Specifically, the Statement provided that the Bureau, until further notice, did not intend to take supervisory or enforcement action against land developers subject to ILSA and Regulation J for:

• Delays in filing annual reports of activity with the Bureau, which Regulation J requires within 30 days of the annual anniversary of the effective date of a developer's initial statement of record (12 CFR 1010.310), provided that developers are making good faith efforts to file these reports within a reasonable time; and

• Delays in filing financial statements with the Bureau, which Regulation J requires within120 days of the close of a developer's fiscal year (12 CFR 1010.212(d)), provided that developers are making good faith efforts to file these reports within a reasonable time.

The Bureau hereby rescinds, as of April 1, 2021, the Statement and provides guidance as to how land developers should now meet this obligation.

The Statement expressed the Bureau's recognition of the serious impact of the COVID-19 pandemic on consumers and on the operations of many entities, including land developers subject to ILSA. The Bureau issued the Statement to provide land developers with flexibility and reduce administrative burden. With regard to the filing of the noted annual reports of activity and financial statements, the Bureau believes that land developers have had sufficient time to adapt to the pandemic and should now be able to file annual reports of activity and financial statements without the flexibility afforded under the Statement. The Bureau notes that immediately following the stay-at-home orders issued by States and other jurisdictions, there was a drop-off in ILSA filings with the Bureau. As States and other jurisdictions have rescinded and modified stay-at-home orders over the course of the pandemic, the Bureau notes that land developers have resumed filing of annual reports of activity and financial statements without significant delays. In addition, because the Statement did not create binding legal obligations on the Bureau or create or confer any substantive rights on external parties, it did not create any reasonable reliance interests for industry participants. Additionally, the Bureau never intended the Statement to be permanent.

The Bureau rescinds the Statement and reminds land developers subject to ILSA and Regulation J to resume:

• Filing annual reports of activity with the Bureau, which Regulation J requires within 30 days of the annual anniversary of the effective date of a developer's initial statement of record (12 CFR 1010.310), and

• Filing financial statements with the Bureau, which Regulation J requires within 120 days of the close of a developer's fiscal year (12 CFR 1010.212(d)).

The Bureau does not intend to take supervisory or enforcement action against land developers subject to ILSA for delays in filing annual reports of activity or financial statements with the Bureau, who make any such delayed submission by April 30, 2021 for the time period that the Statement was in effect beginning April 27, 2020 through April 30, 2021.

The Statement constituted a general statement of policy exempt from the notice and comment rulemaking requirements of the Administrative Procedure Act (APA). It was intended to provide information regarding the Bureau's general plans to exercise its supervisory and enforcement discretion and did not impose any legal requirements on external parties, nor did it create or confer any substantive rights on external parties that could be enforceable in any administrative or civil proceeding. This rescission likewise is a general statement of policy exempt from the notice and comment rulemaking requirements of the APA. It is intended to provide information regarding the Bureau's general plans to exercise its supervision and enforcement discretion and does not impose any legal requirements on external parties or create or confer any substantive rights on external parties that could be enforceable in any administrative or civil proceedings. No notice of proposed rulemaking was originally required in issuing the Statement, and it is not required in issuing this rescission. The Regulatory Flexibility Act also does not require an initial or final regulatory flexibility analysis for this rescission. The Bureau has also determined that the rescission of the Statement does not impose any new or revise any existing recordkeeping, reporting, or disclosure requirements on covered entities or members of the public that would be collections of information requiring approval by the Office of Management and Budget under the Paperwork Reduction Act.

The FAA estimates the following costs to do any necessary replacement that would be required based on the results of the required inspection. The agency has no way of determining the number of aircraft that might need this replacement:

Title 49 of the United States Code specifies the FAA's authority to issue rules on aviation safety. Subtitle I, section 106, describes the authority of the FAA Administrator. Subtitle VII: Aviation Programs, describes in more detail the scope of the Agency's authority.

The FAA is issuing this rulemaking under the authority described in Subtitle VII, Part A, Subpart III, Section 44701: General requirements. Under that section, Congress charges the FAA with promoting safe flight of civil aircraft in air commerce by prescribing regulations for practices, methods, and procedures the Administrator finds necessary for safety in air commerce. This regulation is within the scope of that authority because it addresses an unsafe condition that is likely to exist or develop on products identified in this rulemaking action.

This AD will not have federalism implications under Executive Order 13132. This AD will not have a substantial direct effect on the States, on the relationship between the national government and the States, or on the distribution of power and responsibilities among the various levels of government.

For the reasons discussed above, I certify this AD:

(1) Is not a “significant regulatory action” under Executive Order 12866,

(2) Will not affect intrastate aviation in Alaska, and

(3) Will not have a significant economic impact, positive or negative, on a substantial number of small entities under the criteria of the Regulatory Flexibility Act.

Accordingly, under the authority delegated to me by the Administrator, the FAA amends 14 CFR part 39 as follows:

4. Entities Authorized to Make the Request. The Commission adopts its proposal to require voice service providers to respond to traceback requests from the Commission, civil and criminal law enforcement, and the Consortium. The Commission encourages, but does not require, law enforcement to make such requests through the Consortium, where possible. This will improve efficiency and help ensure that requests are handled in a consistent manner. In requiring response to the Consortium, the Commission does not vest any authority in the Consortium or its members to address non-compliance under the Commission's rules. Instead, the Consortium should inform the Commission when they identify a pattern of non-compliance, and the Commission will take any appropriate action.

5. The Commission adopts a modified version of its proposal to require voice service providers to take steps to effectively mitigate illegal traffic when notified by the Commission. Specifically, the Commission directs the Commission's Enforcement Bureau to identify suspected illegal calls and provide written notice to voice service providers. This requirement builds on the safe harbor it established in the Call Blocking Order, published at 85 FR 56530, September. 14, 2020. That safe harbor permits downstream voice service providers to block calls where an upstream voice service provider failed to effectively mitigate illegal traffic after being notified of such traffic by the Commission. This requirement takes the additional step of holding the notified voice service provider liable for that failure.

6. When providing the notice under this new rule, the Enforcement Bureau shall: (1) Identify with as much particularity as possible the suspected traffic; (2) cite the statutory or regulatory provisions the suspected traffic appears to violate; (3) provide the basis for the Enforcement Bureau's reasonable belief that the identified traffic is unlawful, including any relevant nonconfidential evidence from credible sources such as the Consortium or law enforcement agencies; and (4) direct the voice service provider receiving the notice that it must comply with § 64.1200(n)(2) of the Commission's rules. The Commission generally expects that the Enforcement Bureau will notify either the originating voice service provider that has a direct relationship to the caller or the intermediate provider that is the gateway onto the U.S. network.

7. Upon receiving such notice, the voice service provider must promptly investigate the traffic identified in the notice and either take steps to effectively mitigate the identified traffic, in the manner described below, or respond to the Commission that the service provider has a reasonable basis for concluding that the identified calls are not illegal. If the notified voice service provider determines that such traffic comes from an upstream voice service provider with direct access to the U.S. public switched telephone network, the notified voice service provider must promptly inform the Commission of the source of the traffic and, if possible, take lawful steps to effectively mitigate this traffic. Such steps could include, for example, enforcing contract terms or blocking the calls from bad actor providers.

8. Each notified voice service provider must promptly report the results of its investigation to the Enforcement Bureau, including any steps the voice service provider has taken to effectively mitigate the identified traffic, or an explanation as to why the voice service provider reasonably concluded that the identified calls were not illegal, and what steps it took to reach that conclusion. The Commission emphasizes that a “reasonable basis for concluding that the calls are not illegal” requires sufficient due diligence on the part of the voice service provider making such a determination. For example, the mere existence of a contractual provision forbidding illegal calls on the network is not sufficient to make this determination. Similarly, in cases where a caller makes telemarketing calls that include prerecorded or artificial voice messages, is it not reasonable to rely solely on a caller's written or verbal assurances in lieu of documented proof of prior express written consent from the called parties. Callers that believe that they have been blocked in error can seek review by the Commission through existing mechanisms.

9. For a voice service provider to take steps to “effectively mitigate” the traffic identified, it must first investigate to identify the source of that traffic. Where the source is a customer or some other entity that does not have direct access to the U.S. public switched telephone network, the voice service provider must take steps to prevent that source from continuing to originate such traffic. This could mean ending a customer relationship, limiting access to high-volume origination services, or any other steps that have the effect of stopping this traffic and preventing future, similar traffic. The Commission does not expect that originating voice service providers will need to block calls to comply with this requirement, as such voice service providers have a direct relationship with the customer and can use other mechanisms to address these issues. The Commission notes, however, that blocking may be necessary for gateway providers to comply with these requirements.

10. The Commission anticipates that this requirement will primarily impact originating or gateway voice service providers. If, however, a voice service provider receiving notification from the Commission determines that the source of the illegal traffic is another voice service provider with access to the U.S. public switched telephone network, it must notify the Commission. The originally notified voice service provider must, if possible, take any otherwise lawful steps available to effectively mitigate the identified traffic. Where a voice service provider cannot take immediate action, the Commission encourages voice service providers to use the safe harbor for provider-based blocking the Commission adopted in the Call Blocking Order, once the criteria for that safe harbor have been met.

11. The Commission recognizes that intermediate and terminating voice service providers have limited visibility into the actual source of the traffic. The Commission accordingly does not expect perfection in mitigation, nor do the rules it adopts require an intermediate or terminating voice service provider to block all calls from a particular source. Further, the rules the Commission adopts today only require mitigation steps from originating or gateway voice service providers. While gateway providers may need to engage in blocking to comply with this rule, the Commission does not expect them to block all traffic, and encourages use of other methods where available.

12. The Commission adopts its proposal to require voice service providers to adopt affirmative, effective measures to prevent new and renewing customers from using their network to originate illegal calls. The Commission requires that all originating voice service providers know their customers and exercise due diligence in ensuring that their services are not used to originate illegal traffic. Beyond that, the Commission does not require that voice service providers take specific, defined steps, but instead permits them flexibility to determine what works best on their networks. The Commission does recommend that voice service providers exercise caution in granting access to high-volume origination services, to ensure that bad actors do not abuse such services.

13. While more involved investigations represent some burden, particularly for smaller voice service providers, voice service providers of all sizes should be able to impose and enforce relevant contract terms. Of course, contract provisions are only effective if they are enforced, and voice service providers that refuse to do so fail to satisfy this requirement. The Commission also clarifies that, if voice service providers have already implemented effective measures, they do not need to take further steps at this time. Voice service providers, however, that have not done so will need to comply with this requirement when they accept new customers or renew existing customers after the date of these rules.

14. The Commission adopts its proposal to expand the safe harbor based on reasonable analytics to cover network-based blocking if the network-based blocking incorporates caller ID authentication information where available and otherwise meets the requirements the Commission adopted both in the Call Blocking Order and elsewhere in the Fourth Report and Order. To get the benefit of this safe harbor, a terminating voice service provider must ensure its network-based blocking targets only calls highly likely to be illegal, not simply unwanted. It must also manage this blocking with human oversight and network monitoring sufficient to ensure that the blocking works as intended; this must include a process that reasonably determines that the particular call pattern is highly likely to be illegal prior to blocking calls that are part of that pattern. And the Commission expects voice service providers to demonstrate they have conducted an appropriate process should the Commission inquire about specific blocking. The Commission bases this decision on its previous finding that no reasonable consumer would want to receive calls that are highly likely to be illegal. The safe harbor the Commission adopts today ensures that terminating voice service providers can respond to evolving threats while safeguarding the calls consumers want.

15. Network Blocking Based on Reasonable Analytics. In expanding its safe harbor, the Commission first makes clear that terminating voice service providers may block calls at the network level, without consumer opt in or opt out, if that blocking is based on reasonable analytics that incorporate caller ID authentication information designed to identify calls and call patterns that are highly likely to be illegal. Terminating voice service providers must manage that blocking with human oversight and network monitoring sufficient to ensure that blocking is working as intended. This must include a process to reasonably determine that the particular call pattern is highly likely to be illegal prior to blocking calls. A terminating voice service provider must disclose to consumers that it is engaging in such blocking so that consumers are fully aware of it. In the 2017 Call Blocking Order, published at 83 FR 1566, January 12, 2018, the Commission authorized voice service providers to block, without consumer consent, certain categories of calls on the basis that no reasonable consumer would want to receive such calls. The Commission's decision here is an extension of this decision, making clear that a terminating voice service provider may block any calls that it determines are highly likely to be illegal based on certain defined parameters.

16. For purposes of this safe harbor, the Commission makes clear that terminating voice service providers must have in place a process to reasonably determine that the particular call pattern is highly likely to be illegal prior to blocking calls. Doing so will ensure that important calls, including emergency calls, are not blocked in error based solely on analytics. The Commission does not prescribe the specific steps of this process but instead expects that it will include steps designed to find out whether the calls that are part of the call pattern in question are highly likely to be illegal such as dialing the telephone number from which the apparently illegal calls purportedly originate; reviewing complaint data about calls from the source; or contacting the originating voice service provider.

17. The Commission urges terminating voice service providers to be thorough in this process to avoid blocking errors. In particular, the Commission believes terminating voice service providers will need to use a combination of methods in their processes. For example, call-backs may work well for some types of high-volume traffic, but may not be determinative for emergency alerts since the number may not be set up to receive calls. The Commission further clarifies that, because it only authorizes blocking of calls that are part of a call pattern that indicates the calls are highly likely to be illegal, when a terminating voice service provider learns that calls fitting this pattern are likely lawful, that voice service provider must immediately cease network-based blocking. The Commission notes that a voice service provider may continue to block under this safe harbor while investigating a dispute prior to obtaining information that indicates the calls are likely lawful.

18. Consistent with the TRACED Act and building on its Call Blocking Order and Further Notice, the Commission adopts additional requirements to provide callers and consumers with more transparency to ensure that they can effectively access redress mechanisms, and ensure that a caller can verify the authenticity of its calls. First, the Commission requires terminating voice service providers that block calls to immediately notify the caller that the call has been blocked by sending either a Session Initiation Protocol (SIP) or ISDN User Part (ISUP) response code, as appropriate, and the Commission requires all voice service providers in the call path to transmit these codes to the origination point. Second, the Commission requires terminating voice service providers that block calls on an opt-in or opt-out basis to disclose to their subscribers a list of blocked calls upon request. Third, when a calling party disputes whether blocking its calls is appropriate, the Commission requires terminating voice service providers to provide a status update to the party that filed the dispute within 24 hours. The Commission requires that the point of contact which terminating voice service providers have established to handle blocking disputes also handle contacts from callers that are adversely affected by information provided by caller ID authentication seeking to verify the authenticity of their calls. Finally, the Commission declines to address the issue of erroneous labeling at this time.

19. The Commission expects voice service providers to satisfy those requirements, whether or not they use a third party, to ensure that the safeguards the Commission adopts to prevent erroneous blocking apply across the network. The Commission therefore declines to allow voice service providers to avoid liability solely because they make use of a third-party service. By contrast, a voice service provider is not responsible for blocking done by a blocking service chosen by the consumer such as a blocking app.

20. The Commission requires terminating voice service providers that block calls to immediately notify callers of such blocking. The Commission further directs all voice service providers to perform necessary software upgrades to ensure the codes it requires for such notification are appropriately mapped; voice service providers must ensure that calls that transit over TDM and IP networks return an appropriate code. The Commission requires all voice service providers in the call path to transmit the code, or its equivalent, as discussed below, to the origination point so that callers with the appropriate equipment may receive timely notice of a blocked call. The Commission's requirements ensure that legitimate callers know when their calls are blocked so they can seek redress.

21. Method of Notification. To ensure that callers understand these notifications and can make informed decisions regarding next steps, the Commission requires voice service providers to use specific, existing codes when blocking calls. Callers with properly configured equipment will thereby receive sufficient information to determine whether to access redress or investigate the blocking further. The Commission requires that terminating voice service providers that block calls on an IP network return SIP Code 607 or 608, as appropriate. Both of these codes are designed to be used for call blocking. Because SIP codes are not available on non-IP networks, ISUP code 21 is the appropriate code for calls blocked on a TDM network. Therefore, the Commission requires that terminating voice service providers that block calls on a TDM network return ISUP code 21.

22. Many calls transit both IP-based and TDM networks. The Commission therefore establishes requirements regarding how these codes map, or translate, when call signaling transits between IP and TDM. For signals moving from IP to TDM, the Commission directs voice service providers, regardless of their position in the network, to make any necessary upgrades or software configuration changes to ensure that SIP Codes 607 and 608 map to ISUP code 21. In certain cases, callers may also receive SIP code 603 when calls have been blocked. This is likely to occur when call signaling transits from TDM to IP. The Commission further notes that the specifications for SIP code 608 give some guidance for interoperability, including the playing of an announcement. The Commission strongly encourages voice service providers to use this portion of the specification to eliminate confusion caused by ISUP code 21's multiple uses.

23. Compliance Date. Finally, the Commission gives voice service providers until January 1, 2022, approximately 12 months after the adoption of the Order, to comply with its immediate notification requirements. The Commission recognizes that voice service providers are bearing costs as they work to implement STIR/SHAKEN in the IP portions of their networks, which the Commission has required by June 30, 2021. Any incremental increase in burden could introduce challenges in compliance for some voice service providers. By delaying the compliance date of these requirements until January 1, 2022, which is approximately 12 months from the adoption of the Order, or six months from the STIR/SHAKEN implementation compliance date, the Commission ensures that voice service providers can make the necessary software upgrades without diverting resources from STIR/SHAKEN implementation.

24. The TRACED Act directs the Commission to ensure that consumers, as well as callers, have transparency and effective redress when wanted calls are blocked using robocall blocking services provided on an opt-out or opt-in basis pursuant to its Call Blocking Declaratory Ruling and Further Notice. Consumers may opt out of blocking services at any time as a form of redress, and the Commission already made clear that voice service providers that use blocking programs should disclose to consumers the types of calls they seek to block and make clear that some wanted calls may be blocked. While a consumer knows that wanted calls may be blocked, consumers may not be able to determine whether blocking has occurred. The Commission thus requires any terminating voice service provider that blocks calls on an opt-in or opt-out basis to provide, on the request of the subscriber to a particular number, a list of all calls intended for that number that the voice service provider or its designee has blocked. The list should contain the calling number and the date and time of the call. Consistent with the TRACED Act, this list must be provided at no additional charge to the consumer. To ensure that this information is provided to the subscriber in a timely manner, the Commission requires that the terminating voice service provider provide this list within three business days of receiving the request. To avoid unwieldy recordkeeping requirements, the Commission limits the reporting requirement to calls blocked in no fewer than the 28 days prior to the request.

25. Recipients and Content of List. The Commission requires that the blocked calls list include calls blocked on an opt-out or opt-in basis and that the voice service provider make it available to the subscriber to the called number. The list need only contain calls blocked with consumer consent because consumers can review these calls and make a different choice. This is also consistent with the scope of transparency and effective redress requirement in section 10(b) of the TRACED Act, which applies to “robocall blocking services provided on an opt-out or opt-in basis.” The Commission limits the scope of parties who can request and receive blocked calls lists to the subscriber because callers already have transparency and effective redress through the Commission's other requirements and doing so avoids the additional burden on voice service providers of furnishing lists to a potentially large group of others.

26. Flexibility in Blocked Calls List Mechanism. Consistent with the call for flexibility in the record, the Commission declines to mandate how voice service providers give subscribers this list. The Commission instead leaves the method of providing the information to the judgment of the voice service provider. For example, voice service providers could use a web portal and those using a third-party blocking service may rely on that third party to provide this list. The ultimate responsibility falls to the voice service provider. Should the third party fail to provide the list consistent with the requirements the Commission adopts herein, the voice service provider will be liable for this failure.

27. Time for Response and Length of Recordkeeping Requirement. The Commission requires terminating voice service providers to respond to subscribers' requests for blocked calls lists within three business days of receiving such a request. In establishing this requirement, the Commission balances the burden to the voice service provider against the needs of the subscriber. The Commission determines that three business days is sufficient time for a voice service provider to compile such a list, even if they are using a manual process. It also ensures that the subscriber, who may be requesting the list because they suspect important calls may not be reaching them, has access to this information in a timely manner.

28. To meet this obligation and for purposes of this rule, voice service providers must retain records of blocked calls for a minimum of four weeks or 28 days. This recognizes the need for subscribers to receive meaningful information and seeks to avoid overly burdening voice service providers with unnecessary recordkeeping requirements. Twenty-eight days provides the information subscribers need and imposes a reasonable burden on voice service providers.

29. To ensure that callers can track dispute status and to increase transparency consistent with the TRACED Act, the Commission enhances its existing redress requirements to require voice service providers to respond to blocking disputes they receive through their established point of contact by providing a status update to the party that filed the dispute within 24 hours. In doing so, the Commission does not modify its requirement that disputes are resolved in a reasonable amount of time and at no cost to the caller, if the complaint is made in good faith. Instead, the Commission recognizes both that callers need speedy resolution of disputes and that voice service providers may need additional time to resolve disputes in certain instances. By requiring a status update within 24 hours, the Commission ensures that callers have the information they need while also granting voice service providers flexibility.

30. The Commission requires that the point of contact terminating voice service providers have established to take blocking disputes also handle contacts from callers that are adversely affected by information provided by caller ID authentication seeking to verify the authenticity of their calls. Because the Commission's rules already require blocking voice service providers to have a point of contact, it expects that most terminating voice service providers already have one in place. Any terminating voice service provider that does not block calls, and takes into account attestation information in determining how to deliver calls, must provide a point of contact to receive caller complaints regarding caller ID authentication consistent with the rules the Commission established in the Call Blocking Order, the Call Blocking Further Notice, and the Fourth Report and Order.

31. Section 4(c)(1)(C) of the TRACED Act requires the Commission to establish a mechanism for callers that are adversely affected by information provided by the caller ID authentication framework to verify the authenticity of the calls. This will provide callers with a mechanism for redress where, for example, calls are blocked due to an incorrect attestation. The Commission specifies that the point of contact is not required to resolve all disputes about attestation level and, in fact, is not properly placed to do so; the terminating voice service provider is not the entity that typically attests to caller ID. Instead, the terminating voice service provider should consider whether the decision in question would be appropriate if the same calls were to receive a higher level of attestation and treat future calls accordingly unless circumstances change.

32. The Commission's decision to require the same point of contact to handle blocking disputes and “adverse effects” from caller ID authentication information streamlines the process for both voice service providers and callers. The Commission expects that blocking and authentication concerns will often be interrelated, such as when the adverse effect is blocking. Only the terminating voice service provider can determine whether caller ID authentication was a significant factor in its decision and therefore whether there is a need to adjust its analytics or otherwise change its call-delivery practices. Even when the adverse effects from caller ID authentication and blocking are not directly related, by requiring the same point of contact to receive complaints of both issues, the Commission ensures that a caller only needs to go to one contact at a given terminating voice service provider in order to resolve either issue.

33. Because the TRACED Act requirement seeks to address “adverse effects,” not simply incorrect caller ID authentication information, the Commission finds the terminating voice service provider is in the best position to address callers' concerns. The terminating voice service provider takes the action that represents the adverse effect, such as blocking. Originating voice service providers, by contrast, are not so positioned because they cannot ensure that attestation information reaches the terminating voice service provider. This is because STIR/SHAKEN does not work on TDM networks. Even once voice service providers implement STIR/SHAKEN, some voice service providers may thus be unable to sign calls and some calls may drop the initial attestation when they transit on TDM.

34. The Commission declines to extend redress mechanisms to erroneous call labeling at this time. Rather, the Commission encourages voice service providers and their analytics partners to work in good faith with callers to avoid erroneous labeling so consumers can better decide whether to answer a call.

35. Safe Harbor for Misidentification of the Level of Trust. At this time, the Commission declines to extend the safe harbor to cover the inadvertent or unintended misidentification of the level of trust for particular calls. The Commission is not aware of any sources of liability specifically for the misidentification of the level of trust, including any liability stemming from non-federal sources. The Commission makes clear, however, that it will consider such a safe harbor in the future should parties bring such sources of liability to its attention. In reaching this conclusion, the Commission finds that it has met the TRACED Act's direction to provide such a safe harbor through its provision of a safe harbor for blocking, which is the only potential source of liability of which it is aware.

36. TRACED Act Section 7. The Commission declines to take further action under section 7 of the TRACED Act at this time, but makes clear that the Commission may act in the future, as circumstances warrant. The Commission believes that, at this time, the best approach to protecting consumers from unwanted calls from unauthenticated numbers is through blocking programs that are consistent with the safe harbor it adopted in the Call Blocking Order. The Commission concludes that it has met its statutory obligation under section 7 by seeking comment on additional steps the Commission could take to provide this protection.

37. Other Section 4(c) Issues. The Commission adopts the tentative conclusions proposed in the Call Blocking Further Notice with regard to section 4(c) of the TRACED Act. Specifically, the Commission finds that it has fully implemented section 4(c)(1), except to the extent that it adopts new rules elsewhere in this Order. The Commission similarly concludes that, in establishing the safe harbor adopted in the Call Blocking Order and expanded upon elsewhere in this Order, it has properly taken into account the considerations listed in section 4(c)(2) of the TRACED Act.

1. As required by the Regulatory Flexibility Act of 1980 (RFA), as amended, an Initial Regulatory Flexibility Analysis (IRFA) was incorporated into the Declaratory Ruling and Further Notice. The Commission sought written public comment on the proposals in the Call Blocking Further Notice, including comment on the IRFA. The comments received are discussed below. This Final Regulatory Flexibility Analysis (FRFA) conforms to the RFA.

2. The Fourth Report and Order takes important steps in the fight against illegal robocalls by requiring voice service providers to take certain affirmative steps to prevent illegal calls. Next, the Fourth Report and Order expands the Commission's safe harbor to include network-based blocking based on reasonable analytics that incorporate caller ID authentication information designed to identify calls that are highly likely to be illegal, if this blocking is managed with human oversight and network monitoring sufficient to ensure that blocking is working as intended. The Fourth Report and Order then takes steps to implement the TRACED Act by ensuring that both callers and consumers are provided with transparency and effective redress. Taken together, these steps will provide greater protection to consumers and increase trust in the telephone system while ensuring that consumers continue to receive the calls they want.

3. Affirmative Obligations for Voice Service Providers. The Fourth Report and Order establishes three affirmative obligations for all voice service providers. First, all voice service providers must respond to traceback requests from the Commission, civil and criminal law enforcement, or the Industry Traceback Consortium (Consortium). Second, all voice service providers must take steps to effectively mitigate suspected illegal traffic when notified of such traffic by the Commission through the Enforcement Bureau. The notice from the Enforcement Bureau must be in writing and include specific information as detailed in the Fourth Report and Order and accompanying rules. The notified voice service provider must investigate the suspected illegal traffic and report to the Enforcement Bureau regarding the results of that investigation, including whether the calls came from another voice service provider with direct access to the U.S. public switched telephone network, and any mitigation steps taken. Finally, all voice service providers must take affirmative, effective steps to prevent new and renewing customers from using their network to originate illegal calls.

4. Expanding the Safe Harbor Based on Reasonable Analytics to Network-Based Blocking. The Fourth Report and Order expands the Commission's safe harbor for blocking based on reasonable analytics, which must include caller ID authentication information where available, to cover certain network-based blocking, without consumer opt in or opt out. The blocking must be designed to target only calls highly likely to be illegal and managed with sufficient human oversight and network monitoring to ensure that blocking is working as intended. For purposes of the safe harbor, the Fourth Report and Order makes clear that voice service providers must have a process in place to reasonably determine that a call pattern is highly likely to be illegal prior to initiating blocking without consumer consent, and must cease blocking when the voice service provider learns that calls are likely lawful.

5. Enhanced Transparency and Redress. The Fourth Report and Order establishes several requirements to implement the TRACED Act and ensure that both callers and consumers are provided with transparency and effective redress. First, voice service providers that block calls must return to the caller an appropriate SIP or ISUP code, as appropriate. In order to ensure that these codes reach the origination point of the call, all voice service providers must make all necessary software upgrades and configuration changes to ensure that these codes translate properly when a call moves between TDM and IP-based networks. Providers must comply with this requirement by January 1, 2022. Second, voice service providers that block on an opt-in or opt-out basis must provide, on the request of the subscriber to a particular number, a list of all calls intended for that number that the provider has blocked. Voice service providers have three days to provide the list and the list should include all calls blocked on an opt-in or opt-out basis within the 28 days prior to the request. Third, voice service providers that block calls must respond to any blocking dispute within 24 hours, either with a status update or a resolution. This requirement builds on the Commission's requirements in the Call Blocking Order and Further Notice that voice service providers designate a single point of contact to handle blocking disputes. Finally, consistent with the TRACED Act, the Fourth Report and Order requires that the point of contact previously established to handle blocking disputes also be prepared to handle contacts from callers seeking to verify the authenticity of their calls. Any terminating voice service provider that does not block calls, and takes into account attestation information in determining how to deliver calls, must provide a point of contact to receive caller complaints regarding caller ID authentication consistent with the rules the Commission established in the Call Blocking Order and Further Notice, as well as the Fourth Report and Order.

6. In the Call Blocking Order and Further Notice, the Commission solicited comments on how to minimize the economic impact of the new rules on small business. The Commission received seven comments either directly referencing the IRFA or addressing concerns particular to small businesses. Five of these comments addressed the affirmative obligations. Six addressed small business concerns with transparency and redress requirements. Three of these comments addressed issues raised in the Call Blocking Order and Further Notice that the Fourth Report and Order declines to move forward with, and therefore are not directly relevant to this analysis.

7. Affirmative Obligations. The Fourth Report and Order requires voice service providers to respond to traceback, mitigate illegal traffic when notified of such traffic by the Commission, and take affirmative steps to prevent illegal calls from new and renewing customers. Commenters, including smaller voice service providers, were generally supportive of these requirements. Commenters did urge us to take certain steps to aid smaller voice service providers and ensure that these voice service providers have the information and resources to comply.

8. With regard to the requirement to respond to traceback requests from the Commission, law enforcement, and the Consortium, ACA Connects notes that many smaller voice service providers may be unfamiliar with the process and urges us to work with stakeholders to educate smaller voice service providers. In the Fourth Report and Order the Commission makes clear to voice service providers that it is in their best interest to ensure that they have a clear point of contact at which to receive these requests. The Commission remains open to working with smaller voice service providers and other stakeholders to ensure that they understand the traceback process and how best to handle these requests.

9. There is significant overlap with commenters' concerns regarding the second and third requirements. In general, these concerns urge the Commission to ensure that these requirements, to the extent possible, consider that smaller voice service providers will have fewer resources. INCOMPAS urges us to avoid overly prescriptive requirements and to provide flexibility on the third requirement. Competitive Carriers urges us to ensure that, with regard to new and renewing customers, the requirement “is satisfied so long as a provider takes action once it has actual knowledge of a customer originating illegal calls.” WTA asks that the Commission define particular steps so a voice service provider can be sure it is in compliance. The Fourth Report and Order makes clear that, while the Commission does not define specific steps, the Commission does not expect perfection, and that enforcement of contract clauses is sufficient to satisfy the third requirement. By granting flexibility, the Commission ensures that all voice service providers can determine the approach best suited to their networks.

10. Transparency and Redress. The Fourth Report and Order adopts several transparency and redress requirements, including immediate notification of blocking, provision of a blocked calls list for consumers, and status updates regarding disputes. Commenters raise concerns that prescriptive transparency and redress mandates are particularly burdensome for smaller voice service providers, and generally seek flexibility. They note that smaller providers are more often reliant on third parties, both for blocking services and associated redress. Commenters also raise particular concerns regarding speed of redress for smaller providers. There is, however, some disagreement on this issue, with Telnyx noting that smaller voice service providers may also be disadvantaged if larger voice service providers take too long to resolve disputes. WTA also raises concerns about the burdens associated with requiring a blocked calls list, but does not specifically tie these concerns to voice service provider size.

11. Pursuant to the Small Business Jobs Act of 2010, which amended the RFA, the Commission is required to respond to any comments filed by the Chief Counsel for Advocacy of the Small Business Administration (SBA), and to provide a detailed statement of any change made to the proposed rules as a result of those comments. The Chief Counsel did not file any comments in response to the proposed rules in this proceeding.

27. The Fourth Report and Order requires voice service providers to meet certain affirmative obligations and to provide specific transparency and redress to both callers and consumers. These changes affect small and large companies equally and apply equally to all the classes of regulated entities identified above.

28. Reporting and Recordkeeping Requirements. The Fourth Report and Order requires voice service providers to effectively mitigate illegal traffic once notified of suspected illegal traffic by the Commission through its Enforcement Bureau. As part of this requirement, a notified voice service provider must promptly report the results of its investigation to the Enforcement Bureau, including any steps the voice service provider has taken to effectively mitigate the identified traffic, or an explanation as to why the voice service provider reasonably concluded that the identified calls were not illegal, and what steps it took to reach that conclusion. The Fourth Report and Order also requires voice service providers to provide, at the request of a subscriber, a list of calls blocked on an opt-out or opt-in basis over the prior 28 days. This requires voice service providers that block calls on an opt-out or opt-in basis to retain records regarding such blocking for a minimum of 28 days. The other requirements adopted in the Fourth Report and Order do not include specific recordkeeping or retention requirements. However, voice service providers may find it necessary to retain records to ensure that they are able to resolve blocking disputes, respond to traceback, or demonstrate that they are in compliance with the Commission's rules in the event of a dispute.

29. The RFA requires an agency to describe any significant alternatives that it has considered in reaching its approach, which may include the following four alternatives, among others: (1) The establishment of differing compliance or reporting requirements or timetables that take into account the resources available to small entities; (2) the clarification, consolidation, or simplification of compliance or reporting requirements under the rule for small entities; (3) the use of performance, rather than design, standards; and (4) an exemption from coverage of the rule, or any part thereof, for small entities.

30. The Commission considered feedback from the Call Blocking Order and Further Notice in crafting the final order. The Commission evaluated the comments with the goal of protecting consumers from illegal calls while also ensuring that both consumers and callers receive transparency and effective redress. For example, in establishing affirmative obligations for voice service providers, the Commission ensured that voice service providers have flexibility to determine how best to comply and made clear that the Commission does not expect perfection With regard to transparency and redress requirements, wherever possible, the Commission ensures that prescriptive requirements make use of already-existing mechanisms to minimize the burdens and declined to require resolution of blocking disputes within a specific timeframe. The Commission also delays the date of compliance, setting it at January 1, 2022, to ensure that voice service providers have sufficient time to make any necessary upgrades or configuration changes before they must provide immediate notification of blocked calls by providing a SIP or ISUP code.

31. The Fourth Report and Order carefully weighs the concerns of small voice service providers against those of callers, many of which are also small businesses. In adopting an immediate notification requirement, it makes use of existing mechanisms and delays the compliance date to keep the burden as low as possible while still providing important information to callers. Further, in requiring a status update, but not resolution, within 24 hours, the Fourth Report and Order ensures that small voice service providers have necessary time to conduct investigations while also providing valuable information to callers. The requirements adopted in the Fourth Report and Order will impose some burden on smaller voice service providers, but these burdens are necessary to implement the TRACED Act and ensure that both callers and consumers are provided with transparency and effective redress.

32. The Commission does not see a need to establish a special timetable for small entities to reach compliance with the modification to the rules. No small business has asked for a delay in implementing the rules. Any voice service providers that require such a delay may reach out through the usual processes. Similarly, there are no design standards or performance standards to consider in this rulemaking.

91. Accordingly, it is ordered that, pursuant to sections 4(i), 201, 202, 217, 227, 227b, 251(e), 303(r), and 403 of the Communications Act of 1934, as amended, 47 U.S.C. 154(i), 201, 202, 217, 227, 227b, 251(e), 303(r), 403, this Fourth Report and Order is adopted.

92. It is further ordered that part 0 and part 64 of the Commission's rules are amended as set forth in the Final Rules, with the exception of new §§ 64.1200(k)(9) and (10), and 64.1200(n)(2), and shall be effective 30 days after their publication in the Federal Register .

93. It is further ordered that new § 64.1200(k)(9) shall be effective on January 1, 2022.

94. It is further ordered that new § 64.1200(k)(10) and 64.1200(n)(2) shall be effective 30 days after the Commission's publication of a notice in the Federal Register , which will announce approval of portions of the rules requiring approval by OMB under the PRA.

For the reasons discussed in the preamble, the Federal Communications Commission amends 47 parts 0 and 64 as follows:

In the NPRM, the Board proposed requirements for minimum information to be included on or with Class I carriers' demurrage invoices. 6 NPRM, EP 759, slip op. at 8-11, 14-15. In response, the Board received a significant number of comments from stakeholders, including requests for additional and modified invoicing requirements. 7 The Board subsequently issued a supplemental notice of proposed rulemaking to invite comment on potential modifications and additions to the proposed minimum information requirements and format. The Board received comments and replies in response to the SNPRM. 8 After considering the record in this proceeding, the Board adopts a final rule requiring Class I carriers to include certain minimum information on or with their demurrage invoices and provide, in the format of their choosing, machine-readable  9 access to the required minimum information, as discussed below.

This proceeding arises, in part, as a result of the testimony and comments submitted in Oversight Hearing on Demurrage & Accessorial Charges, 10 Docket No. EP 754. In that proceeding, parties from a broad range of industries raised concerns about demurrage invoicing practices, including issues involving the receipt of invoices containing insufficient information. See NPRM, EP 759, slip op. at 5-6 (providing overview of comments received in Docket No. EP 754 related to the adequacy of demurrage invoices).

After carefully considering the comments and testimony in Docket No. EP 754, the Board issued the NPRM in this docket. As relevant here, the Board proposed requirements for certain minimum information to be included on or with Class I carriers' demurrage invoices. Specifically, the Board proposed the inclusion of:

• The unique identifying information ( e.g., reporting marks and number) of each car involved;

• the following shipment information, where applicable:

○ The date the waybill was created;

○ the status of each car as loaded or empty;

○ the commodity being shipped (if the car is loaded);

○ the identity of the shipper, consignee, and/or care-of party, as applicable; and

○ the origin station and state of the shipment;

• the dates and times of:

○ Actual placement of each car;

○ constructive placement of each car (if applicable and different from actual placement);

○ notification of constructive placement to the shipper, consignee, or third-party intermediary (if applicable); and

○ release of each car; and

• the number of credits and debits attributable to each car (if applicable). NPRM, EP 759, slip op. at 9-10. The Board also proposed to require Class I carriers, prior to sending demurrage invoices, to take “appropriate action to ensure that the demurrage charges are accurate and warranted, consistent with the purpose of demurrage.” Id. at 10 (footnote omitted). The Board proposed to add both the minimum information requirements and the appropriate-action requirement to a new regulation at 49 CFR 1333.4. Id. at 14. In the NPRM, the Board invited stakeholders to comment on the proposed rule, as well as any additional information that Class I carriers could reasonably provide on or with demurrage invoices to help rail users effectively evaluate those invoices. Id. at 10.

In response to stakeholders' comments, the Board issued the SNPRM, which invited comments on modifications and additions to proposed section 1333.4(a) that the Board was considering. The changes proposed in the SNPRM would require that Class I carriers provide certain additional information on or with demurrage invoices, including: (1) The billing cycle covered by the invoice; (2) the original estimated date and time of arrival (ETA) of each car (as established by the invoicing carrier) and the date and time each car was received at interchange (if applicable), either on or with each invoice or, alternatively, upon reasonable request from the invoiced party; and (3) the date and time of each car ordered in (if applicable). SNPRM, EP 759, slip op. at 4-5. In the SNPRM, the Board also asked for comment on a requirement that Class I carriers provide access to demurrage invoicing data in machine-readable format and invited further comment on the proposed demurrage regulations at section 1333.4(b), which would require Class I carriers to take appropriate action to ensure that demurrage charges are accurate and warranted prior to sending demurrage invoices. 11 Id. at 5, 9-11.

As discussed below, rail users express broad support for the minimum information proposed in the NPRM and SNPRM, although many suggest additions and modifications that they argue would improve the rule. Rail users also largely support a machine-readable data requirement and the Board's proposal to require Class I carriers to “take appropriate action to ensure that demurrage charges are accurate and warranted.” Some rail users argue that the rule should apply to Class II and Class III carriers.

Class I carriers oppose the proposed minimum information requirements but argue that, if they are adopted, carriers should be allowed to provide the information on their existing online platforms rather than on or with invoices. Class I carriers also oppose the Board's proposed appropriate-action requirement. ASLRRA supports the proposed exclusion of Class II and Class III carriers from the rule.

The Board now adopts a final rule requiring Class I carriers to include certain minimum information on or with demurrage invoices and provide machine-readable access to the minimum information. The final rule is below.

Class I carriers argue generally that a rule establishing any minimum information requirements is unnecessary; would lead to increased litigation; contradicts Board precedent, the rail transportation policy (RTP) of 49 U.S.C. 10101, and the purpose of demurrage; and would restrict innovation. 12

CSXT and CN argue that rail users have not shown that Class I carriers have a systemwide problem with demurrage invoicing sufficient to justify the rule. (CSXT Comments 4, Nov. 6, 2019; CN Comments 4-5, Nov. 6, 2019; CSXT Comments 3, June 5, 2020; CN Comments 5-6, June 5, 2020.) CSXT, CN, UP, and KCS assert that rail users' complaints do not apply to them because they already provide sufficient information. 13 To the extent some Class I carriers do not provide sufficient information, CP urges the Board to “defer to competitive market pressures to provide the incentive for those railroads to innovate and to catch up with their peers.” (CP Comments 5, Nov. 6, 2019.) CSXT suggests that any problems with carriers' invoicing systems should be addressed on an individualized basis through the Board's formal and informal complaint procedures. (CSXT Comments 4, Nov. 6, 2019.)

CN and CSXT also express concerns about the effect that minimum information requirements would have on demurrage litigation. CN argues that the NPRM suggests that “invoices will not be deemed valid unless they include all eleven specific categories of information,” which would lead to more frequent and more complex litigation. (CN Comments 7, Nov. 6, 2019; see also CSXT Comments 4-5, June 5, 2020 (asserting that minimum information requirements would lead to disputes “over purely technical issues”).) For example, CN suggests that an invoice recipient could argue that an invoice is invalid if “a waybilling error by the originating shipper causes a demurrage bill to show the wrong commodity for a particular car” even if the error has “no material effect on the demurrage billpayer's ability to understand and potentially dispute demurrage for that car.” (CN Comments 7, Nov. 6, 2019; see also CSXT Comments 4-5, June 5, 2020 (contending that under the Board's proposal, “shippers could challenge invoices on the basis of missing or incorrect information whether due to carrier fault or otherwise”).)

Furthermore, CSXT and CN argue that minimum information requirements contradict Maintenance of Records Pertaining to Demurrage, Detention, & Other Related Accessorial Charges by Rail Common Carriers of Property ( Maintenance of Records ), 367 I.C.C. 145 (1982). (CSXT Comments 3, Nov. 6, 2019; CN Comments 7, Nov. 6, 2019; CN Comments 6, June 5, 2020.) CSXT contends that the Board's proposed rule is contrary to Congressional policy and “would turn the clock back to long-rejected policies that mandated paperwork requirements for demurrage bills and that prescribed inefficient nationwide practices.” (CSXT Comments 3, Nov. 6, 2019; see also CN Comments 7, Nov. 6, 2019 (arguing that the Board should refrain from reversing “the principle underlying this [Interstate Commerce Commission (ICC)] decision by adopting requirements for the content of demurrage invoices that would bind all Class I railroads”).)

Additionally, CP and CN contend that minimum information requirements would contravene one of the goals of the RTP at 49 U.S.C. 10101(2), specifically “to minimize the need for Federal regulatory control over the rail transportation system.” (CP Comments 5, Nov. 6, 2019; CN Comments 4, June 5, 2020.) CP also argues that the proposed rule is inconsistent with one of the objectives of demurrage—encouraging the efficient use of rail assets—because the proposal “places the emphasis on empowering customers in their ability to challenge invoiced demurrage charges” after the fact instead of focusing Board policy on encouraging customers “to remain actively engaged in monitoring and managing their supply chains to . . . avoid incurring demurrage charges in the first place.” (CP Comments 4, June 5, 2020.)

Moreover, several carriers allege that minimum information requirements could stifle innovation and discourage carriers from exploring other methods of providing demurrage information to rail users. 14 (CP Comments 5, Nov. 6, 2019; CN Comments 9-10, Nov. 6, 2019; CSXT Comments 1-2, Nov. 6, 2019.)

CN argues that, at most, the Board should adopt a flexible, “performance-based” standard that would require Class I carriers to “ensure that recipients of demurrage invoices have access to sufficient information to be able to understand the basis for the charges and to dispute charges believed to be unwarranted,” which could be “provided either on or with the demurrage invoice or through another electronic means, including through a software platform or portal.” (CN Comments 14-15, June 5, 2020.) CSXT and NSR also support this proposal. (CSXT Comments 5, June 5, 2020; NSR Reply 2, July 6, 2020.)

In their replies, many rail users counter that they are unable to effectively review and understand the demurrage invoices they receive from Class I carriers because the carriers either provide limited information or do not format the information in ways that enable efficient access and auditing. 15 ISRI acknowledges that minimum information requirements may increase costs for carriers but contends that rail users currently “bear the costs and burdens associated with overtime and additional staffing needed to verify the accuracy of the invoices.” (ISRI Reply 10-11, Dec. 6, 2019; see also WCTL & SEC Reply 9, Dec. 6, 2019 (arguing that carriers “have effectively shifted the time and costs for reviewing invoices to shippers”).)

With respect to Class I carriers' concerns about increased litigation over technical issues, joint commenters (ACC, CRA, TCI, and TFI)  16 assert that rail users will not be inclined to dispute appropriate demurrage charges over technical issues since demurrage disputes are costly. (Joint Reply (ACC, CRA, TCI, & TFI) 13, July 6, 2020.) In addition, Dow argues that to the extent Class I carriers face more demurrage claims, those claims are likely to be valid for charges that previously went undetected. (Dow Reply 5, July 6, 2020.)

Several rail users counter CSXT's and CN's argument that minimum information requirements would constitute a return to the ICC's former demurrage rules by arguing, among other things, that unlike the former ICC rules, the Board's proposal would have little, if any, impact on the day-to-day operations of railroads because it would not impose timing requirements, content-organization requirements, or recordkeeping or notification methods. (Joint Reply (ACC, CRA, TFI, & NITL) 7-8, Dec. 6, 2019; see also ISRI Reply 12-13, Dec. 6, 2019 (arguing that the Board's proposal is less stringent than the rules the ICC removed); AM Reply 4, Dec. 6, 2019 (asserting that the minimum information requirements “would not be `re-regulatory' ”).)

Additionally, joint commenters (ACC, CRA, TCI, and TFI) and Dow dismiss CP's argument that minimum information requirements would discourage rail users from taking steps to avoid demurrage charges, asserting that rail users would not choose to incur the time and expense of challenging demurrage charges over preventing them in the first place. (Joint Reply (ACC, CRA, TCI, & TFI) 7, July 6, 2020; Dow Reply 5, July 6, 2020.) Joint commenters (ACC, CRA, TCI, and TFI) and NGFA strongly object to carriers' calls for an alternative performance-based standard because they argue it would allow carriers to exclusively determine the information rail users need to assess the validity of demurrage charges and permit carriers to present the information in formats that would limit rail users' ability to use the information to verify demurrage charges. (Joint Reply (ACC, CRA, TCI, & TFI) 14-15, July 6, 2020; NGFA Reply 12, July 6, 2020.)

The Board finds ample support in the record for adoption of minimum information requirements for demurrage invoices. The Board received many comments in this proceeding  17 and in Oversight Hearing on Demurrage & Accessorial Charges, Docket No. EP 754, 18 from rail users asserting that carriers either do not provide sufficient information or do not present the information in a format that allows rail users to effectively verify demurrage charges. The Board is particularly concerned about rail users' assertions that even with significant time and resources devoted to reviewing demurrage invoices, they find erroneous charges overly difficult to detect under carriers' present invoicing practices. ( See Dow Reply 2-3, Dec. 6, 2019; Joint Reply (ACC, CRA, TFI, & NITL) 5-6, Dec. 6, 2019; ISRI Reply 10-11, Dec. 6, 2019.) While it may be true that certain Class I carriers provide more information, or more accessible information, than others, the Board finds that the comments from a diverse array of shippers served by different carriers demonstrate a widespread issue that justifies the imposition of a uniform set of minimum requirements for all Class I carriers. Because CN's proposed flexible “performance-based” alternative standard lacks objective criteria, Class I carriers would be responsible for determining the amount of information sufficient for demurrage invoices in a manner that would likely continue to result in varied practices, some of which may not provide rail users with information sufficient for them to readily assess the validity of demurrage charges. Likewise, CP's argument that market pressure will encourage carriers to provide better information on demurrage invoices is also unpersuasive because, if the argument were correct, demurrage invoices would already be more complete and informative than they are, and this proceeding would not have been necessary in the first place. 19

Regarding CN's and CSXT's concern that minimum information requirements will lead to increased demurrage litigation because rail users will challenge invoices based upon technical issues unrelated to the validity of demurrage charges, the Board clarifies here that a carrier's failure to comply with the minimum information requirements on a particular invoice does not, by itself, establish that the invoice is invalid. Rather, the Board intends for the final rule to reduce unnecessary litigation by providing rail users with information that enables them to readily assess the validity of demurrage charges and determine when to dispute or accept responsibly for them. Indeed, rail users describe demurrage litigation as a complicated and time-consuming process that they would prefer not to undertake. The Board understands that carriers may make occasional invoicing errors and does not expect that an error would conclusively invalidate an entire demurrage invoice. The question of whether specific demurrage charges are lawful depends on an array of fact-specific factors (including, for example, documentation supporting the charges) that would need to be determined in the context of an individual dispute. Nevertheless, the Board has made clear that transparency and mutual accountability in the billing process are “important factors” in the establishment of reasonable demurrage charges. Pol'y Statement on Demurrage & Accessorial Rules & Charges ( Pol'y Statement ), EP 757, slip op. at 15 (STB served Apr. 30, 2020). Although a carrier's failure to comply with the minimum invoicing requirements to be set forth at section 1333.4 would not be conclusive in litigation regarding a particular demurrage invoice, such noncompliance should be taken into account under 49 U.S.C. 10702 and 10746, along with all other relevant evidence, in determining the reasonableness and enforceability of demurrage charges. 20

Contrary to carriers' arguments, the final rule does not contradict Board precedent, the RTP, or the purpose of demurrage. First, the ICC's decision in Maintenance of Records, 367 I.C.C. 145 (1982), cited by CSXT and CN, does not prevent the Board from adopting minimum information requirements here. As an initial matter, the Board may modify its rules as long as its actions are rational and fully explained. 21 Maintenance of Records itself was a modification based largely on changes in carrier practices due to technological advances. There, the ICC determined that certain recordkeeping requirements, such as those requiring carriers to maintain separate records for each open station, prepare daily car reports, and forward the reports daily to recordkeeping offices were unnecessary because computers could retain the data at a central location in a comparably efficient and less expensive way. Maintenance of Records, 367 I.C.C. at 146. As one rail user points out, however, present rail industry practices and technology are very different than they were when the ICC decided Maintenance of Records in 1982. (ISRI Reply 11-12, Dec. 6, 2019.) Moreover, as the Board observed, carriers use the minimum information in the ordinary course of business today and some carriers already provide certain demurrage information to rail users on online platforms. See NPRM, EP 759, slip op. at 10; SNPRM, EP 759, slip op. at 4. Unlike the more prescriptive rules that predated Maintenance of Records, the Board's final rule in this proceeding gives Class I carriers the flexibility to invoice in the format of their choosing, including electronic options, so long as they include the minimum information requirements on or with the invoices and provide machine-readable access to the minimum information. Accordingly, the final rule does not, as CSXT argues, “turn the clock back to long-rejected policies.” (CSXT Comments 3, Nov. 6, 2019.)

The Board also finds that the final rule adopted here is consistent with the provision of the RTP at section 10101(2), which focuses on minimizing the need for Federal regulatory control and ensuring expeditious Board decisions when required. The record in this proceeding and in Oversight Hearing on Demurrage & Accessorial Charges, Docket No. EP 754, supports the conclusion that limited and focused regulation would help parties resolve future demurrage disputes more efficiently and effectively without the need for costly and time-consuming litigation. Additionally, to the extent that parties may need to litigate demurrage disputes in the future, the minimum information requirements adopted here will facilitate expeditious handling and resolution of those disputes, consistent with section 10101(2), (15). Furthermore, by ensuring that rail users have access to sufficient information to understand demurrage charges, the final rule serves important goals of the RTP to meet the needs of the public and for carriers to remain competitive with other transportation modes. See section 10101(4).

The Board rejects CP's argument that the rule contradicts the purpose of demurrage because it encourages rail users to challenge demurrage invoices rather than avoid demurrage charges in the first instance. The final rule incentivizes efficient asset utilization (and helps to ensure that carriers are compensated when rail cars are unduly detained) by requiring demurrage invoices to contain sufficient information to allow rail users to verify the validity of those charges and modify their own behavior when necessary to avoid future demurrage charges. See NPRM, EP 759, slip op. at 10; SNPRM, EP 759, slip op. at 7. The final rule does not encourage rail users to challenge appropriately assessed demurrage charges. Rather, it ensures that rail users are provided sufficient information about the charges to enable them to take action to avoid future charges and, indeed, rail users have confirmed that incurring the time and expense of demurrage litigation, rather than avoiding the charges in the first place, would not serve their interests. ( See Joint Reply (ACC, CRA, TCI, & TFI) 7, July 6, 2020; Dow Reply 5, July 6, 2020.)

Lastly, the Board is not persuaded by the argument that minimum information requirements will stifle innovation. To the contrary, the final rule allows Class I carriers to choose how to invoice rail users, as long as they include the minimum information required on or with the invoices and provide machine-readable access to the minimum information. 22

Therefore, as discussed below, the Board will adopt the minimum information requirements proposed in the NPRM and the SNPRM .

Rail users generally support the minimum information requirements proposed in the NPRM, asserting that the information would help rail users audit invoices more effectively and learn what actions to take to avoid future demurrage charges. ( See, e.g., TFI Comments 3-4, Nov. 6, 2019; NACD Comments 2-3, Nov. 6, 2019; NITL Comments 9, Nov. 6, 2019.) Additionally, two rail users submit requests for clarification. First, ISRI asks the Board to clarify its proposal that Class I carriers be required to provide “[t]he number of credits and debits attributable to each car (if applicable).” (ISRI Comments 4-5, June 5, 2020.) Specifically, ISRI asks whether this proposal would require carriers to “determine in advance for each car included in an invoice whether credits apply” or whether rail users would need to apply for credits that would appear on future invoices, if granted. ( Id. at 5.) ISRI requests that, if the Board did not intend to require the former, then the Board mandate that credits be carried over for 30 to 60 days before expiring. ( Id. ) Second, ILTA asks the Board to change the “and/or” language in the requirement that Class I carriers provide “the identity of the shipper, consignee, and/or care-of party, as applicable” to “and” so that Class I carriers are required to identify all applicable parties. (ILTA Comments 4, June 4, 2020.) Class I carriers do not respond to ISRI's or ILTA's requests for clarification.

Several Class I carriers state that they already provide rail users with most (or all) of the information proposed in the NPRM, either on invoices or their online platforms. 23 However, KCS and CN express concerns that handling carriers may not always receive complete waybill information from connecting carriers and, therefore, may not have access to the date the waybill was created; the identity of the shipper, consignee, and/or care-of party, as applicable (if not the invoiced party); and the origin station and state of the shipment. (KCS Comments 5, Nov. 6, 2019; CN Comments 7-8, June 5, 2020.) KCS states that it is “willing to work with other carriers to try to obtain this information on a regular basis in the future, but currently does not always have all of the information the Board's rules would require.” (KCS Comments 5, Nov. 6, 2019.) CN asks the Board to specify that “a railroad is only required to provide information that is available to it.” (CN Comments 7, June 5, 2020.)

NGFA objects to CN's proposal. NGFA argues that the proposal would create an incentive for carriers to avoid collecting information needed by rail users so that they would not have to provide the information on demurrage invoices. (NGFA Reply 13, July 6, 2020.)

The Board finds that adopting the minimum information requirements proposed in the NPRM will ensure that rail users have access to information that will help them readily assess the validity of demurrage charges, properly allocate demurrage responsibility, and modify their own behavior, as appropriate, to minimize future demurrage charges. Such actions will help provide for the efficient use of rail assets, consistent with section 10746. Accordingly, the Board will adopt the minimum information requirements proposed in the NPRM with the following clarifications.

To address ISRI's concern that the Board's proposal would require rail users to apply for credits, the Board clarifies that the final rule does not create an obligation for rail users to apply for credits. Rather, the Board intends that Class I carriers will list the number of credits and debits attributable to each car on the invoice (if applicable). 24 Furthermore, the Board declines ILTA's request to change the “and/or” language in the requirement that Class I carriers provide “the identity of the shipper, consignee, and/or care-of party, as applicable” because the “as applicable” language already conveys that Class I carriers should identify all applicable parties on the invoice.

In response to KCS's and CN's concerns about access to select waybill information, the Board clarifies that Class I carriers are not required to provide rail users with information to which the Class I carriers do not have access in the normal course of business from their partner carriers. Although CN and KCS do not quantify the degree to which they may lack information from other rail carriers in a movement, the Board would not expect this situation to occur frequently because Class I carriers have many reasons for collecting the minimum information required by the final rule, including for their own performance metrics and to substantiate demurrage charges should they be challenged, and carriers share information in the ordinary course of business during interchange. Where a carrier cannot provide information required by the rule because it has not received the information from another carrier, the invoicing carrier should make a note to that effect on the invoice. In response to NGFA's concern, the Board observes that it expects that this situation would arise infrequently, and the Board will consider further regulatory action if the situation is becoming widespread.

In the SNPRM, the Board invited comment on a proposal to require Class I carriers to include on or with demurrage invoices the billing cycle covered by the invoice. SNPRM, EP 759, slip op. at 5. Many rail users support the inclusion of the billing cycle, asserting that it would make invoices easier to understand and validate. 25 Dow argues that this information would be particularly useful when demurrage events span more than one invoicing period because some carriers bill demurrage monthly by the date it accrues rather than by the date the demurrage event ends. (Dow Comments 2, June 5, 2020.) Dow also contends that billing cycle information would simplify research into invoice events because many of the carriers' online platforms make demurrage event data available only by billing cycle. ( Id. )

CN opposes this requirement, arguing that there is no basis in the record for it. (CN Comments 10, June 5, 2020.) NSR does not object, but requests additional clarity about whether its current process for providing billing cycle information  26 would satisfy the Board's proposed requirement. (NSR Comments 7, June 5, 2020.) In addition, UP, CN, CSXT, and BNSF state that they currently provide billing cycle information on their invoices or online platforms. (UP Comments, V.S. Prauner 2, June 5, 2020; CN Comments 10, June 5, 2020; CSXT Comments 3, June 5, 2020; BNSF Comments 2-3, June 5, 2020.)

The Board will include a billing cycle requirement in the final rule. The billing cycle information, which is a basic feature on recurring invoices, would help rail users verify demurrage charges that span multiple invoicing periods and compare invoiced charges to the demurrage information available on Class I carriers' online platforms. (Dow Comments 2, June 5, 2020.) Although CN opposes the addition, the record establishes that such information would assist rail users in better understanding their invoices; most carriers indicate that they already provide this basic information; and no carriers indicate that this requirement would be burdensome. In response to NSR's request, the Board clarifies here that providing rail users with the dates of the invoicing period over which rail cars incurred demurrage would be sufficient to satisfy the billing cycle requirement.

As discussed in the SNPRM, several commenters identified the original ETA and, if applicable, the date and time that cars are received at interchange, as information that would give rail users greater visibility into how carrier-caused bunching  27 and other delays may affect demurrage charges. See SNPRM, EP 759, slip op. at 5-8 (describing comments received in response to the NPRM related to the original ETA and the date and time that cars are received at interchange). In response, the Board invited additional comment on revisions to proposed section 1333.4 that would require Class I carriers to provide on or with their demurrage invoices (1) the original ETA of each car (as established by the invoicing carrier); and (2) the date and time that each car was received at interchange, if applicable. Id. For the former, the Board invited comment on how to define “original ETA,”  28 and whether the original ETA may differ depending on whether the rail car is loaded or empty. Id. at 7 n.12. For the latter, the Board invited comment on whether the requirement that Class I carriers provide the date and time that cars are received at interchange, if applicable, should be limited to the last interchange with the invoicing carrier. Id. at 7. Lastly, the Board invited comment on whether Class I carriers should be required to provide these items to the invoiced party upon reasonable request (rather than on or with every invoice) and, if so, what would constitute a reasonable request. Id. at 7-8.

Original ETA. In response to the SNPRM, rail users express additional support for an original ETA requirement. Several rail users contend that by comparing the original ETA to the car's arrival time, rail users will be better able to identify carrier-caused bunching, verify credits when applicable, and know when to dispute demurrage charges. 29 AFPM also contends that this requirement would encourage carriers to apply increased scrutiny to their demurrage invoices before sending them. (AFPM Comments 6, June 5, 2020.) Although several rail users acknowledge that they would need to consider other facts and circumstances besides the original ETA to determine whether demurrage charges arise from carrier-caused bunching, they argue that the original ETA would help them determine when to conduct further inquiries with the carriers. (Dow Reply 3-4, July 6, 2020; Joint Reply (ACC, CRA, TCI, & TFI) 3-4, July 6, 2020; NITL Reply 6, July 6, 2020.) Dow and joint commenters (ACC, CRA, TCI, and TFI) argue that the original ETA and date and time of interchange are the only metrics that allow rail users to identify demurrage charges that may arise from carrier-caused bunching and other delays beyond rail users' reasonable control. (Dow Reply 1, July 6, 2020; Joint Reply (ACC, CRA, TCI, & TFI) 3, July 6, 2020.) Additionally, Dow contends that the original ETA would help inform transit variability so that rail users can “fine tune [their] shipments and the number of cars at a destination to better prevent demurrage.” (Dow Reply 4, July 6, 2020.)

BNSF, CP, CN, UP, NSR, and CSXT oppose an original ETA requirement, although they state that they already provide rail users with ETA information. 30 BNSF, CN, and UP argue that ETAs are most useful when they are consistently updated with current information to account for the variability of traffic movements across the rail network. (BNSF Comments 16, June 5, 2020; CN Comments 8-9, June 5, 2020; UP Comments, V.S. Prauner 2, June 5, 2020.) BNSF asserts that rail users may “keep a historical record of the original ETA and any updates as a car moves across the network, but that original ETA is not meaningful to the customer in its demurrage planning” because “actual events” are more important than “historical estimates.” (BNSF Comments 16-17, June 5, 2020.)

BNSF and UP also contend that original ETAs do not give rail users meaningful information about the causes of demurrage. 31 CN asserts that the relevant data for bunching is not the original ETA but rather “the actual arrival time of shipments, and whether the arrival times for all of a receiver's inbound traffic are clustered in a way that it could prevent the receiver from loading or unloading the cars without incurring demurrage.” (CN Comments 9, June 5, 2020.) UP also argues that “[t]he only way to identify whether and why bunching occurred is through the railroad and customer working cooperatively.” (UP Comments 4, June 5, 2020.)

Additionally, Class I carriers assert that an original ETA requirement would create confusion about carriers' service obligations. CP and UP emphasize that that they do not guarantee specific transit times. (CP Comments 5, June 5, 2020; UP Comments 4, June 5, 2020.) Likewise, NSR and BNSF argue that a carrier's common carrier obligation does not require them to adhere to ETAs. (NSR Comments 7-8, June 5, 2020; BNSF Reply 2-3, July 6, 2020.) BNSF also contends that the Board has recognized in demurrage cases that “transit delays inherent in rail operations do not, on their own, excuse a shipper from demurrage.” (BNSF Reply 3-4, July 6, 2020.) Furthermore, CN argues that the Board's proposal could have unintended consequences, such as disrupting private service agreements between carriers and rail users by suggesting that “failure to deliver by an original ETA could be indicative of service failure” and incentivizing carriers to include a “sizeable cushion” rather than the most accurate ETA forecasts to avoid potential demurrage challenges. (CN Comments 9-10, June 5, 2020; see also BNSF Reply 5, July 6, 2020 (asserting that carriers may change the way they estimate ETAs to avoid litigation with rail users).)

Rail users respond by arguing that an original ETA requirement would not cause the outcomes that carriers describe. Several rail users contend that an original ETA requirement would not create transit-time guarantees since rail users know, via carriers' contracts and tariffs, that carriers do not guarantee transit times. (Dow Reply 4, July 6, 2020; ISRI Reply 4, July 6, 2020; Joint Reply (ACC, CRA, TCI, & TFI) 7, July 6, 2020; NITL Reply 5, July 6, 2020.) Dow and joint commenters (ACC, CRA, TCI, and TFI) suggest that the Board could clarify in the final rule that the original ETA is required for demurrage purposes only and not to create a transit-time guarantee. (Dow Reply 4-5, July 6, 2020; Joint Reply (ACC, CRA, TCI, & TFI) 7, July 6, 2020.) However, joint commenters (ACC, CRA, TCI, and TFI) also argue that “[i]f the railroad caused the shipment to arrive late, even by a day, it should not be entitled to penalize the rail user with demurrage.” (Joint Reply (ACC, CRA, TCI, & TFI) 7, July 6, 2020.)

Dow and joint commenters (ACC, CRA, TCI, and TFI) further argue that, for private cars, original ETA should be defined as “the estimated date and time of constructive placement as determined by the delivering carrier immediately upon proper release of a car by the shipper to the rail carrier (for single-line movements) or the carrier's receipt of a car in interchange (for joint-line movements)” as “determined under applicable AAR interchange rules.” (Dow Comments 3, June 5, 2020; see also Joint Comments (ACC, CRA, TCI, & TFI) 4, June 5, 2020 (proposing a similar definition).) They contend that an original ETA must be estimated immediately upon the delivering carrier receiving control of the car because carrier-caused delays can occur at origins and interchanges. (Dow Comments 3, June 5, 2020; Joint Comments (ACC, CRA, TCI, & TFI) 5, June 5, 2020; see also NGFA Reply 9, July 6, 2020 (expressing concerns that a rail car “could sit idle at origin for days or a week or more due to missed pulls with the original trip plan never generated, creating bunching issues at the origin shipper that are not documented by the carrier”).)  32

NSR states that it opposes a rule that would require carriers to develop an ETA immediately upon receipt of a car in interchange, as this would require “the reconfiguration of [NSR's] trip plan software, an incredibly complicated reconfiguration that would take years to implement due to the legacy infrastructure.” (NSR Reply 4-5, July 6, 2020.) CN opposes any definition that would require it to provide an ETA before a trip plan is created. (CN Reply 5, July 6, 2020.)

In the SNPRM, the Board invited comment on whether the original ETA may differ depending on whether the rail car is loaded or empty. SNPRM, EP 759, slip op. at 7. In response, joint commenters (ACC, CRA, TCI, and TFI) assert that different definitions are not necessary because carriers' demurrage rules generally apply the same calculation and constructive placement methods to all private cars. (Joint Comments (ACC, CRA, TCI, & TFI) 5, June 5, 2020.)

The Board is persuaded that the original ETA provides useful information to rail users for verifying credits, when applicable, and identifying delays that impact demurrage. Although not dispositive as to the cause of bunching, original ETAs will allow rail users to better understand whether there are delays in shipment beyond carriers' initial expectations and will lead to better communication between carriers and rail users about the causes of demurrage. Likewise, original ETAs may give rail users more insight into which demurrage charges to probe further to determine whether carrier-caused bunching is present. Furthermore, original ETAs will assist certain rail users in verifying credits because at least one Class I carrier issues credits based on rail cars that do not meet their original ETAs. ( See NSR Reply 2-3, July 6, 2020.) Given these benefits and the fact that carriers already generate original ETAs in the ordinary course of business, inclusion of the original ETA as a minimum requirement is appropriate.

The Board rejects the argument that updated (or “real-time”) ETAs render original ETAs less useful. The Board recognizes that updated ETAs help rail users account for transit variability and plan for rail cars' arrival; however, they may be less useful when rail users need to verify demurrage charges on invoices that may be issued weeks later. In contrast, allowing rail users to readily compare significant deviations between original ETAs and car arrivals once invoices are issued could lead to better information exchange about the causes of delay. BNSF states that rail users may record original ETAs and updates as needed from the information carriers provide on their online platforms. However, the Board finds it unreasonable to expect rail users to keep records of fluctuating ETAs for all rail cars to prepare for the possibility that some of those rail cars ultimately accrue demurrage. As discussed in the NPRM, minimum information requirements are intended to ensure that rail users do not need to undertake unreasonable efforts to gather information that can be provided by carriers in the first instance. NPRM, EP 759, slip op. at 10.

The Board agrees with the Class I carriers' assertion that rail cars may not be delivered by their original ETAs due to a variety of causes, including rail users' behavior, carrier-caused delays, or other variables. Accordingly, a missed original ETA would not—without more— establish that carrier-caused bunching (or any other event) occurred but rather would give rail users information about delays that may then prompt them to conduct further investigations or adjust their own conduct to better account for transit variabilities and avoid future demurrage charges. In any given case, additional facts and circumstances would need to be considered in determining whether demurrage charges arise from carrier-caused bunching. The fact-specific nature of bunching issues is precisely why the Board has determined that demurrage disputes pertaining to bunching are best addressed in individual cases. See Demurrage Liability, EP 707, slip op. at 23-24; see also Pol'y Statement, EP 757, slip op. at 11-12.

The Board recognizes Class I carriers' concern that rail users may misinterpret original ETAs as guaranteed transit times or as a service standard that would override private agreements between rail users and carriers, and clarifies here that that is not the purpose or effect of the original ETA requirement. The requirement to provide an original ETA established here obligates carriers only to provide rail users with this information on or with demurrage invoices; it does not constitute, or require carriers to provide, service guarantees. The requirement does not create a separate service standard for carriers. Finally, inclusion of an original ETA requirement in the final rule does not change the fact that the Board will determine whether demurrage charges are reasonable under section 10702 and comport with the statutory requirements specified in section 10746 in the context of case-specific facts and circumstances. Accordingly, the existence of the original ETA in the minimum information requirements does not establish whether a delay in shipment renders a demurrage charge unreasonable. ( See Joint Reply (ACC, CRA, TCI, & TFI) 7, July 6, 2020.)  33

With respect to CN's and BNSF's assertions that an original ETA requirement would incentivize carriers to cushion their ETA forecasts, the Board expects that Class I carriers have other motivations to give rail users accurate estimates about rail car arrivals, including to provide good customer service, improve their performance metrics, 34 and ensure that the rail network runs efficiently by giving rail users the best opportunity to plan for the efficient use of rail assets. Although it may be appropriate to make adjustments to ETAs based on real-time information, providing the most accurate estimates available is in the interest of both rail carriers and their customers.

The Board will adopt the definition of original ETA discussed in the SNPRM, which will require Class I carriers to generate ETAs promptly following interchange or release of shipment to the invoicing carrier based on the first movement of the invoicing carrier. SNPRM, EP 759, slip op. at 7. The Board declines to adopt the proposal to require carriers to generate ETAs “immediately” following interchange or release of shipment since the inclusion of the word “promptly” in the definition is sufficient to ensure that there is not undue delay at origin or interchange before ETAs are created for rail cars. The Board also expects that it would not be in Class I carriers' interests, from an efficiency standpoint, to hold rail cars at their yards without trip plans. 35

Interchange Date and Time. Many rail users also support a requirement that Class I carriers provide the date and time that cars are received at interchange, asserting that such information would be useful in identifying upstream carrier-caused bunching. 36 ILTA argues that “having the interchange information would allow rail users to calculate transit time on an upstream carrier's line and allow impacted users to credibly approach the upstream carrier to take responsibility for delays it may have caused.” (ILTA Comments 2, June 4, 2020.) Dow asserts that it would use the date and time of interchange, along with the original ETA, to “identify circumstances that may warrant a deeper inquiry into whether demurrage charges arise from carrier-caused bunching and delays beyond Dow's reasonable control.” (Dow Reply 3-4, July 6, 2020.) NITL acknowledges that “there can be multiple factors causing car delays that result in demurrage” and argues that interchange information, along with original ETAs, “would assist rail customers and railroads in their investigations of invoiced charges.” (NITL Reply 6, July 6, 2020.)

Dow and joint commenters (ACC, CRA, TCI, and TFI) contend that Class I carriers should provide the date and time for every interchange. (Dow Comments 3, June 5, 2020; Joint Comments (ACC, CRA, TCI, & TFI) 5, June 5, 2020.) ISRI and NITL state that information about all interchanges would be helpful but ask that, at a minimum, the Board require Class I carriers to provide information about the last interchange with the invoicing carrier. (ISRI Comments 6, June 5, 2020; NITL Comments 4, June 5, 2020.)

BNSF, CSXT, and NSR state that they provide rail users with the date and time of interchange on their online platforms. (BNSF Comments 18, June 5, 2020; CSXT Comments 3, June 5, 2020; NSR Comments 8, June 5, 2020.) However, BNSF argues that an invoice requirement “would be counterproductive as it would create confusion over the relevance of such data and potentially encourage unnecessary disputes over appropriate demurrage charges.” (BNSF Comments 16, 18, June 5, 2020.) BNSF, UP, and NSR assert that they do not use interchange information to calculate demurrage. (BNSF Comments 18, June 5, 2020; UP Comments 4, June 5, 2020; NSR Reply 2-3, July 6, 2020.) Additionally, NSR argues that this requirement, if adopted, should be limited to the last interchange since it “has no visibility into the operations of its interchange partners, and does not have access to information regarding any trip plan or ETA that may have been generated upstream by other carriers.” (NSR Comments 8, June 5, 2020.)

The Board will require Class I carriers to provide, on or with demurrage invoices, the date and time they received rail cars at interchange, if applicable. The Board finds that interchange information may assist rail users in identifying where delays occurred on joint-line movements, which would in turn allow rail users to know when to adjust their own conduct to account for upstream transit variabilities and conduct further inquiries when necessary. These further inquiries may be especially important when demurrage disputes involve concerns about upstream bunching. See Pol'y Statement, EP 757, slip op. at 11-12. As with the original ETA, however, the Board clarifies that the date and time of interchange does not establish whether upstream bunching occurred and, instead, must be considered in the context of other relevant facts and circumstances.

The Board will limit this requirement to the last interchange with the invoicing carrier. In the SNPRM, the Board stated that Class I carriers would likely have access to the date and time of interchange because this information is used in the ordinary course of business to track car movement and place cars. SNPRM, EP 759, slip op. at 7. According to NSR, Class I carriers do not have access to information about upstream interchanges with other carriers in the ordinary course of business; accordingly, the Board will limit this requirement to the information that Class I carriers can provide without the potential burden of having to consult with other carriers.

Reasonable Request Proposal. Rail users and rail carriers that commented on the Board's alternative proposal to require carriers to provide original ETA and date and time of interchange only upon reasonable request oppose the proposal. Several rail users argue that a reasonable request provision would be burdensome and cause unnecessary delays in collecting information. 37 CSXT and UP also contend that a reasonable request provision is unnecessary because rail users can access the original ETA and date and time of interchange on demand through their online platforms. (CSXT Comments 3, June 5, 2020; UP Comments 5, June 5, 2020.) The Board will not include a reasonable request provision in the final rule because the comments offer no indication that it would benefit rail users or Class I carriers.

Rail users identified the date and time that cars are ordered into a rail user's facility as information that would help them validate invoices more efficiently. In response, the Board invited comment in the SNPRM on a modification to proposed section 1333.4 that would require Class I carriers to provide the ordered-in date and time on or with demurrage invoices. SNPRM, EP 759, slip op. at 8-9.

Rail users replied that access to the ordered-in date and time would allow them to verify demurrage invoices more efficiently by comparing carriers' information to their own records and determining the basis for carriers' demurrage assessments, understand how their own actions impacted the demurrage charges, and calculate credits, if applicable. 38 Dow also emphasizes that this information is a “crucial demurrage metric because demurrage stops accruing at that time” and reiterates that including this information is consistent with the Board's proposal to require the date and time of constructive placement, at which the accrual of demurrage starts. (Dow Comments 5, June 5, 2020.) Joint commenters (ACC, CRA, TCI, and TFI) support this requirement but argue that carriers should be required to provide “the actual date and time the carrier receives the order to place the cars at the receiving facility” since at least one carrier “appears to purposefully record a different ordered-in date and time in its system.” (Joint Comments (ACC, CRA, TCI, & TFI) 7, June 5, 2020.)

In response, CP argues that the Board should not require carriers to provide the date and time the rail user places the order in all circumstances because this information would not always impact the demurrage calculation and could cause administrative confusion. (CP Reply 4-5, July 6, 2020.) For example, CP explains that it allows rail users that operate closed-gate facilities to order in cars while the cars are en route, for which CP records the date that the cars arrive at the serving yard and are available for placement. ( Id. at 3.) In this scenario, CP states that the order for placement upon arrival keeps the demurrage clock from starting. ( Id. at 3-4.) Furthermore, CP states that it allows certain rail users to order in cars for the current day and up to three days in the future and, in these circumstances, records the date selected for car placement because this date stops the accrual of demurrage. ( Id. at 4.) Likewise, NSR states that it provides rail users with the “effective order date” on invoices, which is the date “selected by the customer from their service schedule” and “represents the date the railcar is to be delivered.” (NSR Comments 8, June 5, 2020.) NSR states that the ordered-in date and time that a rail user enters online is not used for demurrage purposes but is provided in an order confirmation email. ( Id. )

CN states that it already provides rail users with the ordered-in date and time but objects to the inclusion of this requirement because rail users would already have this information in their own records. (CN Comments 11, June 5, 2020.) CN also argues that the disputes the Board references in the SNPRM   39 would not be resolved by access to ordered-in date and time information. ( Id. ) CSXT, UP, and BNSF also indicate that, if applicable, they provide the ordered-in date and time on their invoices or online platforms. (CSXT Comments 3, June 5, 2020; UP Comments, V.S. Prauner 2, June 5, 2020; BNSF Comments 2-3, June 5, 2020.)

The comments received in response to the SNPRM do not change the Board's view that the ordered-in date and time, which is essential to the calculation of demurrage at closed-gate facilities, would be valuable on or with demurrage invoices for both demurrage accrual and verification purposes. See SNPRM, EP 759, slip op. at 8. Although, as CN points out, rail users may record the ordered-in date and time themselves, the Board finds that documentation of the ordered-in date and time, which would stop the accrual of demurrage, would be very useful when viewed along with the other information on demurrage invoices, including the event that starts demurrage accrual and the resulting credits and charges, as applicable. Additionally, as rail users explain, having access to the ordered-in date and time recorded by Class I carriers may help rail users identify discrepancies between the carrier's records and the rail user's records. CN argues that the issues stakeholders raised in comments and testimony in Oversight Hearing on Demurrage & Accessorial Charges, Docket No. EP 754, which the Board referenced in the SNPRM, EP 759, slip op. at 9, would not be resolved through the ordered-in date and time, but the Board did not state that the ordered-in date and time would be dispositive in these or any other specific disputes. Rather, as the Board stated in the SNPRM, the ordered-in date and time requirement is intended to give rail users easier access to information for their own verification purposes. Id. Furthermore, the comments from six Class I carriers stating that they currently provide rail users with the ordered-in date and time confirm that providing this information would not be overly burdensome for Class I carriers.

Since Class I carriers' comments demonstrate that the date and time the carrier receives the order from the rail user to place the cars is not used to calculate demurrage in all circumstances, the Board will not define the ordered-in date and time requirement as narrowly as joint commenters request. ( See Joint Comments (ACC, CRA, TCI, & TFI) 7, June 5, 2020.) Rather, the ordered-in date and time will mean the date and time at which demurrage first stops accruing with respect to a closed-gate facility. Depending on the carrier's individual system, this may be the date and time the carrier receives the order to place cars from the rail user, the date selected by the rail user for car placement, or another similar metric. 40

In addition to the proposals discussed above, rail users identify an array of other information that they contend would be useful on demurrage invoices. 41 In response, CSXT states its general opposition to the additional items. (CSXT Reply 2, Dec. 6, 2019.) Moreover, CSXT, NSR, and UP argue that online platforms are the best way to provide an array of information. ( See CSXT Reply 4, Dec. 6, 2019 (stating that CSXT already provides almost all of the requested information “through one of its various platforms”); NSR Reply 1, Dec. 6, 2019 (stating that NSR provides most of the requested information online); UP Reply 3, Dec. 6, 2019 (arguing that an online platform can meet rail users' needs in a “customized, tailored way”).)

The Board declines to incorporate additional items beyond those discussed in the NPRM and SNPRM into the final rule at this time. The Board's minimum information requirements are not intended to encompass every piece of information that may be useful to rail users or that may bear on demurrage. Rather, the minimum information adopted in the final rule represents what the Board has determined will have the greatest impact on rail users' ability to validate demurrage charges, properly allocate demurrage responsibility, and modify their behavior if their own actions led to the demurrage charges. Many of the other items suggested by rail users would provide additional detail about the movement of rail cars but are not as central to an initial assessment of demurrage charges as the minimum information requirements adopted here. Moreover, in adjudicated cases, parties may seek discovery to gain further information about the causes of delays and demurrage. Several Class I carriers indicate that they provide, in some format, much of the information rail users identified in their comments, and the Board encourages them to continue to do so.

In the NPRM, the Board invited comments on the adequacy of other billing or supply chain visibility tools or platforms (other than invoices or accompanying documentation) to provide rail users with access to the proposed minimum information. NPRM, EP 759, slip op. at 9 n.13. In response, Class I carriers state that their existing online platforms provide rail users with most (or, in some cases, all) of the information that the Board proposes. 42 UP asserts that its online platform benefits rail users by allowing them to create custom reports with information unique to their needs. (UP Comments 5, Nov. 6, 2019.) CP indicates that its online platform provides rail users with access to current information as shipments move across the rail network, as well as the ability to log concerns in real time, which obviates the need “to review historical information to identify improper demurrage charges due to railroad-caused bunching.” (CP Comments 3, June 5, 2020.)

Class I carriers ask that, if the Board adopts minimum information requirements, the Board allow them to provide the information on their online platforms, instead of on or with invoices. 43 CSXT argues that the demurrage information currently on its online platform is “easily accessible” and contends that if the Board were to require carriers to provide all of the required information on the invoice itself or determine that “software platforms are acceptable only if all information is made available or downloadable in one central location,” it would have to undertake a substantial and costly software redesign. (CSXT Comments 6, June 5, 2020.) UP also argues that invoices with all of the minimum information the Board proposes would not be useful to rail users since “[a] combination of too many fields and fields that are irrelevant to most customers will make invoices cluttered and unreadable.” (UP Comments 5, June 5, 2020.) Likewise, CSXT objects to a rule that would require it to compile the information into one invoice document because its physical invoice “is already challenged in terms of available physical space” and “[i]t would be difficult to add additional categories without rendering the invoice unreadable.” (CSXT Comments 8 & n.17, June 5, 2020.)

Conversely, several rail users describe carriers' current online platforms as impractical and cumbersome. (WCTL & SEC Comments 10, Nov. 6, 2019; Joint Reply (ACC, CRA, TFI, & NITL) 2, Dec. 6, 2019; Dow Reply 3, Dec. 6, 2019; ISRI Reply 2, July 6, 2020.) Joint commenters (ACC, CRA, TFI, and NITL) explain that locating information is a “multistep process” in which “[a] customer cannot simply enter a demurrage invoice number and download a report of all of the car-event data for each car on the invoice” but rather must access information for each car separately and often in multiple locations on the carrier's online platform. (Joint Reply (ACC, CRA, TFI, & NITL) 3, Dec. 6, 2019.) Dow specifically describes online portals belonging to four Class I carriers as cumbersome and identifies obstacles rail users may face in auditing demurrage invoices on these platforms, such as needing to search for certain information on a car-by-car basis, manually enter car marks, and navigate through multiple pages on the portal to access demurrage data. (Dow Reply 3-6, Dec. 6, 2019.) Likewise, PCA argues that since carriers are “far from consistent in the level of information provided, the ease of access of that information, and the transparency of their demurrage procedures,” rail users are often forced to “cobble together” the information on carriers' online platforms. (PCA Comments 2, June 5, 2020.) Nonetheless, certain rail users state that they do not object to Class I carriers providing the minimum information on their online platforms if they provide it in a format that rail users can download into a single, machine-readable file. (Dow Reply 8, July 6, 2020; ISRI Reply 2, 5, July 6, 2020.)

The record belies Class I carriers' claims that their current online platforms are more useful to rail users than invoices with minimum invoicing requirements. Rail users state they must, in many cases, search for, organize, and consolidate the information themselves from multiple locations on Class I carriers' online platforms. The final rule will ensure that rail users need not make unreasonable efforts to access basic information necessary to efficiently review and validate their demurrage invoices. In addition, Class I carriers will have flexibility to provide the minimum information either on the invoices or with the invoices as accompanying documentation. Furthermore, since demurrage issues may not be apparent until rail cars are delivered and demurrage is charged, the Board is unconvinced by CP's argument that allowing rail users to submit concerns on an online portal while shipments are in transit eliminates the need to review information on or with demurrage invoices.

Accordingly, the Board determines that Class I carriers must provide the minimum information described in section 1333.4 on demurrage invoices or with demurrage invoices as accompanying documentation. Class I carriers may provide the invoices as paper invoices, invoices attached to emails, invoices that are accessible on their online platforms, or other similar formats where the information is consolidated. 44

In response to the NPRM, many rail user commenters voiced a preference for “machine-readable” data containing the minimum information. See SNPRM, EP 759, slip op. at 9-10 (describing comments received in response to the NPRM related to machine-readable data). The Board, therefore, invited additional comment on matters associated with modifying its regulations to require Class I carriers to provide rail users access to machine- readable data in a format to be chosen by the individual Class I carrier, such as a machine-readable invoice, a separate electronic file containing machine-readable data, or a customized link so rail users could directly download data in a machine-readable format. Id. at 10. The Board also invited comment on ways to prevent information inaccessibility for rail users without resources for coding or new upfront costs, and on any other issues pertaining to the accessibility of machine-readable data for small rail users. Id. Finally, the Board invited comment on how to define “machine-readable,” including the following definition proposed by commenters: “A structured data file format that is open and capable of being easily processed by a computer. A format is open if it is not limited to a specific software platform and not subject to restrictions on re-use.” Id.

In response to the SNPRM, rail users broadly support a requirement for machine-readable data, arguing that it will allow rail users to analyze demurrage invoices more efficiently and effectively by reducing the need for manual review, which is resource-intensive and imprecise. 45 Moreover, AFPM states that it supports the flexible compliance options identified by the Board, while other rail users request specific formatting requirements. (AFPM Comments 8, June 5, 2020; Joint Comments (ACC, CRA, TCI, & TFI) 9, June 5, 2020 (requesting machine-readable invoices by email as attachments or direct links); NGFA Comments 7, June 5, 2020 (requesting customized links to machine-readable data); NITL Reply 7, July 6, 2020 (requesting machine-readable data in a “single centralized location”).) In addition to machine-readable data, NGFA contends that rail users should be able to request paper or PDF invoices, (NGFA Comments 6, June 5, 2020), and NACD argues that invoices should “continue to be available in standard format” since small rail users would find analyzing machine-readable data difficult and costly, (NACD Comments 4-5, June 5, 2020).

BNSF, NSR, CP, CN, and UP state that they already provide rail users with some form of machine-readable data. 46 NSR states that it supports the use of machine-readable data but urges the Board not to make a requirement “so prescriptive that it would stifle carrier and technological innovation on carriers' online platforms.” (NSR Comments 2, 6, June 5, 2020.) CSXT states that it does not oppose providing machine-readable data on its online platform as long as “the Board does not mandate any particular format or require that the information be provided in one place only or in a single data file.” (CSXT Comments 6, June 5, 2020.) UP also asks the Board to specify that carriers can meet the machine-readable data requirement by making data available to rail users via their online platforms. (UP Comments 5-6, June 5, 2020.)

Regarding the definition of “machine-readable,” joint commenters (ACC, CRA, TCI, and TFI) and NITL support the definition proposed by some commenters in response to the NPRM:   47 “a structured data file format that is open and capable of being easily processed by a computer. A format is open if it is not limited to a specific software platform and not subject to restrictions on re-use.” (Joint Comments (ACC, CRA, TCI, & TFI) 8, June 5, 2020; NITL Comments 6, June 5, 2020.) Joint commenters (ACC, CRA, TCI, and TFI) contend that this definition would obviate the need for special coding and, therefore, ensure that small rail users can access machine-readable data. (Joint Comments (ACC, CRA, TCI, & TFI) 8, June 5, 2020.) NGFA agrees with this definition but would add the condition that a format is open if it “can be read and interpreted automatically by a computer program without the need for manual intervention.” (NGFA Comments 6-7, June 5, 2020.) In response to the SNPRM, Dow proposes an alternative definition, suggesting that the Board adopt a definition similar to the one used for the Federal Information Policy  48 and define machine-readable as “an open format that can be easily processed by computer without human intervention while ensuring no semantic meaning is lost.” (Dow Comments 6, June 5, 2020.) UP argues that the Board should specify that CSV and Excel files meet the definition of machine-readable data. (UP Comments 5-6, June 5, 2020.)

Rail users convincingly argue that machine-readable data will facilitate efficient auditing by allowing them to validate invoices electronically, thereby reducing the time and resources they must dedicate to manual review. Furthermore, Class I carriers appear to recognize the benefits of machine-readable data, as most provide some machine-readable data now or plan to do so in the future. Accordingly, the Board will adopt a machine-readable data requirement to ensure that all rail users have the option to access machine-readable data containing the minimum information discussed above. As proposed in the SNPRM, the Board will give Class I carriers the discretion to determine how to provide rail users with access to machine-readable data, such as, for example, through a machine-readable invoice, a separate electronic file, a customized link, or another similar option. SNPRM, EP 759, slip op. at 10.

The Board will adopt a definition for machine-readable data that is “data in an open format that can be easily processed by computer without human intervention while ensuring no semantic meaning is lost.” This definition, which is similar to the definition referenced in the SNPRM, is also consistent with the definition adopted for the Federal Information Policy at 44 U.S.C. 3502(18). However, unlike the Federal Information Policy definition, the Board's definition specifies that the data must be provided in an “open format,” to be defined as “a format that is not limited to a specific software program and not subject to restrictions on re-use” so that Class I carriers may choose the program with which to provide machine-readable data in an open format ( e.g., CSV). The open format will also ensure that rail users will not need access to specific software programs to process the data. Moreover, to accommodate those small rail users that state that they would find machine-readable data difficult to manage, the requirement for Class I carriers to provide machine-readable data to rail users will be in addition to, not in lieu of, the requirement to provide the minimum information on or with their standard invoices, as discussed above. 49 The text is set forth in new section 1333.5.

In the NPRM, the Board proposed to require Class I carriers to “take appropriate action to ensure that the demurrage charges are accurate and warranted” prior to sending demurrage invoices. NPRM, EP 759, slip op. at 10. In response to commenters' concerns that this provision would create more uncertainty and potential litigation over its meaning, the Board invited further comment in the SNPRM from Class I carriers about the actions they currently take, and from all stakeholders about the actions Class I carriers reasonably should be required to take, to ensure that demurrage invoices are accurate and warranted. SNPRM, EP 759, slip op. at 10-11.

In response to the SNPRM, rail users propose a variety of actions that they argue Class I carriers should be required to take to ensure invoice accuracy, such as establishing auditing procedures, 50 showing how charges are calculated;  51 providing supporting documentation, 52 offering concise explanations for the charges, 53 certifying practices to the Board, 54 consulting with rail users, 55 and ensuring the accuracy of crew reporting. 56

Class I carriers continue to oppose the appropriate-action proposal as unnecessary and overly restrictive. CN argues that the Board should not mandate any minimum level of appropriate action since carriers “should have flexibility to exercise judgment to pursue an approach that works for [their] particular circumstances, including whether to reasonably rely on technological innovations to enhance accuracy or to enlist more manual review.” (CN Comments 13, June 5, 2020.) CSXT contends that the proposed requirement “places carriers in an untenable situation, as they may either fall short of a vague standard of `appropriateness' or be unable to utilize the prescribed solutions that the Board mandates to ensure accuracy.” (CSXT Comments 9, June 5, 2020.) UP contends that an appropriate-action requirement is unnecessary since it already has “achieved a 95% accuracy rate.” (UP Comments 7, June 5, 2020.) Additionally, BNSF, CN, NSR, and UP detail the actions they currently take to ensure invoice accuracy. (BNSF Comments 14-15, June 5, 2020; CN Comments 12-13, June 5, 2020; NSR Comments 9-10, June 5, 2020; UP Comments 6, June 5, 2020.)

Upon considering the comments on this issue, the Board is persuaded that the proposed appropriate-action requirement should not be adopted in the final rule. Class I carriers convincingly argue that the proposed requirement lacks sufficient detail. Because there are many different reasonable ways to facilitate invoice accuracy, and because deciding whether a particular method is reasonable may depend on a carrier's individual systems and procedures for auditing invoices and potential future advancements in technology, the Board also declines to adopt the specific requirements proposed by rail users.

For these reasons, the final rule adopted in this decision will not include the proposed appropriate-action requirement. Nevertheless, existing requirements, including those at 49 U.S.C. 10702 and 10746, continue to apply to carriers' demurrage invoicing practices. As the Board has made clear previously, it expects that all carriers will take reasonable actions to ensure the accuracy of their invoicing processes and that their demurrage charges are warranted. See Pol'y Statement, EP 757, slip op. at 15-16 (emphasizing that the Board expects rail carriers to “bill for demurrage only when the charges are accurate and warranted, consistent with the purpose of demurrage,” and that rail users should be able to review and dispute charges without incurring undue expense). That being so, the Board strongly encourages carriers to adopt rail users' suggested actions where warranted and practicable, such as conducting regular audits, consulting with rail users when necessary, and providing additional information upon reasonable request. Class I carriers' invoicing protocols and procedures should be considered, in the context of all other relevant facts and circumstances, when determining whether demurrage charges are reasonable and enforceable in individual cases.

Rail users make a variety of other requests, including asking the Board to set a timeframe for Class I carriers to issue invoices, establish dispute resolution procedures, impose penalties for noncompliance with the rule, and apply the rule to accessorial charges. In addition, UP asks that the Board establish a separate process by which Class I carriers can obtain waivers from the final rule. The Board will discuss each of these requests below.

Several rail users ask the Board to set time limits for invoice issuance, ( see NCTA Comments 3-4, Nov. 6, 2019; FRCA Comments 5, Nov. 6, 2019), take further action with respect to dispute resolution, 57 and impose penalties for carriers that issue demurrage invoices that do not comply with the rule, ( see FRCA Comments 5-6, Nov. 6, 2019; FRCA Comments 2, June 5, 2020). No Class I carrier responds directly to these requests.

The Board will not pursue these requests at this time. The Board notes that, by separate decision, it provided guidance on the general principles it expects to consider when evaluating the reasonableness of carriers' invoicing timeframes in future cases and discussed requests to establish additional dispute resolution procedures. 58 See Pol'y Statement, EP 757, slip op. at 16 n.50, 17. With respect to the issue of penalties, the Board expects that Class I carriers will make a concerted effort to comply with the requirements of the rule and finds that it is premature to address specific penalties for non-compliance at this time. 59

NAFCA, AM, and NGFA ask the Board to apply the minimum information requirements to accessorial charges. (NAFCA Comments 2, Nov. 6, 2019; AM Comments 7, Nov. 6, 2019; NGFA Comments 2, June 5, 2020.) Class I carriers did not comment on this issue.

The Board declines to extend the final rule to accessorial charges at this time. There are many kinds of accessorial charges and some, such as those imposed for weighing rail cars or requests for special trains, do not serve the same efficiency-enhancing purpose as demurrage. In their comments, rail users do not identify any specific accessorial charges to which the minimum information requirements should apply, or otherwise justify the extension of the final rule to accessorial charges generally. The Board encourages Class I carriers to provide the minimum information for those accessorial charges designed to enhance the efficient use of rail assets to the extent practicable. Should sufficient evidence be presented in the future that invoicing issues are arising with respect to specific accessorial charges, the Board can revisit this issue and propose any warranted modifications to the rule.

UP requests that, if the Board adopts minimum information requirements, then it also establish a process whereby carriers could obtain waivers from the rule by “attesting that either all of the required information is provided to customers or explain why a particular data set is not provided or unavailable.” (UP Comments 7, June 5, 2020.) UP asserts that this process would need to take place prior to the final rule's effective date so that carriers know whether they need to reprogram their systems. ( Id. )

NGFA objects to this suggestion, arguing that UP's waiver idea is impractical since it would require extensive Board monitoring to ensure that carriers' online platforms do not become noncompliant after waivers had been granted. (NGFA Reply 14-15, July 6, 2020.)

The Board declines to adopt UP's proposal. Pursuant to 49 CFR 1110.9, “[a]ny person may petition the Board for a permanent or temporary waiver of any rule,” and UP fails to explain why the Board's established waiver process is not sufficient to address its concerns. Furthermore, absent unique circumstances, the Board does not anticipate that the waiver process would be used to allow Class I carriers to provide the minimum information by means other than on or with an invoice as described above.

Several Class I carriers request specific amounts of time to comply with the final rule. NSR asks for a minimum of three months to complete its reprogramming, and KCS requests at least six months. (NSR Comments 1, Nov. 6, 2019; KCS Comments 6-7, Nov. 6, 2019.) CSXT contends that if it is required to implement a software redesign “sooner than nine months from the Board's decision,” it will need to delay or reprioritize current projects. (CSXT Comments 7-8, June 5, 2020.) CP likewise states that it could comply within six months but requests at least one year to “minimize disruption to existing projects and allow CP to prioritize its use of its resources appropriately.” (CP Comments 6, June 5, 2020.)

The Board will allow Class I carriers until October 6, 2021, to provide the minimum information on or with demurrage invoices and comply with the machine-readable data requirement, as this timeframe allows Class I carriers a significant amount of time for reprogramming while also ensuring that rail users can benefit from improved demurrage invoicing practices without extended delay.

In the NPRM, the Board explained that it did not propose to require Class II and Class III carriers to comply with the rule because the demurrage issues raised by stakeholders before the Board predominantly pertained to Class I carriers and compliance costs would be more difficult for smaller carriers. NPRM, EP 759, slip op. at 10. The Board invited comment on the proposed exclusion of Class II and Class III carriers. Id.

Although some rail users recognize that demurrage issues most frequently involve Class I carriers, ( see AFPM Comments 8, Nov. 6, 2019; ISRI Comments 10, Nov. 6, 2019), several express concerns about excluding Class II and Class III carriers, 60 particularly those with larger, more sophisticated operations, ( see FRCA Comments 5, Nov. 6, 2019; AFPM Comments 8, Nov. 6, 2019). ISRI urges the inclusion of Class II and Class III carriers for uniformity across the industry, ( see ISRI Comments 10, Nov. 6, 2019; ISRI Comments 10-11, June 5, 2020), and others fear that Class I carriers will seek to evade the rule by tasking Class II and Class III carriers with demurrage invoicing where possible, ( see NITL Comments 10, Nov. 6, 2019; AF&PA Comments 10, Nov. 6, 2019). ILTA acknowledges that Class II and Class III carriers have fewer resources to comply with the rule but argues that small carriers should nonetheless be required to comply since small rail users must pay demurrage charges. (ILTA Comments 4, June 4, 2020.) Some rail users suggest that the Board should apply the rule to all carriers and grant waivers on a case-by-case basis to accommodate the smallest carriers. (NITL Comments 10, Nov. 6, 2019; AF&PA Comments 10, Nov. 6, 2019; AM Reply 5-6, Dec. 6, 2019.) Others suggest that the Board exclude some or all Class III carriers from the rule, but not Class II carriers. (AFPM Comments 8, Nov. 6, 2019 (exclude all Class III carriers, but not Class II carriers); FRCA Comments 5, Nov. 6, 2019 (require Class II carriers and Class III carriers affiliated with large holding companies to comply).)  61

ASLRRA supports the Board's proposal to exclude Class II and Class III carriers. (ASLRRA Comments 3, Nov. 6, 2019; ASLRRA Reply 4, July 6, 2020.) It asserts that more than half of small carriers operate as handling line carriers and, as such, do not always receive all of the information the Board would propose to include in the minimum information requirements from connecting Class I carriers. (ASLRRA Comments 3, Nov. 6, 2019.) ASLRRA further contends that rail users' proposed additions “would place an insurmountable burden on [small rail carriers].” (ASLRRA Reply 4, Dec. 6, 2019.) ASLRRA argues that the suggestion that small carriers could file for individual waivers is unworkable since the waiver process would be too expensive and time-consuming for small carriers with limited resources. ( Id. at 7.) ASLRRA also dismisses rail users' concerns that Class I carriers would assign demurrage invoicing to small carriers to avoid the rule, arguing that Class I carriers will not “want to cede the control of their operations or practices to others or the compensation they receive for the misuse of their rail assets.” ( Id. at 8.)

Nothing in this record undercuts the Board's initial view that the demurrage issues raised by stakeholders in Oversight Hearing on Demurrage & Accessorial Charges, Docket No. EP 754, predominantly pertain to Class I carriers. See NPRM, EP 759, slip op. at 10, 11. Nor do the comments provide a basis for concluding that Class I carriers will seek to avoid the rule by assigning their demurrage invoicing to small carriers. 62 The case-by-case waiver approach for Class II and III carriers suggested by some rail users could be impractical and unduly burdensome for small carriers (and may be problematic for some Class II carriers, where there is a range of capabilities). For these reasons, the Board will not adopt the proposals to make Class II carriers, and, under some proposals, certain Class III carriers, subject to the rule. The Board does, however, strongly encourage Class II and Class III carriers to comply with the rule to the extent they are able to do so. 63

Consistent with this decision, the Board adopts a final rule that requires Class I carriers to include certain minimum information on or with demurrage invoices and provide machine-readable access to the minimum information. The final rule is set out in full below and will be codified in the Code of Federal Regulations.

The Regulatory Flexibility Act of 1980 (RFA), 5 U.S.C. 601-612, generally requires a description and analysis of new rules that would have a significant economic impact on a substantial number of small entities. In drafting a rule, an agency is required to: (1) Assess the effect that its regulation will have on small entities, (2) analyze effective alternatives that may minimize a regulation's impact, and (3) make the analysis available for public comment. Sections 601-604. In its final rule, the agency must either include a final regulatory flexibility analysis, section 604(a), or certify that the proposed rule would not have a “significant impact on a substantial number of small entities,” section 605(b). Because the goal of the RFA is to reduce the cost to small entities of complying with federal regulations, the RFA requires an agency to perform a regulatory flexibility analysis of small entity impacts only when a rule directly regulates those entities. In other words, the impact must be a direct impact on small entities “whose conduct is circumscribed or mandated” by the proposed rule. White Eagle Coop. v. Conner, 553 F.3d 467, 480 (7th Cir. 2009).

As discussed above, the final rule will apply only to Class I carriers. Accordingly, the Board again certifies under 5 U.S.C. 605(b) that this rule would not have a significant economic impact on a substantial number of small entities as defined by the RFA. 64 A copy of this decision will be served upon the Chief Counsel for Advocacy, Office of Advocacy, U.S. Small Business Administration, Washington, DC 20416.

In this proceeding, the Board is modifying an existing collection of information that is currently approved by the Office of Management and Budget (OMB) under OMB Control No. 2140-0021. In the NPRM, the Board sought comments pursuant to the Paperwork Reduction Act (PRA), 44 U.S.C. 3501-3521, and OMB regulations at 5 CFR 1320.11, regarding: (1) Whether the collection of information, as modified, is necessary for the proper performance of the functions of the Board, including whether the collection has practical utility; (2) the accuracy of the Board's burden estimates; (3) ways to enhance the quality, utility, and clarity of the information collected; and (4) ways to minimize the burden of the collection of information on the respondents, including the use of automated collection techniques or other forms of information technology, when appropriate.

The Board estimated in the NPRM that the proposed requirements for minimum information to be included on or with Class I carriers' demurrage invoices would add a total one-time hourly burden of 280 hours (93.3 hours per year as amortized over three years or 40 hours per respondent  65 ) because, in most cases, those carriers would likely need to modify their information technology systems to implement some or all of the proposed changes. 66 NPRM, EP 759, slip op. at 13. In response to comments received from CSXT and CN that this estimate was understated, the Board increased the estimate in the SNPRM to 560 hours (186.6 hours per year as amortized over three years or 80 hours per respondent), which included the time Class I carriers would need to undertake the software redesign necessary to incorporate both the proposed minimum information discussed in the NPRM and the proposed additions discussed in the SNPRM. 67 SNPRM, EP 759, slip op. at 14.