DHS’s 2018 cybersecurity plan could direct federal agencies, and only offer help to everyone else
DHS’s May 2018 cybersecurity plan relied on four kinds of power: directing, offering, investigating and shaping. Only directing reached inside another organization’s systems, and outside two regulated sectors it covered only federal civilian agencies. For states, counties and public hospitals, it was a menu.
Looking Back revisits past weeks one at a time, up to the most recent, and asks what it meant for the banks, health systems and public agencies we serve. The series was written in 2026, with the benefit of hindsight, and each article says which week it looks back on.
Two-minute summaryThe dek and the first sentence of each section, quoted from the article
DHS’s May 2018 cybersecurity plan relied on four kinds of power: directing, offering, investigating and shaping. Only directing reached inside another organization’s systems, and outside two regulated sectors it covered only federal civilian agencies. For states, counties and public hospitals, it was a menu.
- The week On Tuesday, May 15, 2018, the Department of Homeland Security released a document setting out how it would identify and manage national cybersecurity risk.
- Directing: the one lever with compulsion, aimed at federal agencies Goal 2 opened with a strong verb.
- Offering: services, information and help for everyone else Goal 3 used different verbs.
- Investigating: law enforcement as help that does not direct the victim Goal 4 aimed at criminals, not at victims or defenders.
- Shaping: the ecosystem and DHS’s own house Goals 6 and 7 aimed at conditions rather than organizations.
- Why the split followed the law, not the risk The four kinds of power did not track where risk sat.
- Where state and local governments sat The strategy’s goals named two protected groups: federal agencies and critical infrastructure.
- What the strategy promised to measure Each objective ended with an outcome statement.
- How the split fell on different public organizations For a federal civilian agency, the strategy described a supervisor and a service provider in one department.
- Objections to reading the strategy as a menu Four objections deserve a reply.
- What a government technology leader could check The strategy’s four kinds of power translate into a short set of checks for any public organization.
- With hindsight Later in 2018, the department’s cyber component got a new name and statutory footing.
The week
On Tuesday, May 15, 2018, the Department of Homeland Security released a document setting out how it would identify and manage national cybersecurity risk. The department’s announcement said it was directed by the National Defense Authorization Act of 2017. It said the document addressed “the full range of the DHS Secretary’s cybersecurity responsibilities,” and was meant to harmonize and prioritize planning, programming, budgeting and operations across the department’s cybersecurity mission areas. The document is dated May 15, 2018, and its scope section said it gave the department a framework for its cybersecurity responsibilities “during the next five years.”
It organized the department’s work into five pillars and seven goals. Pillar I, risk identification, held Goal 1: assess evolving cybersecurity risks. Pillar II, vulnerability reduction, held Goal 2, protect federal government information systems, and Goal 3, protect critical infrastructure. Pillar III, threat reduction, held Goal 4: prevent and disrupt criminal use of cyberspace. Pillar IV, consequence mitigation, held Goal 5: respond effectively to cyber incidents. Pillar V, enable cybersecurity outcomes, held Goal 6, strengthen the security and reliability of the cyber ecosystem, and Goal 7, improve management of DHS cybersecurity activities.
Goal 1 had a single objective: “Maintain strategic awareness of trends in national and systemic cybersecurity risks.” The document said understanding risk at the strategic level would let DHS allocate resources and prioritize across vulnerabilities, threats and consequences.
Each goal carried objectives, sub-objectives and stated outcomes. Goal 2’s objectives were to improve federal governance, policy and oversight, to provide protective capabilities, tools and services across the federal enterprise, and to protect DHS’s own systems. A footnote said DHS’s authority to administer agency information security policy applied to federal executive branch systems, except national security systems and certain Defense and intelligence community systems. Goal 3 said DHS served as sector-specific lead or co-lead for 10 of the 16 critical infrastructure sectors, and listed all 16. It said DHS had regulatory authority for the chemical and transportation sectors and would “mature existing voluntary and regulatory partnerships.”
Goal 4 said the components with “the most direct responsibility related to countering illicit cyber activity” included the U.S. Secret Service and Immigration and Customs Enforcement’s Homeland Security Investigations. Goal 5’s first objective was to “Increase voluntary incident reporting and victim notification.” Goal 6 covered secure technology, research, international collaboration and the cyber workforce. Goal 7 covered the department’s own policy integration and program evaluation.
An appendix listed the department’s cybersecurity authorities. They included the Homeland Security Act provisions for the National Cybersecurity and Communications Integration Center, the Cybersecurity Information Sharing Act of 2015 and the Federal Information Security Modernization Act of 2014. The appendix said that through the National Protection and Programs Directorate, DHS issued binding operational directives to agencies. It said one Homeland Security Act provision mandated that all agencies “apply and continue to utilize” DHS intrusion detection and prevention technologies. It described other provisions as authorizing analysis and warnings to state and local governments and, upon request, technical assistance to owners and operators of critical information systems. It also listed seven guiding principles, from risk prioritization and cost-effectiveness to collaboration and national values.
A department strategy is written for the department. Read from outside, it is also a statement of what the department can do to the reader, for the reader, and about the reader’s adversaries. The May 2018 DHS strategy was clear on that point, because its appendix listed the legal authority behind each kind of activity. Set the goals beside the authorities and a pattern appears.
We sort the strategy’s seven goals by the kind of power each relied on. Four kinds cover every goal without overlap:
- Directing: compulsory requirements DHS could place on the organizations whose systems it was protecting. In the strategy, that power applied to federal civilian executive branch agencies under Goal 2.
- Offering: assessments, services, information and response help that other organizations could accept or decline. Goals 1, 3 and 5 relied on it for everyone outside the federal civilian enterprise.
- Investigating: law enforcement action against criminals, under Goal 4, which helped victims without directing them.
- Shaping: work on the wider ecosystem and on DHS itself, under Goals 6 and 7, which reached other organizations only indirectly.
Our argument is that only the first kind gave DHS a say over how another organization ran its systems, and outside DHS’s chemical and transportation regulation it covered only federal civilian agencies. For every bank, hospital, state agency and county in the country, the strategy was a menu. That made the reader’s choices decisive: whether to take what was offered, and what to build for itself. After the four kinds of power we ask why the split followed the law rather than the risk, and where state and local governments fit. We then look at what the strategy promised to measure, how the split fell on different public organizations, the objections, and what a leader can check.
Directing: the one lever with compulsion, aimed at federal agencies
Goal 2 opened with a strong verb. DHS “must use all available mechanisms to ensure that every agency maintains an adequate level of cybersecurity, commensurate with its own risks and with those of the larger enterprise.” It acknowledged that agencies must implement their own risk management programs, because they understand their missions. It also said DHS worked with the Office of Management and Budget to ensure adequate security across the enterprise.
The mechanisms behind that verb were listed in the appendix. The Federal Information Security Modernization Act of 2014 gave DHS a central role in securing federal executive branch information systems. Through it, DHS administered government-wide policies, deployed protective technologies and issued binding operational directives. A separate provision, added by the Federal Cybersecurity Enhancement Act of 2015, mandated that all agencies apply and continue to use DHS intrusion detection and prevention technologies. Those are requirements. An agency could not opt out of them.
The objectives under Goal 2 described how DHS meant to use that power. Objective 2.1 called for a clear governance model, new or revised policies, a formal approach to measure and track agency adoption of required controls, and processes to increase agency accountability and compliance. Objective 2.2 called for identifying elements of the federal enterprise that should be centralized and deploying centralized protective capabilities, with tools offered on a reimbursable basis or without reimbursement. The language was about measurement and accountability, which only makes sense where the measurer can require something.
The boundary of the directing power
The strategy drew the boundary itself. Its footnote limited DHS’s authority over agency security to federal executive branch systems, excluding national security systems and certain Defense and intelligence systems. Congress, the courts, states, localities and every private organization sat outside it. A federal agency’s chief information security officer reading the strategy would recognize its directives. A state chief information officer would recognize none.
For federal agencies, that boundary had a practical edge. Directives applied to agency information systems, which can include systems run by contractors on an agency’s behalf. A contractor hosting an agency system could therefore inherit DHS requirements through its contract, even though the strategy did not direct it. That indirect reach is the closest the directing power came to organizations outside government.
What directing could and could not buy
A directive can require an action and a report. It cannot supply the staff, money or system knowledge to act. The strategy acknowledged this by pairing Objective 2.1’s accountability language with Objective 2.2’s tools and services. In our reading, the pairing was the strategy’s recognition that compulsion without capacity produces reports of non-compliance. For an agency, the useful question in 2018 was which DHS services came bundled with each requirement it faced, and which it would have to fund itself.
Directing through shared services and procurement
Objective 2.2 pointed to a second route by which directing power worked. DHS would identify elements of the federal enterprise that should be centralized and deploy centralized protective capabilities. A capability that DHS runs for every agency, such as perimeter intrusion detection, changes each agency’s security without a separate directive. The agency’s choice is narrowed to how it integrates the service, not whether it accepts it.
That route carries a dependency. When a central service fails, is late or does not fit an agency’s environment, the agency inherits the gap. The strategy’s pairing of directives with services meant an agency’s security posture was partly a function of DHS’s delivery. A federal chief information security officer in 2018 had to track both: the requirements DHS set and the services DHS promised, and the dates on which each would arrive.
Procurement was the third route. Objective 2.3 described DHS pursuing innovative acquisition approaches for its own systems and piloting capabilities that could scale across the federal enterprise. Government-wide contracts and shared services can make the secure option the default option for agencies that buy through them. In our reading, this amounts to directing by design of the market, and it reaches agencies through their contracting officers as much as through their security offices.
Reports as the currency of directing
Objective 2.1 sought “a formalized approach to measure and track agency adoption of information security policies, practices, and required controls.” Measurement runs on reports. Each directive and each centralized service generated data flowing from agencies to DHS: compliance status, scan results, remediation progress. For an agency, that reporting was itself work, and the quality of its asset inventory decided whether the reports were accurate. A directive to fix a class of vulnerability presumes the agency knows where that class exists in its systems.
That presumption is why inventory sits beneath every directing power. An agency that could not list its internet-facing assets could not prove it had remediated them. The strategy’s accountability language therefore landed, in practice, on asset management teams as much as on security offices.
Offering: services, information and help for everyone else
Goal 3 used different verbs. DHS would “partner with key stakeholders to ensure that national cybersecurity risks are adequately managed.” It would promote best practices and international standards, provide services such as risk assessments and technical offerings, and improve engagement. Objective 3.1 called for maturing cybersecurity offerings and engagements. Objective 3.2 called for expanding sharing of cyber threat indicators and defensive measures. Objective 3.3 called for improving capabilities available to sector-specific agencies, regulators and policymakers.
The appendix explained why these were offers. The Homeland Security Act authorized the integration center to provide guidance, assessments, incident response support and other technical assistance “upon request.” Another provision authorized analysis and warnings to state and local government entities and, upon request, to owners and operators of critical information systems. The Cybersecurity Information Sharing Act of 2015 gave liability protection to companies that shared indicators with DHS. Each was a reason to participate, not a duty to do so.
The strategy did note two exceptions. DHS had regulatory authority for the chemical and transportation sectors. It said it would use those regulatory authorities “in tailored ways” as a backstop against disruption of essential services. Financial services, healthcare and public health, and government facilities, the sectors this series follows, were not among the regulated ones. Their cybersecurity regulators, where they had any, sat elsewhere.
What accepting an offer costs the recipient
An offer is free at the point of delivery and costly afterward. An organization that requests a vulnerability assessment receives a list of findings that someone must fix. One that joins an indicator-sharing program receives a feed that someone must ingest, triage and act on. One that asks for incident help must give responders access, share logs and coordinate decisions. Each of these steps needs staff, tools and legal review.
The legal review is easy to underestimate. Information-sharing arrangements carry terms about what may be shared, with whom and under what protections. The Cybersecurity Information Sharing Act of 2015 offered liability protection for sharing indicators in accordance with the Act, which made the terms of sharing a matter for counsel. Public organizations also faced their own state records and privacy laws. An agency that wanted DHS help had to settle, in advance, what it could disclose.
The result is that the value of the offering power depended on the recipient’s capacity. A well-staffed state security office could turn DHS assessments and indicators into fixes and blocks. A small agency could receive the same products and lack the people to use them. The strategy did not, and could not, close that gap. It described supply. Demand and absorption were the reader’s problem.
Shared information also varies in usefulness. An indicator is valuable to a recipient only if it arrives in time, in a format the recipient’s tools can read, with enough context to decide what to do. Objective 3.2 promised to expand and improve sharing. The recipient still had to judge which feeds were worth the effort of ingestion, and to tell DHS when they were not. Feedback was part of the offer’s value, and the strategy relied on partners to provide it.
Goal 1 and Goal 5 were offers too
Goal 1, assessing evolving risks, produced analysis for DHS’s own prioritization and for stakeholders. Its value to a hospital or a county depended on whether that analysis reached them and fit their decisions. Goal 5, responding to incidents, made the offer explicit. Its first objective was to “Increase voluntary incident reporting and victim notification to facilitate the provision of response assistance.” The response DHS could give depended on what victims chose to tell it.
Objective 5.2, expanding asset response capabilities, described DHS’s technical help to victims. Objective 5.3 sought cooperation between incident responders so threat response and asset response complemented each other, following Presidential Policy Directive 41. The objectives did not oblige a nonfederal victim to call. In our reading, Goal 5 was a service with a front door, and the strategy’s measure of success depended on how far organizations chose to walk through it.
Prioritizing among those who accept
An offer has a second constraint: supply. The strategy said DHS offerings “must be prioritized to focus on systemic risk or address risk at individual entities that have the greatest potential impact” on national security, public health and safety, and economic security. It added that offerings that did not address identified gaps or give DHS unique information “should be reconsidered.” That was a reasonable way to allocate scarce services. It also meant a small health system or rural county could not assume DHS help would be available to it in particular. The strategy told readers, in effect, that the menu would be rationed by national impact.
Investigating: law enforcement as help that does not direct the victim
Goal 4 aimed at criminals, not at victims or defenders. The Secret Service and Homeland Security Investigations would combat financial and trans-border cyber crime, counter threats to protected persons and special events, build law enforcement capacity and develop investigative capabilities. Objective 4.1 named core investigative responsibilities including financial services and payment systems, computer fraud and abuse, and misuse of cryptocurrencies.
For a bank, that focus mattered directly. The appendix cited the Secret Service’s authority to investigate access device fraud, electronic fund transfer fraud and crimes against federally insured financial institutions. A bank hit by card fraud or account takeover had an investigative partner with jurisdiction over exactly those crimes. The partnership did not direct the bank’s defenses. It pursued the people who attacked them.
For state and local governments, Goal 4 carried the strategy’s clearest commitment to them. Objective 4.3 called for expanded outreach to federal, state, local, territorial and tribal law enforcement, through cyber crime centers and electronic crimes task forces. The appendix cited the Strengthening State and Local Cyber Crime Fighting Act of 2017, which authorized the Secret Service’s National Computer Forensics Institute to train state and local officers, prosecutors and judges. That was capacity building for police and courts. It did not address a county’s own networks.
What an investigation gives a victim
An investigation can identify an attacker, recover proceeds, take down infrastructure and deter repeat attempts. It can also impose demands on the victim: preserve evidence, provide logs, coordinate disclosure. The strategy’s Objective 5.3 recognized the tension by seeking complementary threat response and asset response. For a victim organization, the practical decision was when to involve law enforcement and how to balance its requests against recovery. The strategy offered the partnership. The decision stayed with the victim.
Banks, hospitals and agencies as crime victims
The investigative power treated organizations as victims, and each sector’s victim role differed. For banks, Secret Service jurisdiction over access device fraud and crimes against federally insured institutions made the partnership natural, and banks already had law enforcement liaison functions for fraud. For hospitals, a ransomware attack was a crime as well as an outage, but the hospital’s first duties ran to patients and to its own regulators. For public agencies, an intrusion could be a crime against the government itself, with prosecutors and investigators from the same level of government involved.
In each case, the strategy offered investigators who could pursue attackers across borders and link one victim’s case to others. That benefit reached victims who reported. It did not reach those who chose not to, or those whose incidents fell below investigative priorities. Here again, the reader’s decision shaped the value of the power.
Shaping: the ecosystem and DHS’s own house
Goals 6 and 7 aimed at conditions rather than organizations. Goal 6 sought to foster improved cybersecurity in software, hardware, services and technologies, to prioritize research and technology transition, to expand international collaboration and to improve recruitment, education, training and retention for a cyber workforce. Goal 7 sought to integrate department-wide policy and planning and to evaluate the effectiveness of DHS programs.
Shaping is real power, but it works slowly and indirectly. Better software and a larger workforce eventually reach every organization. They do not change a given hospital’s exposure this quarter. For a reader deciding what to do in 2018, Goals 6 and 7 described where DHS would invest its own effort. They did not describe help a specific organization could request or a requirement it would face.
Research and technology transition, under Objective 6.2, followed the same pattern. DHS would prioritize research to support its own mission objectives and move results into use. The strategy said that research should support protective capabilities for the federal enterprise and critical infrastructure and tools for law enforcement. A public organization looking for research help on its own problems would find the framing instructive: the department would invest where its missions pointed.
Goal 7 deserves one more note. It acknowledged that DHS’s own cybersecurity activities needed integration and evaluation. Objective 2.3 said something similar about DHS systems, noting that “Many DHS information systems remain largely decentralized” and run by components without a standardized approach. A department that set itself to lead federal cybersecurity was candid that its own house needed work.
Workforce as a shaping lever
Objective 6.4, improving recruitment, education, training and retention, is the shaping work that reached public organizations soonest. A larger and better-trained cyber workforce benefits every employer, including states and counties competing with private firms for the same people. The strategy’s appendix also cited the department’s authority to establish cybersecurity positions and fix rates of pay, which concerned DHS’s own hiring. Its effect on a county’s hiring was indirect. In our reading, workforce shaping helped public organizations only to the extent they could compete for the people it produced.
Why the split followed the law, not the risk
The four kinds of power did not track where risk sat. They tracked where DHS had legal authority. The directing power covered federal civilian systems because the Federal Information Security Modernization Act gave DHS a role there. It did not cover a state unemployment system, a county election office or a public hospital’s records system, however consequential an outage might be for residents.
Federal authority works that way, and the strategy did not pretend otherwise. But the result for a reader was a mismatch. Small public bodies, with little capacity to build their own programs, were outside the directing power and behind higher-impact entities in the queue for offered services. Federal agencies received both the requirements and the services. The strategy’s risk prioritization principle applied to how DHS used its offers. It could not move the boundary of its directives.
For a reader, the mismatch meant two questions had to be answered separately. Which risks does federal authority cover in my organization, and which does it leave to me. The answers do not follow from how serious the risks are. They follow from statute. A leader who assumed national attention to a risk meant national coverage of it would plan the wrong way.
Where state and local governments sat
The strategy’s goals named two protected groups: federal agencies and critical infrastructure. State and local governments appeared chiefly as law enforcement partners under Goal 4 and as recipients of analysis and warnings under the appendix’s authorities. Their own networks fell, by our reading, within critical infrastructure: “Government Facilities” is one of the 16 sectors the strategy listed. That placed a state election system, a county 911 center and a city payroll system on the offering side of the line.
That placement had consequences. A state agency could request assessments, receive indicators and ask for incident help. It could not count on being prioritized unless its systems had national impact. It faced no DHS directive and no DHS funding stream tied to the strategy. Its own legislature, auditors and budget determined what it did. The strategy did not hide this. It simply did not address state and local networks as a category with goals of its own.
For a state chief information security officer, the gap had a planning implication. The services DHS offered were real and could be valuable. But the obligation to build a program, fund it and measure it came from state law and state leadership. The DHS strategy could inform that program. It could not require or resource it.
Tribal and territorial governments were in a similar position. The strategy’s law enforcement objective named them as partners. Their networks, like those of states and localities, fell on the offering side of the line, with whatever capacity each government could bring to the services available.
Three questions a state or local leader could put to the strategy
A state or local technology leader reading the strategy in May 2018 could have asked three things of it. First, which of the listed offerings were available to a state or local government on request, and through which DHS office. The appendix answered in part: analysis and warnings, technical assistance and incident support. Second, how DHS would decide whether a given request was a priority, given the strategy’s focus on national impact. The strategy gave the principle but not the threshold. Third, what DHS expected in return, such as incident reports or shared indicators, and on what terms.
The answers would shape the state’s own plan. A state that learned DHS help would be scarce for its smaller agencies might build a state-level service to cover them. One that learned DHS expected reports could set up a reporting path in advance. The strategy did not provide those answers in detail. It showed which questions to ask.
What the strategy promised to measure
Each objective ended with an outcome statement. Goal 2’s first objective promised that DHS-led efforts would result in agencies maintaining an adequate level of cybersecurity. Goal 3’s first objective promised that DHS would reduce the most significant national risks to critical infrastructure; its second, that stakeholders would increasingly use the information DHS shared. Goal 5’s first objective promised that DHS would receive reports of cyber incidents and make timely victim notifications.
The outcomes only partly followed the split in powers. Where DHS could direct, it promised a state of the world: agencies at an adequate level. For critical infrastructure it also promised reduced national risk, but several of its outcomes for offered help described activity: information used, reports received, notifications made. In our reading, that mix showed the difficulty. An agency cannot easily show outcomes in organizations it does not control, so the measures for sectors outside the federal enterprise would more readily show how much help DHS gave than how much risk fell.
For a reader trying to judge the strategy later, that distinction matters. An increase in voluntary incident reports could mean more incidents, more trust or both. An increase in assessments delivered could mean better coverage or more demand from organizations that were already strong. The outcome statements did not distinguish these. Organizations that used DHS services had to measure their own results.
Measures a reader can apply to federal help
Because several of DHS’s outcome statements for nonfederal partners measured activity, a recipient organization had to measure the results itself. Three measures fit the strategy’s offers. For assessments: the share of findings remediated, and how quickly. For shared indicators: the share received that were ingested, matched against the organization’s own logs and acted on. For incident help: whether federal involvement shortened recovery or added to it. Each measure turns a service received into a result an auditor or council can read.
How the split fell on different public organizations
Federal civilian agencies
For a federal civilian agency, the strategy described a supervisor and a service provider in one department. Directives set requirements and deadlines. Intrusion detection was mandatory. Tools and services were available, sometimes without reimbursement. The agency’s planning task was to map each requirement to the DHS service that helped meet it, and to fund the gap. Agencies with systems run by contractors had to carry the requirements into those contracts.
State agencies and large cities
For a state agency or large city, the strategy was a set of offers and an investigative partnership. The useful planning task was to decide which services to request, how to receive and act on shared indicators, and whom to call in an incident. A state that had its own fusion center or cyber command could integrate DHS offerings into its own structure. One without such a structure had to decide who would own the relationship.
Counties, small cities and special districts
For smaller jurisdictions, the strategy’s prioritization by national impact was the binding constraint. A county’s systems were critical to its residents and possibly not to the nation. Its access to DHS services would depend on capacity and priority. Its realistic sources of help were its state, regional partnerships, information sharing organizations and its own vendors. The strategy offered analysis and warnings. Turning those into action required staff that small jurisdictions might not have.
Public hospitals and public banks
A county hospital or a public financial institution sat in two places at once: a government entity and part of a critical infrastructure sector. Its sector regulators, where they existed, set requirements. DHS offered services. Neither substituted for the other. For such an institution, the strategy changed nothing about its compliance duties and added options for help.
That dual position also affected incident decisions. A public hospital facing ransomware might owe notices under health privacy rules, report to its state and consider calling federal law enforcement and DHS. The strategy offered help on the last two. The first two were set by other laws and other agencies. An incident plan that listed each call, its legal basis and its owner kept those duties from being confused with the voluntary ones.
Objections to reading the strategy as a menu
Four objections deserve a reply.
The first is that partnership is the right model. The strategy’s collaboration principle said the internet’s growth had been driven primarily by the private sector. On that view, mandates would provoke resistance and could not keep pace with technology. We agree that partnership was the realistic model for DHS in 2018. Our point is narrower. Under that model, the reader carried the decision and the cost. A strategy built on offers works only when recipients take them up and have capacity to act.
The second is that the line between directing and offering is not sharp. Information sharing shapes behavior, assessments create expectations, and regulators in other departments can turn DHS guidance into requirements. That is true, and it is why we describe the offering power as real. But the legal difference remained. In our reading, a federal agency that ignored a binding directive was out of compliance with federal law. A county that ignored an indicator was not, unless its own law said otherwise.
The line also shapes accountability after an incident. When a federal agency suffers a breach, inspectors general and Congress can ask whether it met DHS directives. When a county suffers one, the questions run to its own council, its state and perhaps its insurers. The strategy’s offers do not create a federal standard against which the county is judged. In our reading, that is why a county’s own written requirements matter so much: they are the standard it will be measured against.
The third is that the strategy was a five-year internal plan, not a promise to outsiders. That is also true. But DHS published it, and it described the department’s responsibilities toward critical infrastructure, state and local partners and victims. Readers outside the department were entitled to read it for what it told them, and it told them the shape of the help available.
A fourth objection is that a strategy cannot create capacity, so criticizing it for not reaching small jurisdictions misses its purpose. We agree, and we do not criticize the strategy for its limits. We read it for what it told public leaders about where their own work began. In 2018, that boundary ran along statutory lines, and the strategy marked it clearly.
What a government technology leader could check
The strategy’s four kinds of power translate into a short set of checks for any public organization.
- List every cybersecurity requirement that binds the organization and its source: federal directive, state law, sector regulator, contract or grant condition.
- List the federal services the organization uses or could request, with the named owner of each relationship.
- Confirm who decides whether and when to report an incident to federal authorities, and what the organization expects in return.
- For any system run by a contractor, check that the contract carries the requirements that apply to the system, including federal directives where relevant.
- Measure the organization’s own results, such as exposure, time to remediate and recovery capability, instead of counting services received.
The checks apply differently by size. A federal agency will find its requirements chiefly in directives and OMB policy. A state will find them in state law, executive orders and sector regulators. A county may find few written requirements at all, which makes the first check more important, not less. Knowing that no rule requires something is itself a finding a council should see.
The checks also guard against double counting. An organization that lists a federal service as a control should confirm it actually uses it, and that someone acts on its output. A scanning service whose reports nobody reads is not a control. Recording the owner and the last action taken for each service keeps the list honest.
Each check produces evidence a legislature, auditor or inspector general can read. Together they show which parts of the organization’s security are required, which are borrowed and which are its own.
With hindsight
Later in 2018, the department’s cyber component got a new name and statutory footing. The Cybersecurity and Infrastructure Security Agency Act of 2018, Public Law 115-278, became law on November 16, 2018. It amended the Homeland Security Act to authorize the Cybersecurity and Infrastructure Security Agency within DHS.
The directing power grew sharper inside its existing boundary. On November 3, 2021, CISA issued Binding Operational Directive 22-01. It set up a catalog of known exploited vulnerabilities and required federal agencies to remediate listed ones on deadlines, within two weeks by default for newer ones. The directive’s page described a binding operational directive as “a compulsory direction to federal, executive branch, departments and agencies.” It applied to federal information systems, including systems operated by another entity on an agency’s behalf. On June 10, 2026, CISA issued BOD 26-04, Prioritizing Security Updates Based on Risk, which revoked BOD 22-01 and the earlier BOD 19-02. The new directive set remediation urgency by asset exposure, catalog status, whether exploitation can be automated and technical impact. As before, it applied to federal civilian executive branch agencies and excluded national security systems.
For state and local governments, the change that mattered was money, not mandates. On September 16, 2022, DHS announced the State and Local Cybersecurity Grant Program, which CISA’s page calls a first-of-its-kind program for state, local, tribal and territorial governments. CISA says Congress established the program and appropriated $1 billion for it, to be distributed over four years. It says the program and a companion tribal program would make more than $400 million available in fiscal year 2023 and more than $300 million in fiscal year 2024. As checked on November 15, 2026, the page announced $91.7 million for the fiscal year 2025 program. Grants are still offers, but offers with funding attached, which in our reading went at the capacity gap the 2018 strategy left open.
The grant page describes the program’s purpose as helping eligible entities “address cybersecurity risks and threats to information systems owned or operated by—or on behalf of—SLTT governments.” That is the category the 2018 strategy had left inside critical infrastructure without goals of its own. Grants came with conditions. CISA’s page says the legislation requires states to pass at least 80 percent of funds to local governments, with at least 25 percent of allocated funds going to rural areas. It says applications may include a Cybersecurity Plan, a capabilities assessment and projects approved by a Cybersecurity Planning Committee and the entity’s chief information officer, chief information security officer or equivalent. Those duties came from accepting the offer, not from a directive, but they gave local governments a defined share of the money.
The 2018 strategy anticipated incident reporting as voluntary. Congress later moved toward mandatory reporting for critical infrastructure through the Cyber Incident Reporting for Critical Infrastructure Act of 2022. CISA published its proposed rule on April 4, 2024. A Federal Register notice of May 26, 2026 scheduled town hall meetings in June 2026 to take input on refining the proposal’s scope and burden. As checked on November 15, 2026, we found no final rule; the reporting duty was still a proposal.
In our reading, the split we described held. Compulsion stayed with federal civilian agencies and became more specific. Everyone else received more offers, some with grant money, and a pending reporting duty. A state or county still decides for itself how to build, fund and measure its program. The services available to it are larger than in 2018, but the obligation to act on them remains its own.
Spotted an error? Tell us and we will check it and correct the article where needed.