The On-Demand Economy Continues to Grow, but Legal Consequences Abound for Employers and Employees in the U.S. and Abroad


by Nancy Cremins

Legal Analysis

The on-demand economy, which consists of independent and frequently short-term temporary employment arrangements, continues to expand in the United States and around the world. The growth of the on-demand workforce has outpaced overall U.S. workforce growth by a multiple of three since 2014, and freelancers are predicted to become the U.S. workforce majority within the next 10 years. Some surveys suggest that as many as 61% of employers plan to switch a significant portion of their full-time permanent positions to contingent jobs in the near future.

While trends indicate that companies intend to continue to move away from standard employer-employee arrangements, the legal landscape for on-demand workers is far from resolved. The classification of on-demand workers as “independent contractors” means there is a growing class of workers that do not have access to employment benefits such as vacation, sick time, and parental leave. Employers are not paying their share of employment taxes. In addition, these workers do not have access to social safety net programs such as workers’ compensation and are not making social security contributions. As more individuals rely on freelance work as their primary means of income, these workers are left without the protections traditional employment provides. As a result, the courts, legislatures, and companies both in the United States and abroad struggle with how to classify these workers and provide them with some access to benefits.  This article summarizes and contrasts some of the different approaches that have been taken since the author first addressed this topic in this journal, and addresses why they matter to businesses in the Commonwealth.

Companies that used independent contractors rather than employees to fuel their global growth, such as Uber, Lyft, and Postmates, continue to face legal challenges to their business practices regarding worker classification.  Several worker-misclassification claims have been settled at significant expense. In April 2017, the food-delivery business DoorDash agreed to pay $5 million to settle a 2015 independent contractor misclassification class action involving 33,744 class members. In March 2017, rideshare company Lyft’s $27 million settlement was approved by the court to resolve a misclassification suit brought by approximately 95,000 drivers.   Although in both cases the companies agreed to pay money to settle claims, neither company agreed to reclassify independent contractors as employees. Instead, each company agreed to clarify its internal policies and provide additional rights for its independent contractor workforce, including limitations on when Lyft may deactivate drivers and an opportunity to be heard in an arbitration paid for by Lyft to challenge the basis for deactivation.

Given that the settlements did not require these companies to re-classify their independent contractors as employees, they may find themselves facing further litigation on this subject. For example, in another case, delivery company Postmates received judicial approval for its $8.75 million settlement of misclassification claims in September 2017 in Singer et al. v. Postmates Inc., Case No. 4:15­cv­01284 (N.D. Cal.). However, Postmates is facing a new misclassification action filed in state court in November 2017, meaning that Postmates will need to settle or litigate these same claims anew.

Only one misclassification suit has made it to trial so far. A six-day bench trial in Lawson v. Grubhub, Inc., Case No. 15-cv-05128 JSC (N.D. Cal.) was concluded with closing arguments at the end of October 2017. On February 8, 2017, the court issued its decision finding that the plaintiff was properly classified as an independent contractor. The court applied California’s 11-factor “economic realities” test set forth in S.G. Borello & Sons Inc. v. Dept. of Industrial Relations, 48 Cal.3d 341 (1989), under which the alleged employer bears the burden of proving that the worker in question is not an employee.  The court found “While some [of the Borello] factors weigh in favor of an employment relationship, Grubhub’s lack of all necessary control over Mr. Lawson’s work, including how he performed deliveries and even whether or for how long, along with other factors persuade the Court that the contractor classification was appropriate for Mr. Lawson during his brief tenure with Grubhub.”

While the 11-factor Borello test[i] sets forth a different independent contractor test than is applied in Massachusetts,[ii] the results of this case will be felt across the U.S. because most prominent on-demand companies are based in California, and California law will likely apply to much of the worker litigation due to California choice of law and choice of venue provisions in agreements with independent contractors across the U.S.[iii]

Of on-demand businesses, ride-sharing giant Uber continues to reign supreme in both its ability to get press coverage and in the sheer volume of legal action it has faced globally for its classification of workers as independent contractors. On balance, the international results for Uber have been decidedly negative, but it is making some progress in the U.S., which reflects how Europe and the U.S. differ on their approaches to worker protection and Europe’s more skeptical view of the use of independent contractors instead of employees.

In April 2016, Uber attempted to resolve its largest worker misclassification class action in O’Connor v. Uber Technologies, Inc., C13-3826 EMC (N.D. Cal.), for $100 million covering 385,000 drivers. However, the proposed settlement was rejected by the court as not “fair, adequate, and reasonable” and because it did not make a determination of how to classify drivers.  This case was complicated by the 9th Circuit’s decision in Mohamed v. Uber Technologies, 836 F.3d 1102, 1008 (9th Cir. 2016), which determined that pursuant to agreements between Uber and the drivers, the drivers’ claims against Uber must be arbitrated.  Following the decision in Mohammed¸ Uber managed to persuade an arbitrator in an arbitration with a single California driver that such driver was properly classified under the Borello test as an independent contractor, and was not an employee.

Further arbitration and litigation was stayed in the Uber suits in California pending the outcome of National Labor Relations Board v. Murphy Oil USA, Inc., No 16-307, which on October 2, 2017 was argued before the Supreme Court and concerns the validity of class-action waivers that bar individuals from pursuing work-related claims on a collective or class basis.

In Europe, some courts and legislative bodies take a decidedly more protective approach for workers.  In 2016, a United Kingdom employment tribunal ruled that Uber drivers should be classified as workers rather than self-employed contractors, which meant that Uber drivers would be entitled to benefits including holiday pay and minimum wage. The decision was upheld on appeal by the U.K. Employment Appeal Tribunal.

In a separate action, the European Court of Justice (“ECJ”) also found that Uber should be regulated as a transportation company, which undercuts Uber’s position that it simply operates as an intermediary between drivers and passengers.

An additional blow to on-demand companies that rely on independent contractors in Europe was delivered by the ECJ in King v. The Sash Window Workshop Ltd., which ruled that misclassified self-employed contractors who are really workers or employees could claim back holiday pay all the way back to the year that the EU’s Working Time Directive[iv] was introduced. Before this ruling, liability was typically limited to one or two years’ back pay in most EU countries.

So, what is being done in the U.S. and Europe at the legislative level to introduce protections for this growing class on freelance workers? In the U.S., an assortment of legislative efforts seek to provide them access to benefits typically provided in the employment relationship. New York City enacted the “Freelance Isn’t Free Act” to impose penalties on companies that fail to pay their contractors, which the city may enforce. New York State established the Black Car Fund, which administers safety and health programs that benefit for-hire drivers, their passengers, and other New Yorkers on the road, and provides workers’ compensation insurance to black car and luxury limousine drivers. Bills proposed in Washington State, New Jersey, and New York would require companies that rely on independent contractors (such as Uber and Grubhub) to contribute to a portable benefits fund that would provide health insurance, time off, workers’ compensation, and other benefits. On the Federal level, in May 2017, Senator Mark Warner (D-VA) and Representative Suzan DelBene (D-WA) introduced legislation to test and evaluate innovative portable benefit designs for freelance workers to give independent contractors access to benefits that to date have only been available for employees.  But where the co-sponsors of this bill are Democrats in a Republican-controlled Congress, it is unlikely that this bill will gain any real traction before the 2018 election cycle.

It appears that additional initiatives in Europe would continue its seemingly more-protective stance. A July 2017 report commissioned by the U.K. Prime Minister on gig economy working practices sets forth a series of recommendations to improve working conditions for on-demand workers, including the proposal to create a new classification for workers on tech platforms, like Lyft and Postmates. In November 2017, the European Pillar of Social Rights, a set of policy priorities, was jointly signed by the European Parliament (the EU’s legislative body), the European Council (which sets EU policies), and the European Commission (the EU’s executive body) which sets forth 20 key principles, structured around three categories: (1) equal opportunities and access to the labor market; (2) fair working conditions; and (3) social protection and inclusion. These key principles will be implemented over time through legislation across the EU member states.

While Massachusetts’ strict “ABC test” to determine whether a worker is an employee or an independent contractor remains one of the toughest in the nation, other states that use similar tests are finding ways to determine that on-demand workers are independent contractors. However, there is presently no evidence Massachusetts courts are currently prepared to move in that direction.

Currently, the only new legislation in Massachusetts that impacts any on-demand companies involves the 2016 regulations on ridesharing companies that imposed a fee on ride-sharing services and established requirements for background checks, inspections, and insurance. Another new development impacting on-demand companies in Massachusetts occurred in  2017, when Uber introduced a pilot program that allows Massachusetts drivers to purchase workers’ compensation coverage that offers $1 million maximum coverage for medical costs and lost earnings due to a work-related accident.

The on-demand economy shows no sign of slowing down. As a result, innovative legislation or company-led initiatives that will protect these workers by providing new worker benefit programs are essential for the growing freelance workforce’s health and stability.  Under the Trump administration, the Department of Labor rolled back Obama era guidance of the “economic realities test” leaving more room for independent-contractor classification than the prior administration. However, the outcome of still pending litigations will likely force changes in policies and the practices of on-demand businesses that will result in additional worker protection. As freelance work continues to grow, Massachusetts may follow other states in finding new ways to provide benefits and protections for these workers. Regardless, when guiding your clients in Massachusetts, the safest bet is still providing full employment benefits to all workers they retain.

[i] The 11-factor Borello “economic realities” test sets forth the following factors used to determine whether an individual is properly classified as an independent contractor:

  1. Whether the person performing services is engaged in an occupation or business distinct from that of the principal;
  2. Whether or not the work is a part of the regular business of the principal or alleged employer;
  3. Whether the principal or the worker supplies the instrumentalities, tools, and the place for the person doing the work;
  4. The alleged employee’s investment in the equipment or materials required by his or her task or his or her employment of helpers;
  5. Whether the service rendered requires a special skill;
  6. The kind of occupation, with reference to whether, in the locality, the work is usually done under the direction of the principal or by a specialist without supervision;
  7. The alleged employee’s opportunity for profit or loss depending on his or her managerial skill;
  8. The length of time for which the services are to be performed;
  9. The degree of permanence of the working relationship;
  10. The method of payment, whether by time or by the job; and
  11. Whether or not the parties believe they are creating an employer-employee relationship may have some bearing on the question, but is not determinative since this is a question of law based on objective tests.

[ii] Massachusetts applies a strict “ABC test” for properly classifying an independent contractor. Specifically, “[a]n employer who wants to treat someone as an independent contractor rather than an employee has to show that the work:

  1. is done without the direction and control of the employer; and
  2. is performed outside the usual course of the employer’s business; and
  3. is done by someone who has their own, independent business or trade doing that kind of work.” Massachusetts Attorney General’s Fair Labor Division on M.G.L. c. 149, s.148B

[iii] The long-standing Borello test is also undergoing scrutiny in California. Currently, in Dynamex Operations West, Inc. v. Superior Court of Los Angeles County, the California high court is considering whether to adopt a revised test to determine whether an individual is an employee or independent contractor. The California Supreme Court requested briefs addressing whether California law should use an “ABC” test similar to the one used in Massachusetts.  Plaintiff’s counsel in the Grubhub case filed a Notice of Supplemental Authority with the District Court regarding the Dynamex matter, however, the court did not delay its decision in Gruhub instead opting to apply the current Borello test in its decision.

[iv] The EU’s Working Time Directive (2003/88/EC) requires EU countries to guarantee the certain minimum rights for all workers and regulates the amount of time people can spend at work in order to protect the health and safety of the European workforce.


Nancy Cremins is the Chief Administrative Officer & General Counsel of Globalization Partners, an International Professional Employer Organization helping companies expand in 150+ countries without the pain of setting up an entity.

This article was updated on March 8, 2018 to reflect the latest development in the case ofaweson v. Grubhub.

Sanctuary Cities: Distinguishing Rhetoric from Reality



by Inez Friedman-Boyce, Jennifer Luz, Sarah J. Fischer, Alexandra Lu, and Louis L. Lobel

Legal Analysis

Less than a week after his inauguration, President Trump signed Executive Order No. 13,768 (“Order”), threatening to “crackdown on sanctuary cities that refuse to comply with federal law and that harbor criminal aliens” by cutting off federal grant money.[1] This article examines the current political and legal landscape affecting sanctuary cities and the policies that define the “sanctuary city” designation.[2]

What Is A “Sanctuary City”?

There is no single, legal definition of a “sanctuary city”; rather, the designation refers generally to cities and counties that have policies—whether formally or informally adopted—that are intended to further public safety by mitigating against any deterrent effects that immigration status might have on residents’ cooperation with local law enforcement officials and by distinguishing between local police and federal immigration officials. Studies that inform sanctuary policies indicate that victims of and witnesses to crimes are less likely to come forward to report and assist with the investigation and prosecution of crimes if they fear deportation as a possible result.[3] Despite some variation, sanctuary cities share the common policy objective: to build community trust in order to “promote public safety and confidence in local law enforcement.”

What Are Sanctuary Policies?

Sanctuary policies differ across jurisdictions to accommodate local needs and priorities. Some have written policies that expressly prohibit police from inquiring about immigration status or direct local law enforcement not to comply with civil detainer requests by the U.S. Immigration and Customs Enforcement (“ICE”) to hold noncitizens for up to 48 hours to provide ICE agents extra time to take them into federal custody for deportation purposes. Others identify as sanctuary cities but have no written policies. Florida’s Miami-Dade County’s policy, until recently, was to refuse detainer requests except where the suspect had been charged with a non-bondable offense or had previously been convicted of a violent felony. Meanwhile, California’s Santa Clara County refuses to honor all detainer requests.

Several Massachusetts communities have sanctuary policies that limit local police cooperation with ICE, including Arlington, Boston, Cambridge, Chelsea, Holyoke, Lawrence, Newton, Northampton, and Somerville. Chelsea declared itself a sanctuary city in June 2007, adopting a policy that “immigration status (or lack thereof) … is not and shall not be a matter of local police concern or subsequent enforcement action by the [Chelsea Police Department] unless there exists through reliable and credible information a potential threat to public safety and/or national security.”[4] The policy only governs civil immigration matters and does not prohibit Chelsea Police from assisting with criminal matters. Lawrence adopted its Trust Ordinance in August 2015 “to increase public confidence in Lawrence Law Enforcement by providing guidelines associated with federal immigration enforcement, arrests, and detentions.”[5] Pursuant to the Ordinance, Lawrence police will not detain an individual based solely on an immigration hold or administrative warrant—or absent a warrant signed by a judge and based on probable cause—but will allow ICE officers with criminal warrants to use their facilities.

Since Trump’s election, more Massachusetts communities have galvanized to consider “sanctuary city” status. The Massachusetts Legislature also is considering a state-wide sanctuary policy, the Safe Communities Act, which would prohibit, inter alia, the use of state and local law enforcement resources or the Massachusetts Registry of Motor Vehicles record-keeping system for immigration enforcement purposes, and the arrest or detention of individuals solely on the basis of civil detainer requests or administrative warrants.[6] Police would not be prevented from pursuing immigrants who commit crimes subject to applicable federal laws and constitutional standards. Because sanctuary policies have broad support across the Commonwealth, two exceptions have attracted disproportionate press attention: in January 2017,the Republican sheriffs of Bristol and Plymouth County each signed agreements with ICE to deputize their correctional officers to detain individuals for immigration violations under Section 287(g) of the Immigration and Nationality Act.

The Order

On January 25, 2017, President Trump signed the Order entitled “Enhancing Public Safety in the Interior of the United States.” By its plain language, the Order threatens “all Federal grant money” received by “sanctuary jurisdictions.” The Order includes several internally inconsistent and ambiguous definitions of sanctuary jurisdictions. Section 1 defines “sanctuary jurisdictions” as those that “willfully violate Federal law in an attempt to shield aliens from removal.” Section 9(a) defines the term more broadly as jurisdictions that “willfully refuse to comply with 8 U.S.C. § 1373” (“§ 1373”), which states that “a Federal, State, or local government entity or official may not prohibit or in any way restrict, any government entity or official from sending to, or receiving from, the [INS] information regarding the citizenships or immigration status, lawful or unlawful, of any individual,” “or which has in effect a statute, policy, or practice that prevents or hinders the enforcement of Federal law.” Section 9(b) orders a retroactive identification of sanctuary jurisdictions based on a list to be publicized weekly including “any jurisdiction that ignored or otherwise failed to honor any detainers.” The Order also gives the Secretary of the Department of Homeland Security (“DHS”) unfettered “authority to designate, in his discretion and to the extent consistent with law, a jurisdiction as a sanctuary jurisdiction.”

The impact of the Order was felt immediately nationwide, with reports of decreased utilization of police, health, and social services by immigrant communities. And cities that had enacted sanctuary policies in effort to address the very fear and community distrust the Order has revived are now faced with deciding between prioritizing community safety or abandoning their sanctuary policies to avoid potentially losing critical federal funding. In letters dated April 21, 2017 sent to nine jurisdictions, the Department of Justice (“DOJ”) demanded proof of compliance with § 1373, coupled with the threat of terminating certain grants. Confronted with the Order, some jurisdictions, including Miami-Dade County, Florida and Dayton, Ohio  rescinded their sanctuary policies,[7] and other cities like Quincy, Massachusetts, have declined to adopt a proposed policy. Yet other communities chose to fight back, declaring that challenging the Order is “just as much about protecting residents as it is about protecting federal resources.”

Legal Challenges to the Order

On January 31, 2017, San Francisco filed the first lawsuit challenging the constitutionality of Section 9(a) of the Order, which states: “jurisdictions that willfully refuse to comply with 8 U.S.C. § 1373 (sanctuary jurisdictions) are not eligible to receive Federal grants, except as deemed necessary for law enforcement purposes by the Attorney General or the Secretary [of Homeland Security].” Other suits quickly followed by Santa Clara County and Richmond in California; Chelsea and Lawrence, Massachusetts; and Seattle, Washington.[8]  Santa Clara, consistent with its long-standing position that it does not comply with § 1373, asserted only constitutional arguments, but the remaining jurisdictions sought declarations they complied with § 1373 and therefore were not “sanctuary jurisdictions” subject to the Section 9(a) sanctions.

These cases assert the following constitutional challenges to the Order:

·       violation of the separation of powers doctrine (legislates a penalty and imposes new conditions on federal grants that only Congress can authorize and impermissibly refuses to spend funds already appropriated by Congress);

·       void for vagueness under the Fifth Amendment (fails to specify the prohibited conduct that would subject the local jurisdiction to defunding, includes no guidance on what constitutes a “sanctuary jurisdiction” subject to penalties, and has “expansive standardless language” open to arbitrary and discriminatory enforcement);

·       violation of procedural due process under the Fifth Amendment (jeopardizes local jurisdictions’ entitlement to money appropriated by Congress without administrative or judicial procedure);

·       violation of the spending clause of the Tenth Amendment (imposes, without notice, vague conditions after funds have already been accepted, with no nexus between the federal funds threatened and the Order’s purpose, and uses coercive financial inducements); and

·       violation of the principles of federalism and state sovereignty under the Tenth Amendment (compels local jurisdictions to administer or enforce federal immigration policies and programs through coercion, and may subject cities to Fourth Amendment liability; imposes a blanket restriction on local policymaking discretion regarding how to treat immigration status of residents and a specific restriction on the regulation of law enforcement priorities and policies to address the best interest of residents).[9]

On April 25, 2017, Judge William H. Orrick III of the Northern District of California ordered a nationwide preliminary injunction against enforcement of the Order’s defunding provision in the Santa Clara and San Francisco cases. Judge Orrick rejected the DOJ’s arguments that: (1) the claims were not “prudentially ripe” because the harms are too contingent, and the DOJ and DHS have not determined the terms of the Order, (2) there was no loss of funds or cognizable harm because neither Santa Clara nor San Francisco had been named “sanctuary jurisdictions” pursuant to the Order, (3) the Order did not change existing law, as it would be enforced only “to the extent consistent with the law,” (4) it was restricted to three DOJ and DHS “grants that are already conditioned on compliance with § 1373,” and (5) it was therefore “merely an exercise of the President’s ‘bully pulpit’” that “highlight[ed] a changed approach to immigration enforcement.” Judge Orrick wrote: (1) “[t]here is no doubt that Section 9(a), as written, changes the law” and “purport[s] to give the Secretary or Attorney General the unilateral authority to alter [§ 1373],” a power reserved to Congress, and (2) standing is established “by demonstrating a well-founded fear of enforcement and a threatened injury that is ‘sufficiently real and imminent,’” and Santa Clara and San Francisco, are likely to be designated “sanctuary jurisdictions” under the Order given their policies, and withdrawing review would result in hardship that is more than financial loss. Further, Judge Orrick found a high likelihood of success on the merits of the constitutional claims, that there was impending irreparable harm based on budgetary uncertainty and constitutional injury, and that the balance of equities and public interest squarely tips in favor of the injunction. Finally, Judge Orrick found “a nationwide injunction is appropriate” because the constitutional violations had nationwide consequences.

The Chelsea and Lawrence Lawsuits

On February 8, 2017, Chelsea and Lawrence filed their complaint, challenging the Order on the previously discussed constitutional and declaratory relief grounds. Their motivation in filing suit underscores what is at stake for many sanctuary cities nationwide.[10] Simply put: “[i]t is impossible [for a sanctuary city] to create a budget when it is unclear what effect the Executive Order will have on its funding.” The crippling consequence is especially stark in communities like Chelsea and Lawrence. Chelsea is a working-class city where over 60% of its residents identify as Hispanic or Latino, over 40% are foreign-born, and over 20% live below the poverty level with a per capita income of $21,722.00. Chelsea counts on the federal government for about 10%, or $14 million, of its $170 million annual budget. Similarly, Lawrence is a working class city where over 70% of its residents identify as Hispanic or Latino, over 35% are foreign-born, and over 25% live below the poverty level with a per capita income of $17,167.00. Lawrence counts on the federal government for over 15%, or $38 million, of its $245 million annual budget. The Order threatened large portions of these impoverished cities’ budgets because of policies they deemed necessary for their communities’ public safety. In early May, while the DOJ’s motion to dismiss was pending, Judge Orrick’s national injunction issued; the DOJ and Chelsea and Lawrence subsequently agreed to a stay, pending resolution of the injunction.

Where We Are Now

On May 22, 2017, Attorney General Sessions issued a “Memorandum on the Implementation of the Executive Order” (“Memo”), codifying arguments advanced by the DOJ at the preliminary injunction hearing. Relying on the Memo, in late May, the DOJ moved for reconsideration of the nationwide injunction in the San Francisco and Santa Clara cases. The DOJ then filed motions to dismiss on procedural and substantive grounds in the San Francisco, Santa Clara, Richmond, and Seattle cases. On July 20, 2017, Judge Orrick issued an order denying the motions for reconsideration and motions to dismiss in the Santa Clara and San Francisco cases, finding that the Memo did not impact his prior conclusion regarding standing, ripeness, and likelihood of success on the merits. Additionally, he concluded that San Francisco had stated a claim for declaratory relief.


While a nationwide preliminary injunction has been entered, many questions remain. The interplay between federal and state law regarding ICE detainers remains unclear; the constitutionality of § 1373 is still undecided; and future federal actions against sanctuary cities remain real possibilities. The Memo, purporting to narrow the definition of “sanctuary jurisdictions” and limit the sources of federal funding that are threatened by the Order, is arguably inconsistent with the terms of the Order itself, does not have the force of law, and is subject to change. It remains to be seen to what extent local policy makers are able to prioritize public safety over federal immigration enforcement without jeopardizing critical federal funding.

[1] Exec. Order No. 13,768, 82 Fed. Reg. 8799 (Jan. 25, 2017). Joseph Spector Cities Back Lawsuit to Block Immigration Order, USA Today, Feb. 17, 2017.

[2] See Benjamin Gonzalez, Loren Collingwood, Stephen Omar El-Khatib, The Politics of Refuge: Sanctuary Cities, Crime, and Undocumented Immigration, Urban Affairs Review, May 7, 2017.

[3]For example, one study found that 70% of undocumented immigrants and 44% of Latinos are less likely to contact law enforcement if they are victims of a crime for fear that the police will ask about immigration status, and 67% of undocumented immigrants and 45% of Latinos are less likely to report crimes because of the same fear. See Nik Theodore, Insecure Communities: Latino Perceptions of Police Involvement in Immigration Enforcement (Dep’t of Urban Planning and Policy, Uni. of Ill. at Chicago, Chicago, IL), May 201, at 5-6.

[4] Chelsea, MA, Resolution of the City of Chelsea, Massachusetts (Jun. 4, 2007).

[5] Lawrence, MA, Lawrence Trust Ordinance, Chapter 9.20 (June 8, 2015).

[6] S.1305 and H.3269, 190th Gen. Court of the Commonwealth of Mass. (Mass. 2017).

[7] As Mayor Carlos Giménez of Miami-Dade explained, “It’s really not worth the risk of losing millions of dollars … in discretionary money from the feds.” Ray Sanchez, Florida’s Largest County to Comply with Trump’s Sanctuary Crackdown, CNNpolitics, Updated Jan. 27, 2017.

[8] City and County of San Francisco v. Trump et al., 4:17-cv-00485-DMR (N.D. Cal. Jan.31, 2017), County of Santa Clara v. Trump et al., 5:17-cv-00574-LHK (N.D. Cal. Feb. 3, 2017); City of Chelsea, et al. v. Trump, et al., 1:17-cv-10214-GAO (D. Mass. Feb. 8, 2017); City of Richmond v. Trump et al., 3:17-cv-01535-SK (N.D.Cal. Mar. 21, 2017); City of Seattle v Trump et al., 2:17-cv-00497-BAT (W.D. Wash. Mar. 29, 2017), The San Francisco, Santa Clara and Richmond, California cases have been related before Judge William H. Orrick, III.

[9] See, e.g., Creedle v. Gimenez, et al., 1:17-cv-22477 (S.D. Fl. filed on Jul. 5, 2017); Commonwealth v. Lunn, SJC-12276 (decided July 24, 2017) (holding federal civil detainers unconstitutional under Massachusetts Constitution).

[10] Elizabeth Ross, How Can “Sanctuary Cities” Resist Trump? This Lawsuit Could Provide a Blueprint, Public Radio International, Updated Apr. 12, 2017.

Inez Friedman-Boyce is a partner, Jennifer Luz is counsel, and Sarah Fischer, Alexandra Lu, and Louis Lobel are associates at Goodwin Procter LLP. Ms. Friedman-Boyce is a past co-chair of the BBA Class Actions Committee and the current co-chair of the Lawyers’ Committee for Civil Rights and Economic Justice. Along with the Lawyers’ Committee, they are all counsel for the Cities of Chelsea and Lawrence in litigation pending in the U.S. District Court for the District of Massachusetts challenging President Trump’s sanctuary city executive order.

Executive Order: Strike of a Pen, Law of the Land?

starkeystarkeyby M. Patrick Moore and Kate R. Cook

Legal Analysis


The President of the United States and the Governor of Massachusetts have the implied power to issue executive orders that, in certain contexts, will have the force of law. Focusing on the federal system and the Massachusetts state system, this article will address the concept of the executive order, how it has changed over time, and why executive orders are used to further wide-ranging policy goals. The article will also address the judicial scrutiny of executive orders, including, in particular, whether they are owed any deference or presumption of lawfulness.


Throughout history, executive orders have addressed issues of profound national and local importance. Our system of classifying national security information, for example, is set by executive order. See Exec. Order No. 13526 (Dec. 29, 2009). So, too, is the process by which national security agencies determine who may have access to such information. See Exec. Order No. 13764 (Jan. 17, 2017); see also 50 U.S.C. § 3161 (instructing the President to issue such an order). Executive orders define the process by which nearly every federal and state agency may regulate. See generally Exec. Order No. 12291 (Feb. 17, 1981) (centralizing federal regulatory planning and review); Mass. Exec. Order No. 562 (Mar. 31, 2015) (synthesizing state regulatory review). And they provide the definition of an unfunded mandate, at least at the local level. Mass. Exec. Order No. 145 (Oct. 21, 1978). Executive orders have both advanced principles of equal protection and hindered them. See Exec. Order No. 9981 (July 26, 1948) (desegregating the armed forces); Exec. Order No. 10450 (Apr. 27, 1953) (excluding gay and lesbian officials from government service during all administrations from Eisenhower to Clinton); Exec. Order No. 12968 (Aug. 2, 1995) (establishing that the federal government no longer will “discriminate on the basis . . . [of] sexual orientation in granting access to classified information”).

Despite the importance of executive orders, neither the United States nor the Massachusetts Constitutions expressly grants the power to issue them. Rather, the power is implied from the executive’s core authority to administer the laws. See U.S. Const. art. II (“The executive Power shall be vested in a President . . . .”); id., art. II, § 3 (“[H]e shall take Care that the Laws be faithfully executed . . . .”); Mass. Const. pt. II, ch. 2, § 1, art. I (“There shall be a supreme executive magistrate, who shall be styled, the Governor of the Commonwealth of Massachusetts . . . .”). The scope of that power has been, and continues to be, the focus of significant debate. In this article, we focus on the legal framework of that ongoing debate.

The Supreme Court tells us that history is an important guide to the scope of executive power. See, e.g., NLRB v. Noel Canning, 134 S. Ct. 2550, 2560 (2014). On that score, the record is clear. Dating back to President Washington, each President has interpreted his executive power to encompass the authority to issue executive orders. See Phillip J. Cooper, By Order of the President: Administration by Executive Order and Proclamation, 18 Admin. & Soc’y 233, 236 (1986) (describing Washington’s use of executive orders and proclamations). Massachusetts Governors likely have exercised the same power for just as long, though Governor Saltonstall (who served from 1939-1945) was the first to track the orders formally. See Mass. Legis. Research Council, Report Relative to Gubernatorial Executive Orders, submitted as 1981 House 6557, at 22 (1981) (“Mass. H. Rep.”).

Historically, the number of presidential and gubernatorial executive orders peaked during periods of national emergency and war, most notably the Civil War, World War I, the Great Depression and World War II. See Cooper, supra, at 236-37. Governor Saltonstall, for example, began issuing a disproportionately high number of orders at the outset of World War II. See Mass. H. Rep. at 22, 80. No executive, however, used the authority with such regularity as President Franklin Roosevelt, who issued 567 orders in 1933 alone and a total of 3,727 orders before his death in 1945. See Cooper, supra, at 237. President Roosevelt averaged nearly 285 executive orders per year—a trend that increased steadily through the conclusion of World War II. See John Contrubis, Cong. Research Serv., Rep. No. 95-772, Executive Orders and Proclamations, at CRS-25 tbl.1 (1999). Since his administration, no President has averaged more than 78 per year. Id; Tara L. Branum, President or King? The Use and Abuse of Executive Orders in Modern Day America, 28 J. Legis. 1, 27-29 (2002).

Perhaps as a result of the constitutional silence on the issue, there is no codified definition of an executive order. The most commonly referenced definition is set forth in a 1957 report from a committee of the U.S. House of Representatives. It reads, in pertinent part: “Executive orders . . . are directives or actions by the President. When they are founded on the authority of the President derived from the Constitution or statute, they may have the force and effect of law. . . . Executive orders are generally directed to, and govern actions by, Government officials and agencies. They usually affect private individuals only indirectly.” H. Comm. on Gov’t Operations, 85th Cong., Executive Orders and Proclamations: A Study of a Use of Presidential Powers 1 (1957); see also Vivian S. Chu & Todd Garvey, Cong. Research Serv., RS20846, Executive Orders: Issuance, Modification, and Revocation (2014); see generally Contrubis, supra. Massachusetts law similarly lacks a statutory definition of executive order, though the Supreme Judicial Court has recognized such orders as the “formal” exercise (or delegation) of powers granted to the Governor by the constitution or the legislature. See Opinion of the Justices, 368 Mass. 866, 874-75 (1975).

Regarding the substance and reviewability of executive orders, policymakers and courts alike are guided by the three categories set forth in Justice Robert Jackson’s seminal concurrence in Youngstown Sheet and Tube Co. v. Sawyer, 343 U.S. 579 (1952). That case famously involved President Truman’s attempt to seize most of the nation’s steel mills to quell labor unrest during the Korean War, which Truman attempted to accomplish via executive order directing the Secretary of Commerce to take possession of and operate the mills. Id. at 583. In analyzing Truman’s order, Justice Jackson set forth three “practical situations in which a President may doubt, or others may challenge, his powers.” Id. at 635.

Though Justice Jackson described the “three practical situations” as “over-simplified,” they continue to be the benchmarks against which the exercise of executive power is measured. Id. at 635-38. First, when acting with express or implied legislative authorization, the executive exercises both the power vested in the executive and the authority of the legislature, and the executive’s actions are presumed valid. Second, when the executive acts in the “zone of twilight” defined by legislative silence or inaction on the issue, executive authority will be addressed on a case-by-case basis. Third, when the executive contravenes legislative action, the executive’s action is presumptively impermissible. Id. at 637-38.

The most common executive orders fall into the first category, i.e., action where the executive is expressly authorized to act. Such orders are generally directed to, and govern actions by, government officials and agencies. The reason is simple. Executive orders are not self-executing; most exist against the backdrop that an executive official who fails to comply with the order will be removed. See generally Myers v. United States, 272 U.S. 52 (1926). The majority of executive orders pertain to internal matters of government administration, the creation of task forces, and the commissioning of reports. Typically, the order will set forth clearly the basis of its authority. Cf. 1 C.F.R. § 19.1(b) (2017) (“The order or proclamation shall contain a citation of the authority under which it is issued.”).

Though the legal framework is routine, the subject matter of such orders is often rich and significant. On the state level, for example, the Governor is expressly afforded the constitutional authority to nominate and appoint judges. See Mass. Const. pt. II, ch. 2, § 1, art. 9. In 1975, Governor Dukakis established the Judicial Nominating Commission (JNC) by executive order, on the theory that the “high quality of judicial officer appointments can be best assured by the use of a non-partisan judicial nominating commission composed of outstanding laymen and lawyers.” Mass. Exec. Order No. 114 (Jan. 3, 1975). The Supreme Judicial Court upheld the Governor’s authority to create the JNC by executive order, which established “formally and publicly” the “enlist[ment] [of] such aid as he deems necessary to investigate the availability of qualified candidates for judicial office.” Opinion of the Justices, 368 Mass. 866, 874-75 (1975). Though each successive Governor has slightly revised the JNC, the tradition of a formalized nominating process continues.

Even where the executive has acted pursuant to an express grant of authority, however, there are checks. For example, when President Trump established his most recent travel ban, see Exec. Order No. 13780 (March 6, 2017), he purported to act under his express statutory authority to bar the immigration of certain individuals he “finds . . . would be detrimental to the interests of the United States.” See 8 U.S.C. § 1182(f). Subsequent litigation has focused on whether President Trump actually made such a finding and, if he did so, whether it was improperly infected with religious considerations (in violation of the First Amendment) or was otherwise arbitrary (in violation of the Fifth or Fourteenth Amendments). See, e.g., Int’l Refugee Assistance Project v. Trump, 857 F.3d 554 (4th Cir. 2017); Hawaii v. Trump, 859 F.3d 741 (9th Cir. 2017). The Supreme Court will hear these cases during the first session of October Term 2017. See Trump v. Int’l Refugee Assistance Project, 137 S. Ct. 2080 (2017) (granting the Government’s petitions for certiorari, staying in part the lower courts’ injunctions, and holding that the Executive Order cannot be enforced against foreign nationals who have a credible claim of a bona fide relationship with a person or entity in the United States).

Executive orders in Justice Jackson’s second category, where the scope of executive authority is not well-defined, are evaluated on a case-by-case basis. A reviewing court will scour the Constitution and consider “all the circumstances which might shed light on the views of the Legislative Branch towards such action, including congressional inertia, indifference, or quiescence.” Dames & Moore v. Regan, 453 U.S. 654, 668-69 (1981) (internal quotation omitted). In Dames & Moore, the Supreme Court analyzed a series of executive orders that extinguished American liens against Iranian property and barred claims against Iranian property in American courts. There, the Court recognized that a “‘long-continued [executive] practice, known to and acquiesced in by Congress, would raise a presumption that the [action] had been [taken] in pursuance of [legislative] consent.’” Id. at 686 (second and third alterations in original) (quoting United States v. Midwest Oil Co., 236 U.S. 459, 474 (1915)). In other words, the absence of legislative disapproval following executive action on uncertain terrain will be a relevant consideration. Id. at 686-88.

Sometimes, particularly regarding long-standing executive orders, legislative approval is express. President Reagan’s Executive Order No. 12333, for example, established a framework for the collection of foreign intelligence. The order itself relied on executive authority granted “by the Constitution” and by the 1947 National Security Act which, most prominently, created the National Security Council but which contained no express delegation of authority to the President. The order remains in place and, reportedly, is the legal basis for much of the National Security Agency’s data collection. See Erica Newland, Executive Orders in Court, 124 Yale L.J. 2026, 2030 (2015). The order is now expressly referenced many times in the United States Code. Indeed, Congress even requires the regular reporting of any violations of the order. See 50 U.S.C. § 3110.

Executive orders in the third category, where the executive has acted in contravention of legislation, generally are not permitted. For example, the Governor may not seize facilities of the Massachusetts Bay Transportation Authority where such action is barred by the General Laws. See Mass. Bay Transp. Auth. Advisory Bd. v. Mass. Bay Transp. Auth., 382 Mass. 569, 578-79 (1981). Nor may the President bar a federal contractor that has hired permanent replacements for striking workers where the contractor is expressly afforded by federal law the right to hire such replacements. See Chamber of Commerce v. Reich, 74 F.3d 1322, 1338-39 (D.C. Cir. 1996).

Where executive orders are well-anchored in existing law, they can be attractive tools for policymakers, in part because of the minimal process associated with them. As a senior aide to President Clinton once quipped: “Stroke of the pen, law of the land. Kind of cool.” Tara L. Branum, President or King? The Use and Abuse of Executive Orders in Modern-Day America, 28 J. Legislation 1, 1 (2002). Accordingly, an executive order may be an expeditious and tangible step towards accomplishing a policy goal. See, e.g., Elena Kagan, Presidential Administration, 114 Harv. L. Rev. 2245, 2300 (2001) (describing President Clinton’s use of “deliverables,” like announcing the issuance of an executive order, to advance his political agenda). Little is required of the executive other than publishing the order in the respective federal or state register. See 44 U.S.C. § 1505; G.L. c. 30A, § 6.

But the same qualities that make orders attractive can make them perilous, particularly where the subject matter strays beyond routine administrative or ceremonial purposes. Other types of executive action, most prominently including administrative rulemaking, include formalized stakeholder input before the action is finalized. The notice-and-comment regulatory process, for example, requires the executive branch to hear—and, where appropriate, address—stakeholder comments and concerns before a regulation is finalized. See 5 U.S.C. § 553; G.L. c. 30A, §§ 2-3. An executive order may be issued without such external input. Although robust internal legal and litigation risk analyses should be undertaken before an order is signed, that is a matter of practice rather than law. Compare 1 C.F.R. § 19.2 (setting forth typical process of Attorney General review), with Ryan Lizza, Why Sally Yates Stood Up to Trump, New Yorker (May 29, 2017) (noting that the Acting Attorney General first learned of Exec. Order No. 13768, President Trump’s first travel ban, from media reports).

Where an executive order that substantively affects the rights or property interests of stakeholders is issued, litigation is likely to follow, particularly where its legal foundation is uncertain or untested. Recent experience is illustrative. See, e.g., Int’l Refugee Assistance Project, 137 S. Ct. at 2088 (granting certiorari review of Exec. Order No. 13780, commonly referred to as the “travel ban”); County of Santa Clara v. Trump, 2017 WL 1459081 (N.D. Cal. Apr. 25, 2017) (enjoining enforcement of Exec. Order No. 13768 regarding grant funding to so-called sanctuary cities on several constitutional grounds); cf. United States v. Texas, 136 S. Ct. 2271 (2016) (per curium) (equally divided court affirming Fifth Circuit decision invalidating Department of Homeland Security memoranda regarding deferred action on certain undocumented immigrants). As Justice Jackson teaches us, that is as it should be: “Presidential claim to a power at once so conclusive and preclusive must be scrutinized with caution.” Youngstown, 343 U.S. at 638 (Jackson, J., concurring).


Executive orders can function as a formalized statement to the public regarding how executives intend to solve an administrative problem or discharge their duties. To withstand judicial review, executive orders must be rooted in constitutional or statutory authority and comply with the relevant constitution. Moreover, because executive orders lack the procedural protections that accompany administrative rulemaking and the deliberative process that shapes legislation, executive orders outside the obvious bounds of executive power deserve particular scrutiny.

Kate R. Cook is co-chair of the Boston Bar Association’s Civil Rights and Civil Liberties Section and Of Counsel at Sugarman Rogers.  She previously served as Chief Legal Counsel to Governor Deval Patrick, where she assisted in drafting numerous Executive Orders.

M. Patrick Moore Jr. is co-chair of the Boston Bar Association’s Government Lawyers Forum and Counsel at Hemenway and Barnes LLP. He previously served as Associated Counsel and Advisor for Presidential Personnel in the White House of President Barack Obama, and as Deputy Legal Counsel to Governors Deval Patrick and Charlie Baker. In those roles, he assisted in drafting, analysis and review of executive orders. 

The Administrative Law of Deregulation: The Long Road for the Trump Administration to Undo Obama-Era Regulations

lyonsby Daniel Lyons

Legal Analysis

Dysfunction among Washington’s elected branches has thrust the administrative state—the alphabet soup of federal agencies that together do the day-to-day work of governing the republic—into the nation’s political spotlight. Frustrated by his inability to get the Republican-led Congress to pass his legislation, President Obama governed by “pen and phone,” leaning on the administrative state to accomplish much of his second-term agenda. From immigration reform to transgender rights to climate change, agencies became the primary engines driving the White House’s policy agenda.

Unsurprisingly, President Trump has focused on undoing many of these agency initiatives. As a candidate, Trump campaigned on a platform of deregulation, arguing that agency regulations inhibit economic growth. And as president, he moved quickly to reverse certain Obama-era administrative initiatives and established agency-level task forces to identify others for repeal. White House advisor Stephen Bannon described these efforts as a fight for the “deconstruction of the administrative state.”

But how, precisely, can an agency conduct an about-face and reverse a policy decision that it only recently adopted? The answer is surprisingly complex, turning in part upon the way the agency formulated its policy and how long it has been in effect. The answer is also extremely important. Each year, agencies enact more rules than Congress and hear far more cases than the federal court system. Administrative process is the primary legal check on what is often called the “headless fourth branch of government.” The Trump administration’s deregulation campaign will withstand court challenges only to the extent that it gets agency process right.

Policy Statements

The easiest agency decisions to reverse are so-called interpretive rules and policy statements. These are agency statements that, in theory, create no new rights or responsibilities for regulated entities, but rather simply clarify what the agency currently believes that existing law already requires. Because these so-called “nonlegislative rules” do not change the law, they are exempt from the Administrative Procedure Act’s notice and comment process. As a result, agencies are generally free to issue or revoke these statements at will, subject only to the agency’s own procedures for doing so.

Perhaps the most high-profile interpretive rule in President Trump’s crosshairs was the Education Department’s guidance regarding transgender students. In January 2015, the Department of Education’s Office of Civil Rights issued an opinion letter explaining that under the agency’s regulations implementing Title IX, schools are permitted to separate bathroom facilities by sex, but “must treat transgender students consistent with their gender identity.” The agency followed this with a second letter, in May 2016, explaining that a student’s sex under the statute equates to his or her gender identity. The opinion letters had legal consequences. The Fourth Circuit deferred to the first letter in G.G. v. Gloucester County School Board, a 2016 decision reversing the dismissal of a transgender student’s suit for bathroom access.[1] Four months later, a federal district court in Texas enjoined the letter on the ground that it changed the law and therefore was not merely an interpretive rule.[2] In October 2016, the Supreme Court granted certiorari in Gloucester County.[3]

On February 22, 2017, the agency issued a new opinion letter that withdrew and rescinded the earlier two letters. It explained that the earlier letters “do not contain extensive legal analysis or explain how the position is consistent with the express language of Title IX, nor did they undergo any formal public process.” The Supreme Court then vacated and remanded Gloucester County back to the Fourth Circuit to reconsider its decision in light of the new letter. In the meantime, several states including Massachusetts have stepped up to fill the void in transgender rights left by the agency’s most recent action.[4]

Recent Regulations: The Congressional Review Act

Regulations enacted during the final months of the Obama administration were susceptible to rescission via the Congressional Review Act, a 1995 law enacted as part of Newt Gingrich’s Contract with America to encourage greater legislative oversight of agency rulemaking.[5] The Congressional Review Act requires agencies to report to Congress whenever enacting a rule with a $100 million impact on the economy. Congress then has 60 legislative days to pass a joint resolution disapproving the rule. The Act provides expedited debate procedures, including a prohibition on filibusters. If both houses pass the resolution and either the president signs it or Congress overrides a presidential veto, the rule is voided and the agency is prohibited from issuing any rule “substantially the same” as the rejected rule.

In theory, the Congressional Review Act gives Congress greater oversight of agency action, restoring some of the power lost in INS v. Chadha[6] (which invalidated the legislative veto, a process that permitted the House or Senate to unilaterally invalidate certain agency action). But in practice, a successful joint resolution is exceedingly difficult to achieve, because absent a veto-proof majority, it requires the approval of the president, who is unlikely to agree to repeal one of his agencies’ major rules. In fact, prior to President Trump’s inauguration, the Congressional Review Act had been successfully deployed only once, to invalidate a 2001 Department of Labor ergonomics rule passed in the twilight moments of the Clinton administration and repealed shortly after President George W. Bush’s inauguration.

But the current Congress successfully passed, and President Trump signed, a record fourteen resolutions of disapproval.[7] One of the most far reaching of these was the voiding of the Department of the Interior’s Stream Protection Rule, which took effect on the final day of Obama’s term. The rule would have prohibited mining practices that adversely affected streams and water supplies and would have required mining companies to restore streams and mined areas such that they could support all uses that they could have supported prior to mining activities. Congress passed a resolution of disapproval in January 2017 and President Trump signed it on February 16. As a result, the agency reports, the rule was nullified and “the regulations that are now in effect are the same as those that were in effect on January 18, 2017,” the day before the new rule would have taken effect. The agency is “in the process of amending all the regulations altered by the Stream Protection Rule back to the form in which those regulations existed on January 18, 2017. When complete, these amendments will be published in the Federal Register.”

Older Regulations: Notice and Comment Rulemaking

Older regulations may be rescinded primarily via the notice and comment process outlined in the Administrative Procedure Act—the same process the agency went through to enact the regulation.[8] First the agency must promulgate a notice of proposed rulemaking, which usually includes an explanation why the agency seeks to take action and the text of the rule the agency proposes to adopt (but in the deregulatory posture, it could instead identify the text the agency proposes to delete). The agency then must invite public comment, which occurs in two rounds: an initial comment period and a reply comment period. After the comment period has closed, the agency must review the comments filed and will ultimately issue a final rule, which becomes binding after publication in the Federal Register (and expiration of the 60-day Congressional Review Act clock).

Deregulation via notice and comment rulemaking differs in two important ways from the other avenues discussed above, both of which can potentially derail the agency’s plans. First, the public has the opportunity to provide input into the agency’s process. One need not be a lawyer to comment on a proposed rule, including a proposal to rescind an existing rule. Most open agency proceedings are listed at, where one can file a comment by uploading a document or even simply by typing into a text box and pressing “submit.” This comment process is not merely window-dressing. By law, agencies must review and respond to these comments in its final rule. More specifically, while an agency need not address every point in every submission, it must respond to significant comments in a reasoned manner to show that the major policy critiques were considered by the agency.

Second, opponents can seek judicial review of the agency’s decision. Like all final agency action, a decision to revoke or modify an existing rule is subject to review under the Administrative Procedure Act’s “arbitrary and capricious” standard. In the deregulatory context, the touchstone case is Motor Vehicle Manufacturers Association vs. State Farm Insurance Co., in which petitioners challenged the Reagan administration’s rescission of a Carter-era rule mandating that cars contain either automatic seatbelts or airbags.[9] The Supreme Court explained that both adopting a new rule and repealing an existing rule change the legal baseline, and therefore in both cases the agency must “examine the relevant data and articulate a satisfactory explanation for its action.”[10] Furthermore, because repeal involves a reversal of the agency’s previous views on the issue, this explanation should include a discussion of why the agency had changed its position.

In State Farm, the Department of Transportation justified the repeal by arguing that, contrary to the agency’s original estimate that 60% of cars would have automatic seatbelts and 40% would have airbags, the agency now believed that over 99% of cars would choose the seatbelt option. And this was problematic because new studies showed that most Americans would deactivate the automatic seatbelts, meaning the regulation would not measurably improve automobile safety. The Supreme Court unanimously vacated this repeal because the agency failed to explain why it did not simply adopt an “airbags only” rule instead, which, from a safety perspective, would seem to be a logical response if automatic seatbelts were deemed ineffective. The Court explained that the agency could repeal the rule completely, but it had to explain why it chose that path rather than the other options that were on the table.

Learning from State Farm, the Trump administration is beginning the notice and comment process to repeal select Obama-era regulations. One of the first out the gate is the Federal Communications Commission’s notice of proposed rulemaking to repeal the classification of broadband providers as common carriers. In a series of orders between 2002 and 2005, the Commission had classified broadband access as a lightly-regulated “information service” under the Communications Act. In 2015, to bolster its legal arguments in support of net neutrality, the agency reversed those orders and instead classified broadband access as “telecommunications service” subject to common carriage regulations. The Commission’s 2015 decision was a 3-2 party-line vote. The two dissenting Republican commissioners now constitute a 2-1 majority, and have proposed to repeal the 2015 decision. Consistent with State Farm, the notice of proposed rulemaking stresses the unintended consequences of the 2015 order, including a decline in broadband investment and a weakening of privacy laws. It also stresses that repeal would restore the agency’s original classification, and thus draws upon the agency’s prior deliberations to justify its current decision. Unsurprisingly, many of the comments have challenged these factual assertions, and the final rule must address these comments if the Commission decides to proceed with the repeal.

Politics and the Administrative Law of Deregulation

Of course, adhering to State Farm is a bit of a fiction. Ultimately, in these cases the agency has not suffered a change of heart, but of personnel. The Trump administration’s decision to repeal Obama-era regulations stems from a change to a political regime that favors less regulation. The same was true of the Reagan administration’s deregulation efforts, which drove the agency’s position in State Farm. In partial dissent in that case, then-Justice Rehnquist suggested that the Court should call a spade a spade and allow politically driven reversals. He would have held that “[a] change in administration brought about by the people casting their votes is a perfectly reasonable basis for an executive agency’s reappraisal of the costs and benefits of its programs and regulations.”[11] The majority did not directly respond to this point, although one might argue that if it worked reciprocally, such an explanation would eviscerate judicial review of agency action. A general regulatory or deregulatory mood may explain an agency’s broad goals, but it says nothing about why any particular rule is justified on the facts relevant to its merits. The Court was right to require a policy, rather than political, explanation for an agency’s action.

Lessons for the Trump Administration

By reversing policy statements with which it disagreed and passing Congressional Review Act resolutions against Obama’s eleventh-hour initiatives, the Trump administration has identified the low-hanging fruit in its deregulatory project. The lesson of State Farm is that repealing other, longer-standing policies will not be as easy. It will require agencies to undergo the same time-consuming process as they did to enact the rules that they seek now to undo—and to overcome similar opposition from those who support the status quo.

And this is where the Trump administration’s focus on short-term gains may hurt its longer-term goals. As an initial volley against the administrative state, President Trump announced a government-wide hiring freeze.[12] As Obama-era employees departed, this attrition has left agencies short of the personnel needed to do the blocking and tackling required under State Farm. His proposed agency budget cuts compound the problem, by denying agencies the resources they need to implement deregulation.[13]

As he left office, President Harry S. Truman remarked of his successor, “He’ll sit there and say, ‘do this’ and ‘do that’ and nothing will happen. Poor Ike—it won’t be a bit like the Army.” The Trump administration may soon learn a similar lesson the hard way. If the White House is serious about achieving lasting deregulatory change, it must do so methodically, on an agency-by-agency, regulation-by-regulation basis, and avoid securing minor budget victories at the expense of losing the agency resources necessary to effect that change.

[1] G.G. ex rel Grimm v. Gloucester County School Board, 822 F.3d 709 (4th Cir. 2016), vacated and remanded, Gloucester County School Bd. v. G.G. ex rel Grimm, 137 S.Ct. 1239 (2017).

[2] Texas v. United States, 201 F. Supp. 3d 810 (N.D. Tex. 2016).

[3] Gloucester County School Bd. v. G.G. ex rel Grimm, 137 S.Ct. 369 (2017).

[4] See Massachusetts Session Laws 2016 c.134, available at

[5] 5 U.S.C. § 801 et seq.

[6] 462 U.S. 919 (1983).

[7] Pub.L. 115-4 (Feb. 14, 2017) (SEC Disclosure Rule); Pub.L. 115-5 (Feb. 16, 2017) (Stream Protection Rule); Pub.L. 115-8 (Feb. 28, 2017) (SSA Rule implementing NICS Improvement Act of 2007); Pub.L. 115-11 (Mar. 27, 2017) (NASA rule on government contracts); Pub.L. 115-12 (Mar. 27, 2017) (DOI rule on land use plans); Pub.L. 115-13 (Mar. 27, 2017) (Education Department accountability rule); Pub.L. 115-4 (Mar. 27, 2017) (Education Department teacher preparation rule); Pub.L. 115-17 (Mar. 31, 2017) (DOL Drug Testing Rule); Pub.L. 115-20 (Apr. 3, 2017) (DOI Rule on Alaska Wildlife Refuges); Pub.L. 115-21 (Apr. 3, 2017) (DOL employee injury recordkeeping rule); Pub.L. 115-22 (Apr. 3, 2017) (FCC Data Privacy Rule); Pub.L. 115-23 (Apr. 13, 2017) (HHS Title X rule); Pub.L. 115-24 (Apr. 13, 2017 (DOL rule on municipal savings plans); Pub.L. 115-35 (May 17, 2017) (DOL rule on state savings plans).

[8] See 5 U.S.C. § 553.

[9] 463 U.S. 29 (1983).

[10] Id. at 30.

[11] Id. at 59 (Rehnquist, J., concurring in part and dissenting in part).

[12] See Presidential Memorandum Regarding the Hiring Freeze, January 23, 2017, available at

[13] See President Trump’s Taxpayer First Budget, available at; see Gregory Kreig and Will Mullery, Trump’s Budget By the Numbers: What Gets Cut and Why,, May 23, 2017, available at

Daniel Lyons is an Associate Professor of Law (with tenure) at Boston College Law School, where he specializes in administrative law, telecommunications, and Internet law.


Early Lease Termination Under G.L. c. 186, § 24: An Essential Escape Route for Tenants Who Are Facing Domestic Violence, Sexual Assault, or Stalking

lyonsby Julia Devanthéry

Legal Analysis


Survivors of sexual assault, domestic violence, or stalking often have to leave their homes and relocate to a safer place with little notice or planning in order to avoid harm by a perpetrator who knows where they live.  My client Olivia[i] (whose name and identifying information I have changed to protect her privacy) was raped by a college classmate. Olivia’s rapist was also her upstairs neighbor. After the incident she saw him coming and going from the building, in the elevator, and around other common areas. She was terrified of what he might do to her and ashamed every time she saw him. Olivia knew she had to get far away from the person who assaulted her as soon as possible. When she asked her property manager about moving out of her apartment, he very politely directed her to the paragraph of her lease that provided for a three-month rent penalty for early lease termination. Olivia could not pay three months’ rent in addition to moving costs for a new apartment.  Frustrated, scared, and confused, she hid from her rapist by locking herself into her apartment, missing school and work.  After several weeks, a local rape crisis center referred her to me. Fortunately, G.L. c. 186, § 24, “Termination of rental agreement or tenancy by victim of domestic violence, rape, sexual assault or stalking” (“Section 24”), was signed into law in 2013, after decades of law-reform efforts by survivors and housing advocates. Its purpose is to help tenants like Olivia leave their homes for safety reasons without incurring financial penalties. With our help, Olivia asserted those rights and moved out of her apartment in less than 24 hours.

Olivia’s case illustrates that because Section 24 is not well known or understood, survivors across the Commonwealth remain trapped in unsafe living situations and unable to break free from abuse for fear of financial liability. This article aims to familiarize the practitioner with the provisions of Section 24 so that its benefits may be more broadly utilized.

A Brief History of G.L. c. 186, § 24.

In recognition of the unique housing needs of survivors of domestic violence, sexual assault, and stalking, the Massachusetts Legislature passed a comprehensive statute aimed at decreasing homelessness among victims of violence and increasing their possibilities for escaping harm without financial repercussions, titled An Act Relative to Housing Rights for Victims of Domestic Violence, Rape, Sexual Assault and Stalking (“the Act”). The Act was patterned after the federal Violence Against Women Act (“VAWA”), 42 U.S.C. § 14043e-11, which also contains special housing protections for survivors. The Act established four new tenant protections for survivors of domestic violence, sexual assault, or stalking in the Commonwealth: the right to have locks changed at home, G.L. c. 186, § 26; the right to call upon law enforcement or the courts for protection against abuse without reprisal from a property owner, G.L. c. 239, § 2A; the right to terminate a lease early for safety reasons, G.L. c. 186, § 24; and a prohibition against discrimination based on a prospective tenant’s use of any of the lease-breaking or lock-change protections. G.L. c. 186, § 25. Section 24, the right to terminate a rental agreement early, mirrors a similar VAWA provision allowing early termination of Section 8 leases. See 24 C.F.R. 982.354(b)(4). Section 24 extends the right to flee an unsafe housing without penalties to all tenants in the Commonwealth, not just those who received federal housing subsidies.

After the Act became law, Lieutenant Governor Murray, who then chaired both the Governor’s Council to Address Sexual and Domestic Violence and the Interagency Council on Housing and Homelessness, summed up the intent of the legislation by saying, “[a]fter facing the emotional and physical trauma of abuse or assault, victims of sexual and domestic violence often struggle with ongoing concern for personal safety, housing instability or potential homelessness if they need to leave their residence. By partnering with the legislature and community based advocates, we are helping to improve the safety of victims in their own home and providing opportunities to improve their safety without further financial penalties.”[ii] Taken as a whole, and in light of the purpose expressed by lawmakers, it is clear that the law aims to protect survivors’ access to housing and to allow them as much flexibility as possible to decide whether relocating is their safest solution. The Legislature sent a firm message to tenants and landlords alike: survivors should not lose housing because of abuse, nor should they be trapped in unsafe housing by the financial obligations of a tenancy agreement if they must leave.

The Mechanics of G.L. c. 186, § 24.
Who Qualifies for Protection Under G.L. c. 186, § 24?

Pursuant to Section 24, a tenant may terminate a rental agreement or tenancy by giving written notice to the property owner that a member of the household has been a victim of domestic violence, rape, sexual assault, or stalking within the past three months, or that the tenant is in fear of imminent serious physical harm from such abuse. G.L. c. 186, § 24(a). Where the violence took place, the identity of the perpetrator, whether the perpetrator is part of the household, and whether the owner had any prior knowledge of the abuse are all irrelevant to the question of whether a tenant is eligible for early lease termination under Section 24 (nevertheless these are all issues landlords have raised with me in seeking to avoid complying with the law). The only threshold eligibility criteria are: (1) that the tenant or household member has experienced domestic or sexual abuse in the past three months, or is in imminent fear of serious physical harm, (2) that the tenant requests termination of the tenancy to the owner in writing, and (3) that the tenant vacates the unit within three months of giving the early termination notice.  G.L. c. 186, § 24(b).[iii]

The landlord may, but is not required to, ask for proof of the domestic violence, sexual assault, or stalking. G.L. c. 186, § 24(e). The statute provides three possible ways for a tenant to prove she is a victim eligible for early lease termination.  She can provide the owner with (1) a copy of a G.L. c. 209A or G.L. c. 258E protection order, (2) a record from a court or law enforcement agency documenting an act of domestic violence, rape, sexual assault, or stalking, stating the name of the perpetrator if known, or (3) a written verification by a qualified third party to whom the victim reported violence. G.L. c. 186, § 24(e)(1)-(3). Under the third method for verification, a qualified third party can be a police officer or law enforcement professional, a medical care provider, an employee of the Department of Children and Families or of the Department of Transitional Assistance, a licensed social worker, a licensed mental health professional, a sexual assault counselor, or a domestic violence victims’ counselor. G.L. c. 186, § 23(iii). The letter must be verified by the victim in a statement signed under the pains and penalties of perjury. G.L. c. 186, § 24(e)(3). Counsel for tenants should be attentive to issues of privacy and dignity when advising a client on which form of verification to provide, and given a choice between multiple methods, the one with the least detail is often preferable.

Once an owner obtains proof of the tenant’s status as a qualified victim he or she must keep such documentation confidential and may not share the content with any other person or agency unless the victim provides written authorization or unless required by a court or by government regulation. G.L. c. 186, § 24(f). Confidentiality is essential to victims, as the verification documents often contain private, embarrassing, or otherwise sensitive material. Owners cannot share this information with co-tenants or anyone else who may inquire about the reason for the early lease termination.

When Does Termination Occur?

If a tenant seeks early termination under Section 24, the tenancy terminates on the date that a tenant surrenders her interest in the premises, or the “quitting date”. G.L. c. 186, § 23. If the tenant has already vacated the unit, the quitting date is the date that she gave notice to the landlord of her intent to abandon the unit and not return. If a survivor left on the 1st of the month, but didn’t inform her landlord that she vacated until the 15th, the quitting date is the 15th. Id. If the tenant gives notice prior to vacating the unit, the quitting date is either (A) the date the tenant intends to vacate or (B) the actual date that she vacates after providing the notice. Id. If the tenant is still in the unit and informs the landlord of an intent to leave by the 1st day of the following month, but doesn’t turn over possession of the unit until the 15th, the quitting date is the 15th.

What Financial Obligations Exist Between the Parties?

Tenants are responsible for rent through the quitting date, but are not liable for rent after the quitting date. G.L. c. 186, § 24(c). This provision of the statute is drafted unartfully, and therefore it is the source of some misunderstanding among owners and their attorneys who routinely ask tenants to pay for the month after the quitting date before agreeing to release them from ongoing obligations under the rental agreement. This practice amounts to asking survivors to pay double rent in the same month (one month to the former owner and one to the next landlord at her new, safer apartment). Such a request is unaffordable for many and has the effect of trapping victims in unsafe apartments. Moreover, the practice of asking for a month’s early termination penalty is in direct contravention of the legislative purpose expressed by lawmakers involved in writing and sponsoring the bill. For example, Senator Jamie Eldridge, one of the co-sponsors of the legislation, issued a press release stating, “This bill will help those victims who face a financial and/or legal barrier to leaving their home and . . .  to vacate their lease or rental agreements without financial penalty.”[iv] Senator Cynthia Creem, another co-sponsor of the legislation, was quoted in the Governor Deval Patrick’s press release as saying that, “[t]his new law will provide important protections to victims of domestic violence, sexual abuse and stalking by giving them the ability to choose whether to stay in their residences or move without having to weigh their personal safety against financial considerations.[v] The statements of lawmakers involved in drafting and passing the statute repeatedly confirm that qualifying survivors should not suffer financial consequences for seeking protection under Section 24, and therefore owners should not seek rent from tenants for periods after the quitting date. Section 24 shifts the financial burden associated with an emergency relocation away from survivors and on to landlords, at least in terms of the cost of turning a vacated apartment over and re-renting it. This legislative choice likely reflects lawmakers’ conclusion that property owners are in a better position to absorb these costs than survivors of domestic violence or sexual assault.

After termination, owners are responsible for refunding the tenant any pre-paid rent, (including the last month’s rent) and the security deposit, if applicable, in compliance with G.L. c. 186, § 15B. G.L. c. 186, § 24(c).

Special Considerations with Regard to Co-Tenants.

Section 24 does not terminate the rental obligations of co-tenants who are not qualified to terminate their leases for safety reasons. G.L. c. 186, § 24(d). The practical implications for co-tenants is that while one tenant may be able to terminate the lease early and on very short notice, the remaining tenant continues to be bound by the terms of the tenancy with the owner. If the co-tenants were jointly and severally liable under the lease (which is very common), the remaining tenant continues to be liable for the entire monthly rent after the survivor vacates. On the other hand, if roommates had individual tenancy agreements with the landlord, the remaining roommate’s share would stay the same after the survivor vacated, and the landlord would receive less rent for the unit until a replacement tenant moves in. Given the incredibly low vacancy rates and high demand for rental housing in Massachusetts,[vi] owners working closely with remaining tenants should be able to identify a replacement co-tenant in a timely manner.

The Bar Can Give G.L. c. 186, § 24 Meaning by Raising Awareness and Respecting Its Precepts.

Section 24 has been law for over four years, yet there is a pervasive lack of awareness about the possibility of early termination, and an alarming level of misunderstanding among landlords and tenants alike about how the law works. Section 24 was enacted to allow maximum flexibility and minimal financial barriers to the most vulnerable tenants in the Commonwealth: those in the process of extricating themselves from abusive relationships and those who are fleeing homes with which their rapists or stalkers are familiar. Olivia, who is a rare example of a survivor who was able to access legal representation to terminate her lease early, now lives in a new apartment where she can focus on her studies and rebuild her life without the fear and trauma of living in the same building as her rapist. All qualifying survivors should be able to benefit from Section 24, not just those who can afford or find legal counsel. Increasing awareness and compliance with G.L. c. 186, § 24 will ensure that more victims have the opportunity to break free from abuse and claim the dignity of living in a home free from violence.

Julia Devanthéry is a Clinical Instructor at the WilmerHale Legal Services Center of Harvard Law School where she co-teaches and supervises students enrolled in the Housing Clinic and the Housing Justice for Survivors Project. Ms. Devanthéry received her juris doctor from Northeastern School of Law in 2009 and her B.A. in Sociology from Brown University in 2004.

[i] This article uses gendered pronouns to describe victims and survivors as “she.” This is in part to reflect the fact that all of my clients seeking the protection of G.L. c. 186, § 24 have been women, and also because there is ample research to support the assertion that most domestic violence and sexual assault cases involve female victims. See Molly Dragiewicza and Yvonne Lindgren, The Gendered Nature of Domestic Violence: Statistical Data for Lawyers Considering Equal Protection Analysis, 17 Am. U. J. Gender Soc. Pol’y & L. 229, 242-257 (2009).

[ii] Press Release, Governor Deval Patrick Press Office, Governor Patrick Signs Legislation to Provide Housing Rights and Protections for Victims of Sexual Assault and Domestic Violence (Jan. 3, 2013).

[iii] If the tenant fails to vacate within three months, the notice to terminate the tenancy is considered void.

[iv] Press Release, Team Eldridge, Housing Committee Advances Bill to Help Victims of Domestic Violence, Sexual Assault (March 26, 2012). (Emphasis added.)

[v] Press Release, Governor Deval Patrick Press Office, Governor Patrick Signs Legislation to Provide Housing Rights and Protections for Victims of Sexual Assault and Domestic Violence (Jan. 3, 2013). (Emphasis added).

[vi] For example, the vacancy rate for rental housing in Greater Boston is 3.4%, lower than at any time since 2001. The Greater Boston Housing Report Card 2016, The Trouble with Growth, How Unbalanced Economic Expansion Affects Housing, Prepared by The Kitty and Michael Dukakis Center for Urban and Regional Policy, Northeastern University for the Boston Foundation (2016).

Julia Devanthéry is a Clinical Instructor at the WilmerHale Legal Services Center of Harvard Law School where she co-teaches and supervises students enrolled in the Housing Clinic and the Housing Justice for Survivors Project. Ms. Devanthéry received her juris doctor from Northeastern School of Law in 2009 and her B.A. in Sociology from Brown University in 2004.

Fasten Your Seatbelt: The SJC Revises the Standard for Anti-SLAPP Motions

lyonsby Richard J. Yurko

Legal Analysis

In Blanchard v. Steward Carney Hospital, 477 Mass. 141 (2017), a surprise decision rendered in May, the Supreme Judicial Court altered significantly its “well-established” test—reflected in over two decades of precedent—for how trial courts are to evaluate special motions to dismiss under the Massachusetts anti-SLAPP statute. Whether this decision, which creates a new way to defeat the anti-SLAPP special motion to dismiss, effectively guts the statute or is just a bump on the road to a better understanding of it, will be determined by years of decisions in cases that have yet to arise. However, no less of an authority than the Court of Appeals for the First Circuit has already observed that Blanchard “dramatically shifted” the law under the anti-SLAPP statute.[i]


The anti-SLAPP statute, G.L. c. 231, § 59H, was enacted in 1994 over the veto of Governor William Weld to protect citizens from “strategic litigation against public participation,” i.e. “SLAPP” lawsuits.  The immediate impetus for the statute was a long and expensive case against a group of Rehoboth residents who opposed a local real estate development. The legislature concluded that persons who were exercising their rights of petitioning and speech were being punished for doing so by litigation in which their petitioning was the basis for claims against them. Effectively, what the statute does is create a statutory qualified immunity from suit for non-sham petitioning activities and an expedited procedure for dismissing claims arising from such petitioning, typically before the expense of discovery. The broad statutory definitions evidence a broad prophylactic purpose.

The new statutory immunity was heir to a long line of federal cases protecting the First Amendment right to petition.[ii] What is different in the anti-SLAPP statute, and in similar statutes in other states, is the expedited dismissal procedure (supplanting the normal rules for dispositive motions) and the fact that it was a positive enactment by the legislature rather than a rule developed by judges over time. Perhaps for these two reasons, the statute has not always been a favorite of the courts.  See, e.g., Adams v. Whitman, 62 Mass. App. Ct. 850 (2005)(extensive dicta worrying that the anti-SLAPP statute needlessly eliminates litigation-based tort claims).  For twenty years, however, the Supreme Judicial Court sought to apply the statute as written—regardless of some of the justices’ apparent doubts concerning its scope—and often corrected trial courts and appellate panels that strayed too far from the words of the statute. E.g., Duracraft Corp. v. Holmes Products Corp., 427 Mass. 156, 163 (1998) (noting that legislature “ignored [the] potential uses [of the statute] in litigation far different from the typical SLAPP suit”); Town of Hanover v. New England Regional Council of Carpenters, 467 Mass. 587, 594 (2014) (reversing trial court’s denial of special motion to dismiss and concluding that granting the motion was in accord with the reasons underlying the anti-SLAPP statute’s enactment). No longer.

The Original Duracraft Test

Two decades ago in Duracraft, the Supreme Judicial Court announced what it would later call the “well-established” two-part test for deciding a special motion to dismiss under the statute. E.g., Benoit v. Frederickson, 454 Mass. 148, 152 (2009); Wenger v. Aceto, 451 Mass. 1, 5 (2008). First, the moving party who asserts that she is being sued based upon her petitioning activities (typically the defendant) must demonstrate that the claims against her have no substantial basis other than her petitioning activities. If she succeeds in doing so, then the burden shifts to the nonmoving party (typically the plaintiff) to show, in the words of the statute, that the defendant’s petitioning “was devoid of any reasonable factual support or any arguable basis in law and [that] the moving party’s acts caused actual injury to the responding party.” G.L. c. 231, § 59H. Essentially, this second step in the analysis places a high, but not insurmountable, burden on the plaintiff to show early on that the petitioning was an unsupportable sham. Because the burden in the second step of the analysis was indeed steep, many plaintiffs focused their defense to a special motion to dismiss on the first step in the analysis, i.e. whether the defendant had shown that the plaintiff’s claims were based solely on defendant’s petitioning.

There are several key points about the Duracraft framework. First, the test is objective, just as the test in the statute is objective—whether the claim is based on petitioning and whether the petitioning has, or does not have, “reasonable factual support.”[iii] 427 Mass. at 165. Second, the focus is not on the ultimate question of the validity of the plaintiff’s claim but rather on the non-sham nature of defendant’s petitioning, as befits an immunity. Id. Third, the decision on whether immunity exists is generally to be made at the start of the case, otherwise the benefits of qualified immunity are lost. Id. at 161; see also Pearson v. Callahan, 555 U.S. 223, 231-32 (2009) (stressing importance of resolving immunity questions at earliest possible stage in litigation).

The Subjective Blanchard Addition

In Blanchard, the Court left the prior Duracraft framework intact but added an alternative way to defeat a special motion to dismiss. Rather than scale the high peak of showing the petitioning was a sham under the second part of the Duracraft test, plaintiffs can now convince a trial court judge that their “primary” motivation for each claim “was not to chill the special movant’s legitimate petitioning activities.” To make this alternative showing, “the nonmoving party must establish, such that the motion judge may conclude with fair assurance, that its primary motivating goal in bringing its claim, viewed in its entirety, was not to interfere with and burden defendants’ petition rights, but to seek damages for the personal harm to it from the defendants’ alleged legally transgressive acts.” Blanchard, 477 Mass. at 160 (internal quotations and citations omitted).

This “motivation” standard finds no antecedent in the language of the statute itself which, as noted, focuses strictly on the merits of the petitioning activity. One might wonder what plaintiff will not now file an affidavit calmly assuring the trial court that burdening the defendant’s petitioning was the farthest thing from its mind? Would not the Rehoboth developer, beset by neighborhood opposition, have done just that? By grafting a subjective branch onto an otherwise objective test, the Court has opened a Pandora’s Box of unanswered questions like:

(a)  How does one prove or disprove, without discovery, the plaintiff’s “primary motivating goal”?

(b)  If one of the chief goals of the anti-SLAPP statute was the earliest possible termination of potential SLAPP suits before the time and expense of discovery, how can such discovery be avoided if the question now turns on the plaintiff’s “primary” motivation?

(c) How is a trial court, or for that matter a reviewing court, supposed to weigh an affidavit from a plaintiff swearing it does not seek to burden the defendant’s petitioning, against objective—but inconclusive—evidence to the contrary?


None of the parties in Blanchard argued for this shift. The Court did not request briefs from the parties or amici on a possible new test. Yet the Court altered two decades of its own jurisprudence on the subject. So, one has to wonder why the Court was moved to announce a new test.[iv]

The Court proffered one stated rationale for the new test.  It suggested that the Duracraft “framework,” which simply implemented the objective language of the statute, was much broader than the “fundamental statutory concern” of just eliminating SLAPP suits. 477 Mass. at 155. According to the Court, without the benefit of a citation to the statute, the statutory language sweeps more broadly than the legislative intent because “[i]t is only … the actual ‘SLAPP’ suit,” i.e. the “meritless claim[] targeting legitimate petitioning activity,” that “the Legislature intended to stop early in its tracks.”  Id. at 157.

This justification contradicts extensive Massachusetts precedent by purporting to create a clash between the intent of a statute and its words. As this very Court confirmed just three weeks later, Massachusetts courts are supposed to look for the statute’s intent in the words of the statute. “Where the language of the statute is clear and unambiguous, it is conclusive as to legislative intent.”  AIDS Support Group of Cape Cod, Inc. v. Town of Barnstable, 477 Mass. 296, 300 (June 14, 2017)(collecting cases). The words of the anti-SLAPP statute broadly defined petitioning activities, defined a qualified immunity for those activities, and crafted an objective test and an expedited procedure for enforcing that immunity.  That the immunity was prophylactic and certainly broader on its face than the one SLAPP suit that prompted the statute is evident from the face of the statute.  No linguistic gymnastics can derive a narrow “intent” standard from such broad language, especially where the statute was enacted over the veto of a governor who argued for a much more narrow law. See, e.g., David Kluft, The Scalpel or the Bludgeon? Twenty Years of Anti-SLAPP in Massachusetts, 58 B.B.J. 5 (2014). Had the Court followed its own precedent, which it reaffirmed three weeks later in the Barnstable case, it would not have set up a false clash between the words of the statute and its true “intent.”

The Court may have acted for two other reasons, implied but not fully stated in the decision. First, perhaps the Court might have thought the new test was necessary to keep the statute constitutional. Blanchard, 477 Mass. at 157-158. Second, the Court might have thought that a new test was necessary to save the plaintiff nurses’ claims, which didn’t seem to the Court to be SLAPP claims at all. Id. at 154. Neither of these intimated reasons is correct or, for that matter, particularly satisfying.

In Duracraft, the Court twenty years ago came up with the two-part test to resolve the supposed constitutional tension between preferring the defendant’s prior petitioning to the plaintiff’s new petitioning (where the latter seeks to hold the defendant liable for its petitioning). Duracraft, 427 Mass. at 167-68. Other than its flawed legislative “intent” argument, the Blanchard Court does not satisfactorily explain why the Duracraft balancing was now insufficient. Moreover, the Court never grappled with a fundamental flaw in the “constitutional balancing” argument to begin with. In enacting the anti-SLAPP statue, the legislature actually changed the substantive law of the Commonwealth,[v] as it is entitled to do. By creating what is, effectively, a qualified immunity from suit for non-sham petitioning, the legislature altered the substantive tort law of the Commonwealth. Upon such enactment, over a governor’s veto, the legislature effectively made claims based solely on meritorious petitioning activity no longer valid. There is no right, constitutional or otherwise, for a plaintiff to petition for redress under old, supplanted law. There is no tension between two types of valid petitioning, but rather only a tension between former law and new law—an easy “tension” to resolve.

Moreover, the Court could have held for the plaintiffs in Blanchard simply by applying Duracraft. In Duracraft, the defendant’s special motion to dismiss was deemed properly denied because the plaintiff’s claims were not based solely on the defendant’s petitioning. Duracraft, 427 Mass. at 168. There existed a written contract (a nondisclosure and confidentiality agreement) between plaintiff and defendant. Id. That contract was an additional, non-petitioning factor in the claims that arose from the defendant’s petitioning (his testimony at a deposition).

In Blanchard, the nurses sued their former employer, Steward Carney Hospital, for defamation allegedly arising from the employer’s comments to newspapers and in emails to all employees about the hospital’s termination of them. Although the claim may have been for defamation, it unquestionably arose from the employer-employee relationship and against an alleged termination “for cause” required by the nurses’ union contract with the employer. See 477 Mass. at 145 n.7 (noting the collective bargaining agreement and an arbitrator’s finding in favor of the nurses).  In short, what made the statements defamatory was the necessary suggestion that there had been cause for the nurses’ termination (or, in the words of Blanchard, the nurses’ “culpability,” 477 Mass. at 142), even though none of the nurses had been accused of any individual wrongdoing. The nurses’ defamation claim was not subject to the anti-SLAPP statute under a straightforward application of Duracraft because it rested, in part, on the employer-employee relationship and the union contract between the plaintiffs and defendant–just as the contract in Duracraft provided a substantial basis to the claim in addition to the petitioning. Thus, no new test was required for the Court to preserve the nurses’ defamation claim.

There is, of course, always the possibility that in subsequent decisions, the Supreme Judicial Court will return to an objective standard derived from the actual words of the anti-SLAPP statute. Often, decisions that seem like a wrong turn are, on further judicial reflection, treated as sui generis until a decade or two later when they are overruled. E.g., Lawrence v. Texas, 539 U.S. 558 (2003) (overruling Bowers v. Hardwick, 478 U.S. 168 (1986)).

But it is perhaps much more likely that, with a new judicially-created subjective defense to an anti-SLAPP special motion to dismiss, trial and appellate courts will spend a decade or more trying to reconcile the objective words of the statute with this new subjective judicial exception to a legislatively-crafted immunity. Given the judicial equivocation under the statute even before this latest decision, my prediction is that courts and judges will scatter in how loose or how narrow to construe this new judicially-created exception to legislatively-created immunity.  We are all in for a bumpy ride.

[i]  Steinmetz v. Coyle & Caron, Inc., 2017 U.S. App. LEXIS 11916 at *5 (1st Cir. June 29, 2017).  In Steinmetz, the First Circuit has certified another important anti-SLAPP question to the Supreme Judicial Court: is a paid consultant who is aiding a group in its petitioning covered by the statute?  The SJC’s occasional dicta on this question is internally inconsistent, and conflicts with its and the Massachusetts Appeals Court’s holdings in cases like Town of Hanover, infra.

[ii] See, e.g., BE&K Construction Co. v. NLRB, 536 U.S. 516, 524-25 (2002); Stern v. United States Gypsum, Inc., 547 F.2d 1329, 1342 (7th Cir. 1977), cert. denied, 434 U.S. 975 (1977); In re Primus, 436 U.S. 412, 426 (1978); United Trans. Union v. Michigan Bar, 401 U.S. 576, 585 (1971); United Mine Workers of America v. Ill. State Bar Ass’n, 389 U.S. 217, 222 (1967).

[iii] Indeed, the Supreme Judicial Court repeatedly eschewed any inquiry into defendant’s subjective motive for petitioning in the first part of the test. E.g., Benoit at 153; Office One at 122; Fabre v. Walton, 436 Mass. 517, 523-24 (2002).

[iv] The Blanchard case did have one significantly redeeming aspect.  It made short work, confined largely to a footnote, to the argument that somehow the anti-SLAPP statute’s special motion to dismiss was a violation of state or federal guaranties of the right to a jury trial in certain civil cases.  Blanchard, 477 Mass. at 158 n.24.

[v] In Steinmetz, the Court of Appeals for the First Circuit recognized that the Massachusetts anti-SLAPP statute (like the Maine anti-SLAPP statute, which was the subject of an earlier decision) was substantive law, not procedural law, and therefore ought to be applied in federal court under Erie RR v. Tompkins, 304 U.S. 64 (1938), in diversity cases.  Steinmetz, 2017 U.S. App. LEXIS 11916 at *2, 6 (citing Steinmetz v. Coyle & Caron, Inc., 2016 U.S. Dist. LEXIS 99631 at *6-8 (D. Mass. July 29, 2016).

Richard Yurko is the founding shareholder of Yurko, Salvesen & Remz, P.C. and has frequently authored briefs on behalf of amicus curiae parties on the anti-SLAPP statute.

SJC Remakes Search-and-Seizure Law to Keep Pace with Modern Realities of Smartphone Technology and Race Relations

omeararangavizby Ruth O’Meara-Costello and David Rangaviz

Legal Analysis

In recent decisions, the Supreme Judicial Court (“SJC”) has cast an increasingly skeptical eye on law enforcement activities in two areas of perennial controversy:  the search and seizure of cell phones and electronic data, and police encounters with young black men.  The SJC’s review of search and seizure matters has been stringent, as the court has demanded a specific evidentiary basis for searches in both the digital and physical realms. These cases implement in practice the principles that absent reasonable suspicion, an individual may voluntarily terminate a police encounter; before obtaining a warrant, the police must have a particularized reason to believe that evidence will be found in a place to be searched (including a specific folder within an electronic device); and officers need individualized suspicion of a suspect’s involvement in a crime before stopping and seizing the individual. In a series of cases, the court has breathed new life into these oft-stated and staid legal rules, particularly in the context of digital searches.

The court has also explicitly addressed the role of race in interactions between the police and the minority residents of the communities they serve. In doing so, the court has recognized the reality in which many black targets of police investigations live. The SJC has forced the criminal justice system – and the overwhelmingly-white players within it – to imagine what it is to be African-American in an over-policed and underrepresented community.  By analyzing what probable cause means in the context of digital searches and relying on social science to understand interactions between police and African-American suspects, the court has brought an added degree of rigor in applying Fourth Amendment principles to the realities of modern American life.

Digital Searches

First, in Commonwealth v. Dorelas, 473 Mass. 496 (2016), the SJC reviewed whether a warrant to search an iPhone was supported by probable cause.  Police had reason to suspect the defendant was involved in a shooting, and that his iPhone might contain incriminating evidence because the victim had been receiving threatening calls and texts.  But the warrant did not authorize a search of just call and text history; it allowed officers to search all of the phone’s other contents, including photographs. Executing the warrant, officers found a photo of the defendant holding a gun and wearing clothing similar to that of the alleged shooter. The defendant sought to suppress the photograph, arguing that there was no probable cause to support the search of the photographs (as opposed to call or text history) and that the warrant did not identify the items to be seized or places to be searched with sufficient “particularity.”

The SJC rejected both arguments in a 4-3 decision, but announced a more demanding standard for searches of the digital contents of a smartphone.[i]  The majority noted that given the vast “volume, variety, and sensitivity” of information stored in or accessed through a smartphone, permitting a digital search to extend anywhere targeted information could be found is a “limitation without consequence” in the digital world, because “data possibly could be found anywhere within an electronic device.” In light of those “properties that render an iPhone distinct from the closed containers regularly seen in the physical world,” searches of such electronic data require “special care” and must satisfy a “more narrow and demanding standard” than physical searches. But the majority reasoned that the search into the phone’s stored photographs met that standard because threatening photos received or sent via text could have been stored separately from the texts themselves.

The dissent argued that the potential connection to a threat did not justify a search of the phone’s photographs.  It emphasized a forensic examiner’s testimony that extraction of call and text history would have retrieved photographs attached to messages, eliminating any need to search all photographs separately stored on the device.  The dissent also argued that the warrant failed to satisfy the Fourth Amendment’s “particularity” requirement because it authorized a general search of the entire iPhone.  Given the expansive capacity of today’s smartphones, the dissent likened this to “limiting a search to the entire city.”  The dissent thus fully rejected the traditional “container” analogy that generally permits a search of any “container” or file that is capable of containing the evidence sought.

Dorelas reflects a closely-divided court struggling over how to translate analog constitutional rules to modern digital reality. Both the majority and dissenting opinions appreciated the need for a heightened standard on cell phone searches, though they took different approaches when considering the obligation to limit the search’s intrusiveness.

A few months later, in Commonwealth v. Broom, 474 Mass. 486 (2016), the SJC provided further guidance on the kind of evidence needed to justify a cell phone search.  The defendant in Broom was charged with the first-degree murder and rape of his former neighbor. His statements to police put at issue his whereabouts the night before the murder.  A search of “cellular site location information” (CSLI) – location data associated with the defendant’s cell phone – undercut the defendant’s claims about that night.  A search of the contents of his cell phone call log and text messages yielded a crude text message from the defendant to his fiancé suggesting that he was sexually frustrated.  On appeal, the defendant challenged admission of both the CSLI and the text message.

The court concluded that probable cause did not exist to search the cell phone.[ii] The court emphasized the heightened Dorelas standard, and concluded that the affidavit in support of the search warrant failed to describe “particularized evidence” that the defendant’s phone would contain evidence relating to the crime. The court completely discounted the detective’s statement that, in his training and experience, cell phones “store vast amounts of electronic data” and thus “there is probable cause”, explaining that such a “general, conclusory statement adds nothing to the probable cause calculus.” While the court found the error in Broom to be harmless, its decision put lower courts on notice that they cannot authorize digital searches merely based on an officer’s training and experience without the kind of specific supporting information present in Dorelas.[iii]

In Commonwealth v. White, 475 Mass. 583 (2016), the court made explicit what it had implied in Broom: “Probable cause to search or seize a person’s cellular telephone may not be based solely on an officer’s opinion that the device is likely to contain evidence of the crime under investigation.”  The search warrant affidavit’s factual basis for the request to search the cell phone in White amounted to two things:  (a) there was evidence that the defendant had participated with others in a robbery-homicide, and (b) the officer’s “training and experience” suggested that cell phones generally contain incriminating evidence of communications in multi-defendant cases.  The court found this basis insufficient, emphasizing that the existence of probable cause to arrest does not necessarily provide probable cause to search a suspect’s cell phone; the latter requires particularized evidence that the phone was reasonably likely to contain evidence related to the crime.  Absent such particularized evidence, a suspect’s cell phone cannot be searched.

Police Encounters

The court has also recently taken on the challenge of applying Fourth Amendment rules to the reality of modern racial dynamics. In Commonwealth v. Warren, 475 Mass. 530 (2016), the unanimous court held that an African-American defendant’s flight from the police does not give rise to probable cause for a subsequent search. The SJC emphasized reasons other than consciousness of guilt that an African-American might flee a police encounter: “Such an individual, when approached by the police, might just as easily be motivated by the desire to avoid the recurring indignity of being racially profiled as by the desire to hide criminal activity.” Citing an ACLU of Massachusetts report about the disproportionate impact of police stops on African-Americans, the court held that flight “add[s] nothing to the reasonable suspicion calculus.”  (That study, examining the Boston Police Department’s “stop and frisk” activity, concluded that 63% of Boston police-civilian encounters from 2007 to 2010 targeted African-Americans, who are less than 25% of the city’s population. The Department itself acknowledged that “[t]he study did show some racial disparities that must be addressed.”)

The Warren opinion recognizes the importance of perspective in applying legal doctrine.  It attempts to defeat stereotypes that only guilty people flee police encounters, and reconciles the justice system with the reality that black men in Boston have an innocent and legitimate reason to flee the police.

The court’s analytical approach is also noteworthy. As the foregoing cases make clear, the court has not hesitated to change the law to keep pace with changes in technology.[iv] Similarly, the SJC’s opinion in Warren suggests its willingness to alter criminal practice and procedure based on emerging social science research. This forward-thinking perspective is unusual – appellate practitioners are trained to rely upon legal sources: statutes, legislative history, constitutional provisions, and precedent.  Indeed, the defense attorney litigating Warren never cited the report about racially-biased police stops in his brief to the Appeals Court and SJC – justices of the Appeals Court cited the study in dissent, and the SJC relied on it to effect a sweeping change in doctrine.[v]  The court’s recent receptiveness to this type of outside-the-record social science information is worth noting by appellate advocates.[vi]

Finally, in Commonwealth v. Meneus, 476 Mass. 231 (2017), the court held that a search of a group of young black men who happened to be located near a crime scene was unconstitutional.  After gunshots struck a woman’s car, she described having seen a group of young black men run away. The SJC held that such a vague description – “a group of young black males” – falls far short of justifying a search of all people fitting that description. In the court’s words: “[T]he mere presence of a nondescript group of young black males standing near the scene of a reported shooting did not, standing alone, sufficiently narrow the range of possible suspects to include this group of individuals.”[vii] As in Warren, the court refused to rely on the defendant’s flight to find reasonable suspicion. Ultimately, despite the seriousness of the crime under investigation, the court’s decision in Meneus was a rebuke to the conduct of the police. In its emphasis on the need for specific evidence to support suspicion and rejection of the importance of proximity to a crime or presence in a high-crime neighborhood, Meneus complements Warren and emphasizes the court’s determination to stringently uphold constitutional protections for minority groups who may be unfairly targeted by law enforcement.

The complex legal issues posed by digital searches, and the reality of racial profiling, will undoubtedly continue to confront the criminal justice system in Massachusetts and elsewhere.  With a quartet of new members, and an additional seat to be filled in the near future, it remains to be seen how the SJC’s search and seizure jurisprudence will grapple with these questions going forward.

[i] The Majority opinion was written by Justice Cordy, and joined by Chief Justice Gants and Justices Spina and Botsford; Justice Lenk wrote the dissent, joined by Justices Duffly and Hines. The defendant was represented by an attorney in the CPCS Public Defender Division Appeals Unit. David Rangaviz, co-author of this piece, had no involvement in the case.

[ii] As to the CSLI, the SJC had previously ruled that the Commonwealth may obtain CSLI only pursuant to a warrant. Commonwealth v. Augustine, 467 Mass. 230 (2014). The Broom court held that the Commonwealth should have sought a warrant for the defendant’s CSLI, but that the error did not require reversal. The SJC found no prejudice in the evidence’s admission because (1) the CSLI was only for the day of and day before the murder, and (2) in light of the defendant’s DNA on the victim police had sufficient probable cause to retrieve his CSLI for those two days anyway. The court thus seemed to suggest that there was no prejudice because a warrant would have issued if sought. (The court has, however, previously rejected the notion that “an illegal warrantless search could be cured by proof that a search warrant, if sought, would have been issued and the evidence inevitably discovered.”  Commonwealth v. O’Connor, 406 Mass. 112, 115 (1989).)

[iii] The admission of the contents of the defendant’s cell phone was thus error, but the court upheld the conviction based on the strength of other evidence against the defendant, coupled with the fact that only a single text message was erroneously admitted.

[iv] Another recent opinion follows this trend. In Commonwealth v. Martinez, 476 Mass. 410 (2017), the court held that probable cause that the user of a certain IP address possesses child pornography is generally sufficient to justify a search of the residence assigned that IP address. The court nonetheless recognized that its holding may not “always” hold true as future technology “may further erode the connection between an IP address and a physical address” and “analysis hinges on fluid and rapidly changing technologies.” The court has recently heard argument in Commonwealth v. Keown (SJC-10593), in which the defendant argues that a warrant to search his laptop was insufficiently particularized, and therefore is likely to weigh in again on this issue in the near future.

[v] Justices Peter Agnes and Peter Rubin first cited the study in their dissenting Appeals Court opinions.  After their views did not carry the day – a three-justice majority of Chief Justice Rapoza and Justices Cypher and Green disagreed – a unanimous SJC embraced the dissenters’ opinion and rationale.

[vi] The SJC’s interest in evidence-based rulemaking is also apparent in recent decisions (all written by Chief Justice Ralph Gants) regarding eyewitness identification. In Commonwealth v. Crayton, 470 Mass. 228 (2014) and Commonwealth v. Collins, 470 Mass. 255 (2014), the court cited social science to limit the admissibility of in-court identifications. In Commonwealth v. Gomes, 470 Mass. 352 (2015), the court changed its model jury instruction regarding eyewitness identification to incorporate updated research, while “acknowledg[ing] the possibility that, as the science evolves, we may need to revise our new model instruction[] . . .”. Similarly, in Commonwealth v. Silva-Santiago, 453 Mass. 782 (2009), the SJC described a protocol, designed to decrease the risk of misidentification, for police to use before providing an eyewitness with a photographic array of potential suspects. The court recently reaffirmed this protocol’s importance in Commonwealth v. Thomas, 476 Mass. 451 (2017). The court will determine whether to extend Crayton and Collins in Commonwealth v. Dew (SJC-12225), currently pending.

[vii] The court also discounted the relevance of a police claim that the events occurred in a “high-crime area” and reiterated calls for caution regarding that claim in a reasonable suspicion analysis.

David Rangaviz is a staff attorney in the Appeals Unit of the Public Counsel Division of CPCS.

Ruth O’Meara-Costello is a partner at Zalkind Duncan & Bernstein LLP. Her practice focuses on criminal defense and student disciplinary matters