Power and Telecommunications Resilience: What the FFIEC BCM Booklet Says About Generators, Carriers, and Testing

Correction (updated October 13, 2026): An earlier version of this post described a “March 2026 update” to the FFIEC Business Continuity Management booklet and attributed specific generator-testing and utility-diversity requirements to it. No such update exists. The booklet’s power (IV.A.7) and telecommunications (IV.A.6) sections date from its November 2019 release and set no testing cadence. This post has been rewritten to reflect what the booklet actually says.

TL;DR

• Power (IV.A.7) and telecommunications (IV.A.6) have had their own sections in the FFIEC Business Continuity Management booklet since it was issued in November 2019. There was no 2026 rewrite of either section.
• IV.A.7 tells management to consider alternate energy sources, fuel, generator load capacity, maintenance, and “Testing of generators.” It sets no cadence, no load percentage, no duration.
• IV.A.6 is more specific: diversified connections, last-mile single points of failure, carriers’ physical paths, and the federal TSP, GETS, and WPS priority programs.
• Cadence is your policy’s job. Write a generator and telecom testing standard, follow it, and keep the logs. That’s what examiners can actually hold you to.

The FFIEC’s entire section on power resilience is one short paragraph and five bullets. That’s not a loophole. It’s the design of the booklet: it tells you what to think about, then expects your own policy to say how often and how hard you test it, and expects you to prove you did.

Which is why “we have a generator” fails so often in practice. The gap usually isn’t hardware. It’s that nobody wrote down which systems the generator carries, how long the fuel lasts, when it was last run under real load, or whether both “redundant” internet circuits leave the building through the same conduit.

Here’s what the booklet actually says, and how to turn its “consider” lists into a testing program you can defend.

What the FFIEC BCM Booklet Actually Says (and When It Said It)

The FFIEC issued the Business Continuity Management booklet in November 2019. The FDIC’s FIL-71-2019 notes it replaced the Business Continuity Planning booklet from February 2015 and that “The changes do not impose new requirements on examined entities.”

Power and telecommunications sit under Section IV.A, Resilience, alongside physical, cyber, data backup, personnel, third-party, and change-management resilience. Both sections have been there since 2019. If you’ve seen claims of a 2026 revision that added generator test cadences or utility-independence mandates, they don’t match the booklet text.

IV.A.7 Power: the full list

The booklet opens by noting that “A long-term power outage can negatively impact an entity’s resilience,” then says:

“Management should implement measures to provide electricity in the event of a short-term power disruption. Furthermore, management should develop plans to provide electricity in the event of a long-term power disruption. As part of its short-term and long-term plans, management should consider the following: Alternate energy sources (e.g., generators, multiple power grids). Fuel requirements, both for fuel on-hand and contracts with suppliers for deliveries during events, and any potential impediments to obtaining fuel. Load capacity of generators (e.g., length of time, useful life, level of power supplied). Continued maintenance of generators. Testing of generators.”

That’s it. No quarterly test, no annual four-hour run, no load percentage. The Appendix A examination procedures mirror the list: examiners “Determine whether management considers the following as part of the entity’s power resilience strategies,” ending with “Testing of generators.”

Two other passages fill in the picture:

• III.A.2 Interdependency Analysis tells management to identify single points of failure, which “may include telecommunication lines, network connections between branches, backups that become corrupted, reliance on one power source, or data center locations in close geographic proximity.”
• VII.G.4 Tests lists characteristics of a successful test, including “Certifying facility controls (e.g., environmental, backup power, and physical security).”

IV.A.6 Telecommunications: more specific than you’d expect

The telecom section does more of the work for you. It opens: “Given the critical nature of telecommunications, management should ensure appropriate redundancy levels in the entity’s telecommunications infrastructure.” It warns that the infrastructure “may contain single points of failure that are outside the control of a single entity,” and that “multiple carriers may rely on the same telecommunications backbone.”

The key aspects management should consider include:

• “Identifying and mitigating single points of failure across the entity’s infrastructure.”
• “Establishing redundant telecommunications links with each of the entity’s third-party service providers through a contractual arrangement, which allows either party to switch its connection to an alternate communication path.”
• “Developing guidelines, commensurate with the entity’s size, complexity, and risk profile, to diversify connections to mitigate the risk of a telecommunications failure.”
• “Assessing the communications technology that bridges the transmission distance between the telecommunications service provider and the entity, sometimes referred to as the ‘last mile,’ for single points of failure.”
• “Inquiring about the physical paths used by telecommunications providers and verifying that system redundancies have been properly implemented.”

Then it points to three federal priority programs, which give participants priority access to telecommunications during a widespread event:

• Telecommunications Service Priority (TSP) — “TSP-designated circuits are recovered first in an emergency.” The booklet says management may contact its primary federal regulator about whether the entity qualifies, and if it does, “management should integrate the TSP program into the entity’s BCP.”
• Government Emergency Telecommunications Service (GETS) — priority calling on congested landline networks. The booklet says management may request GETS cards by applying through the entity’s primary federal regulator.
• Wireless Priority Service (WPS) — the wireless complement to GETS.

If your BCP doesn’t say whether you’ve evaluated TSP eligibility or who holds GETS/WPS access, that’s a cheap gap to close.

For broader context on how the booklet is organized, see our FFIEC BCM requirements breakdown.

Generator Testing: Building the Cadence the Booklet Leaves to You

Because IV.A.7 stops at “Testing of generators,” the cadence comes from your own exercise and test policy. Section VII.B says that policy should “Establish minimum frequency, scope, and reporting requirements.” So write one for power, then follow it.

What follows is common practitioner practice, not FFIEC text:

Start with what facilities already does. Most generators are already maintained to the manufacturer’s schedule or to NFPA 110, the Standard for Emergency and Standby Power Systems, where local code adopts it. Your facilities team or generator vendor probably has logs. The BCM gap is usually that those logs never make it into the BCP evidence file.

Routine exercise runs under real load. Starting the generator and letting it idle proves it starts. It doesn’t prove the automatic transfer switch moves your actual critical load. Set a routine cadence (monthly is common) and make sure at least some runs carry real or simulated load.

A periodic extended run. Once a year or on whatever interval your risk assessment supports, run long enough to see how the unit behaves under sustained load. A load bank is the usual answer when you can’t put production on it.

Capacity and fuel math. IV.A.7 explicitly calls out load capacity (“length of time, useful life, level of power supplied”) and fuel, including “contracts with suppliers for deliveries during events, and any potential impediments to obtaining fuel.” Document which systems each generator supports, run time on a full tank at current load, and who delivers fuel when every other business in the region is also calling.

Document results and remediation. Date, duration, load, systems carried, pass/fail, issues found, and who’s fixing what by when. A failed test with a closed remediation item is better evidence than a perfect log nobody believes.

Telecommunications Resilience: What Redundancy Actually Means

Most institutions have two internet connections. IV.A.6 is essentially a checklist for asking whether those two connections are actually two.

Carrier diversity. Two circuits from the same carrier fail together when that carrier has a regional problem. The booklet’s warning that “multiple carriers may rely on the same telecommunications backbone” cuts the other way too: two carriers can still share infrastructure. Ask.

Physical path diversity. The booklet tells you to inquire “about the physical paths used by telecommunications providers.” In practice that means asking each carrier which central office serves you and how the circuit enters the building. Two carriers sharing a conduit can be cut by one backhoe.

The last mile. The segment between the carrier and your building is called out by name. That’s often where “diverse” circuits converge.

Third-party connections. IV.A.6 asks for “a plan to address an outage in the telecommunications lines with the entity’s primary third-party service providers” and redundant links with each of them. If your core processor connection has one path, that’s your single point of failure, not your internet circuits.

Out-of-band coordination. V.F.3 says crisis or emergency management plans “should address simultaneous disruptions of telecommunications and electronic messaging, including between the entity and third-party service providers.” If incident coordination runs entirely on Teams or Slack over your primary network, a network outage takes out your recovery channel at the same time. Our communications failure BCP framework covers how to build that backup layer.

Alternate Site Power and Telecom: The Analysis Examiners Can Ask About

The booklet doesn’t say an alternate site must be on a different utility feed. What it does say is enough to make the analysis worth doing:

• IV.A.1 says management, “when possible, should diversify telecommunication lines, establish redundant connections between branches and data centers, create backups, identify multiple power sources, and verify geographic diversity of key entity locations.”
• III.A.2 lists “reliance on one power source” and “data center locations in close geographic proximity” as single points of failure to identify.
• III.B.2 asks management to assess whether third-party providers consider “the proximity to critical infrastructure (e.g., power grids, telecommunications…).”

The practical move: document the electric utility, telecom carriers, and serving central office for both your primary and alternate sites, and note anything they share. For community banks and credit unions that recover to another branch or a contracted hot site, the honest answer may be “we don’t know.” That’s the gap.

If your alternate site is a vendor’s hot site, ask the vendor for its power and telecom diversity documentation. If it can’t provide any, record that in your third-party risk assessment and decide whether it’s an accepted risk or a contract conversation.

Exercise vs. Test: Use the Booklet’s Definitions

Programs often run tabletops and log them as “tests.” The booklet’s own definitions make the difference clear.

From Section VII: “An exercise is a task or activity involving people and processes that is designed to validate one or more aspects of the BCP or related procedures.” And: “A test is a type of exercise intended to verify the quality, performance, or reliability of system resilience in an operational environment.” Tests “use quantifiable metrics,” and the booklet’s own example is “what happens as a result of removing power from a system or system component.”

Section VII.G lists four methods: full-scale exercise, limited-scale exercise, tabletop exercise, and tests. On tabletops, VII.G.3 is blunt: “By themselves, tabletop exercises are likely insufficient to validate recovery capabilities, because they are limited to a discussion-based analysis of policies and procedures.”

So a generator transfer under load or a telecom circuit cutover is a test. A conversation about what would happen if the power went out is an exercise. Both are useful. Only one shows the generator works.

If you only run tabletops, see our business continuity testing guide for how to layer in tests without breaking production.

Documentation: Build the Evidence File Before Anyone Asks

The booklet doesn’t hand you a document list for power and telecom. But between the IV.A.6/IV.A.7 “consider” lists, VII.B’s policy requirements, and the Appendix A examination procedures, a good evidence file usually looks like this:

• Your exercise and test policy, with the minimum frequency you set for generator and telecom tests
• Generator inventory, systems supported, and capacity/fuel analysis
• Generator maintenance and test logs showing your policy’s cadence was actually met
• Fuel supplier contracts and delivery assumptions for regional events
• Telecom carrier list, serving central offices, and physical-path documentation for primary and alternate sites
• TSP eligibility decision and GETS/WPS holder list
• Failover test results, including anything that didn’t recover
• Remediation tracking through closure

If your policy says quarterly and the logs show two runs last year, the finding writes itself. A modest cadence you actually follow is stronger than an aggressive one you don’t.

The BCP/DR Kit gives you a place to put this: the Dependency Mapping tab records which processes rely on which power and telecom paths and flags single points of failure, the Test & Exercise Log records generator and failover tests, and the Emergency Contacts tab holds your utility and carrier contacts with stale flags. The Tabletop Exercise Kit’s regional hurricane scenario takes power and telecommunications out for days.

So What? 5-Step Action Plan

Step 1: Check your policy for a number. Does your exercise and test policy set a minimum frequency for generator and telecom failover testing? If not, VII.B says it should. Pick a cadence you can sustain and write it down.

Step 2: Pull your generator logs into the BCP file. Get the maintenance and test records from facilities or your vendor. Compare them against your policy’s cadence. Fix gaps before an examiner finds them.

Step 3: Ask your carriers the IV.A.6 questions. Which central office serves each circuit? Do the physical paths share a conduit or entry point? What’s the last-mile setup? Get the answers in writing.

Step 4: Map the alternate site. Document electric utility, carriers, and central offices for your recovery site and compare them to your primary. Record what’s shared and whether you accept it.

Step 5: Settle TSP, GETS, and WPS. Decide whether you qualify for TSP, and confirm who holds GETS and WPS access. Then add an out-of-band coordination channel to your incident plan. The crisis communication plan framework covers how to document it.

Power and telecom are the dependencies everything else sits on. The FFIEC gave you a short list and left the cadence to you. Use that freedom to write a standard you’ll actually meet, then keep the receipts.