannot name who publishes each end of it. If you need a current number for a specific market, it has to come from an engineering study of that market rather than from a national average.

What Congress expected the band to fetch at auction was never tested, since the auction never happened. The comparison that did get made repeatedly during the debate was to 800 MHz rebanding, the Nextel-related reconfiguration the FCC ordered in 2004, which came with an identified payer and still took many regions well over a decade to complete. T-Band relocation was ordered with neither an identified payer nor identified destination channels, which is the difference that made the mandate unworkable rather than merely difficult.

The repeal in December 2020

Legislation to undo the mandate was introduced in successive Congresses without passing, under the name the Don’t Break Up the T-Band Act. The version that succeeded was the Don’t Break Up the T-Band Act of 2020, enacted as part of the Consolidated Appropriations Act, 2021, Public Law 116-260, which was signed on December 27, 2020. It repealed section 6103 of the 2012 act, which removed the reallocation and auction requirement and with it the two-year relocation clock that would have started once proceeds were available.

The repeal arrived about two months before the statutory auction deadline of February 2021, which tells you how close the affected agencies came to a formal reallocation proceeding. It is worth being precise about what the repeal accomplished, because it is easy to describe it as a win in terms broader than the text supports. It ended the obligation to leave the band. It did not appropriate money, it did not create new public safety channels anywhere, it did not change the sharing relationship with UHF television, and it did not give back the planning years that the mandate consumed.

For the agencies involved, the practical effect was that a long-term investment in T-Band infrastructure became defensible again. A radio manager in one of these markets could finally answer the capital committee’s question about which band the next system would operate in, and could do it without attaching a footnote about pending federal legislation. That answer arrived in 2021 for systems that in many cases had been due for replacement since the middle of the previous decade.

What the uncertainty did to the replacement cycle

A public safety radio system runs on a capital cycle measured in decades, and the decision points inside that cycle are few and heavy. A city commits to a system architecture, spends five to ten years paying for it, runs it for fifteen or twenty with midlife upgrades, and starts planning the successor while the current one is still working. Nearly nine years of statutory uncertainty removed one entire decision point from that cycle in the affected markets, because no elected body will approve a bond issue for infrastructure in a band that federal law says must be vacated, and no responsible radio manager will recommend one.

The observable damage shows up in a few specific places. Infrastructure that should have been replaced on schedule ran past the manufacturer’s published end-of-support dates, which changes the failure mode from a repair to a scavenging exercise, because the parts come from a decommissioned system somewhere else or from a third-party broker. Subscriber fleets aged past the point where the model is still built, so the agency buys refurbished portables to keep the fleet consistent instead of migrating to current hardware. Coverage problems that would ordinarily be fixed by adding a site went unfixed or were addressed with in-building solutions and vehicular repeaters, because a new T-Band site required navigating an application freeze.

There is a staffing dimension that gets overlooked. Nine years is longer than many people stay in a radio shop or a communications director’s chair, so the institutional memory of why the system was frozen leaves the building along with the person who negotiated the freeze. I have watched much shorter gaps do this. A project that stalls for three years comes back as a project nobody in the room designed, with assumptions in the requirements document that no current employee can explain and a coverage study whose input data predates the last round of construction in the service area.

Money did not sit still either, and this is the part that turns a policy delay into a permanent cost. Deferred capital work becomes deferred maintenance, deferred maintenance becomes emergency repair, and emergency repair is bought at whatever price and schedule the vendor offers because there is no time to compete it. An agency that entered 2012 with a fifteen-year-old system and a funded replacement plan came out of 2020 with a twenty-four-year-old system and a plan that had to be written from scratch against 2021 prices.

Where the band stands, and what to verify with the FCC

As of this writing on September 2, 2026, the statutory relocation requirement is repealed and public safety operation in 470 to 512 MHz continues under the ordinary Part 90 rules in the urbanized areas where the band is available. That is the durable part of the picture. Nearly everything else about the band is a matter of current FCC administration rather than statute, and it can change without Congress doing anything, which is precisely why I am dating the statement.

The application freeze imposed in 2012 was addressed by the Commission after the repeal, and the filing posture for new and modified T-Band facilities is whatever the Public Safety and Homeland Security Bureau says it is today. If you are designing a site addition, a system expansion or a license modification in this band, confirm the current status with the Bureau or with your frequency coordinator before you spend engineering money, and get the answer in a form you can attach to the project file. The same goes for the interaction with television. The incentive auction and the repack that followed it moved stations around inside the remaining television channels, and the channel-by-channel occupancy above and below your assignments in your specific market is a coordination question with a market-specific answer.

The broader lesson generalizes past this band. The 4.9 GHz public safety band has been through more than one FCC decision and reversal since 2020, and agencies that built plans on any single one of those orders learned the same thing the T-Band licensees learned. Spectrum policy is not a settled input to your engineering; it is a variable with its own schedule, and the only honest way to handle it in a plan is to name the assumption, date it, and identify who checks it.

Writing a capital plan that survives a policy reversal

The technique that works is separating spending that is band-dependent from spending that is not. Towers, shelters, grounding and lightning protection, generators and battery plant, fiber and microwave paths, dispatch console positions, logging recorders and the site security you should have already fixed are all things you will need regardless of which band the radios end up in, and most of them carry forward through a technology change with modest modification. Antennas, feedline, combining and filtering, base station RF decks and the subscriber fleet are the band-dependent items. When a policy question is genuinely open, put the money into the first list and hold the second list at the design stage rather than stopping the whole program.

Write the policy assumption down as an explicit line item with an owner. One paragraph in the capital plan naming the assumption, the authority it rests on, the date it was verified and the person who rechecks it annually is enough, and it converts an invisible risk into an agenda item. The governance body that already reviews your radio system, whether that is a county communications committee, a regional interoperability board or a single department’s staff meeting, is the right place for that review, because it needs to happen in front of the people who control the money.

Keep the license clean while you wait. An agency in a contested band that lets construction deadlines lapse, fails to renew on time, or lets its authorized parameters drift out of alignment with what is actually on the air has weakened its own position for reasons that have nothing to do with the policy fight. Pull your own records from the FCC’s licensing system, compare the authorized emission designators, power levels, antenna coordinates and heights against the equipment as installed, and fix the differences before somebody else finds them.

Finally, be honest in grant applications about which parts of the project are contingent. Federal and state grant programs generally have periods of performance measured in a few years, and a project whose spectrum future is genuinely unsettled will not complete inside one. The T-Band decade produced agencies that returned or reprogrammed award money because the underlying project could not proceed, which is a poor use of a competitive award and it is noticed the next time you apply.

What to do at your agency

• Have your radio manager pull your agency’s own license records from the FCC licensing system this month and reconcile the authorized frequencies, emission designators, power, antenna coordinates and heights against what is physically installed at each site, then document and correct any differences.
• Ask your infrastructure vendor in writing for the published end-of-support dates for your current system release and for every subscriber model in your fleet, and put those dates into the capital plan next to the replacement year you are currently assuming.
• Add one standing item to the agenda of the committee that already reviews your radio system, which is an annual restatement of the regulatory assumptions the plan rests on, with the date each was last verified and the name of the person who verified it.
• If you operate in a T-Band market, have your frequency coordinator or your consulting engineer confirm the current FCC filing posture for new and modified 470 to 512 MHz facilities before you commit engineering money to a site addition, and file the answer with the project record.
• Split your next five-year communications capital request into band-dependent and band-independent line items, so that a future policy change stalls the radio purchase without stalling the tower, power, backhaul and console work.
• Have whoever writes your grant applications review any open award for tasks that depend on an unresolved spectrum question, and either resequence those tasks inside the period of performance or raise the issue with the grant administrator now rather than at closeout.
• Write one paragraph into your existing communications plan recording what band each of your systems operates in, what the regulatory basis for that operation is, and the date the paragraph was written, so the next person in your chair does not have to reconstruct it.

Takeaways

• The T-Band is land mobile spectrum at 470 to 512 MHz overlaid on UHF television channels 14 through 20, available only in a table of urbanized areas listed in the FCC’s Part 90 rules and subject to distance separations that protect television stations.
• Section 6103 of the Middle Class Tax Relief and Job Creation Act of 2012, Public Law 112-96, signed February 22, 2012, required the FCC to reallocate the band and to begin competitive bidding within nine years, with public safety relocating within two years after auction proceeds became available.
• The mandate identified no replacement spectrum and provided no up-front funding, making relocation contingent on auction receipts that were never tested because the auction never occurred.
• The National Public Safety Telecommunications Council’s 2013 T-Band report estimated the national relocation cost at roughly $5.9 billion and concluded that sufficient alternative spectrum did not exist in most of the affected markets, and that figure should be read as a 2013 advocacy-side estimate rather than a current price.
• The FCC’s narrowbanding deadline of January 1, 2013 applied to the T-Band, so the same licensees were required to spend money modernizing equipment in a band that statute told them to abandon.
• The Don’t Break Up the T-Band Act of 2020, enacted in the Consolidated Appropriations Act, 2021, Public Law 116-260, signed December 27, 2020, repealed section 6103, which ended the relocation requirement without appropriating money, creating channels, or restoring the lost planning years.
• Nearly nine years of uncertainty pushed infrastructure past manufacturer support, aged subscriber fleets past production, deferred coverage fixes behind an application freeze, and cost the affected agencies institutional memory through staff turnover.
• Spectrum policy is a dated variable rather than a fixed input, so every regulatory statement in a capital plan needs its source and its verification date, and the current status of anything in this band should be confirmed with the FCC’s Public Safety and Homeland Security Bureau before money is committed.