Work, once a matter of years of commitment to a single trade, career track, or even one employer, has in recent years undergone a radical transformation. From measuring work in units of decades, all pointed toward an engraved watch at retirement, we have swung far in the other direction, toward a world in which work will sometimes consist of brief transactions with multiple clients, customers, or employers. New technology has enabled apps such as TaskRabbit and Lyft, which allow us to measure work in units of minutes, seconds, or fractions of a mile moved.

But it is easy to be distracted by the sparkle of new technology, or the changes in our own experiences as consumers, and fail to see the underlying transformation of work for many Americans—from something that generally provided security and self-sufficiency, toward a world in which workers themselves bear most of the economic and personal risk. The jobs in the “gig” or “on-demand” economy that have transfixed journalists and pundits are still a relatively small part of the overall labor market, but the situation of workers in gig economy jobs is not starkly different from those in less technology-centric jobs whose work falls outside of traditional labor and employment law protections—taxi drivers, domestic workers, delivery drivers, and temp agency workers. We call this broader set of work experiences the “patchwork economy”1—a patchwork of jobs and a patchwork of protections—encompassing gig workers and non-gig workers alike, all of whom struggle with gaps in the safety net in various ways.

Politicians, academics, and the media have been absorbed in an argument about whether the gig economy is really something new and important,2 or how large the gig economy is. The Department of Labor has not conducted its Contingent Workers Survey since 2005, and thus has not yet been able to capture the rise of the gig economy in its count (although it expects to resume this survey in 2017). Some estimates of nontraditional work put levels as low as 8 percent of the workforce (4 percent as self-employed independent contractors, and another 4 percent as temporary workers), while others put the percentage of people doing at least some freelance work as high as 34 percent of the workforce.3 Economists Larry Katz and Alan Krueger recently found that the percentage of workers in alternative work arrangements (temporary help agency workers, on-call workers, contract workers, and independent contractors or freelancers) had increased from 10.1 percent of the workforce in February 2005 to 15.8 percent in late 2015, with online platforms like Uber still only 0.5 percent of the 2015 workforce.4 Debates over the numbers are important; better data makes for better social policy. But we do not need to answer all questions about the scope or importance of the gig economy to know that we need to make the safety net work better for those of us living in the patchwork economy. Designed for the workplaces and families of the 1950s, today’s safety net does not reflect the realities of life, work, and family in the twenty-first century.

Concurrent with the growth of the patchwork economy, though not entirely independent of it, there has been an astonishing rise in economic inequality and economic insecurity in the United States. The growth in inequality is, at this point, well-documented; as University of California, Berkeley professor Emmanuel Saez notes, from 1993 to 2014, Americans in the top 1 percent of incomes captured 55 percent of total U.S. income growth.5 Just as importantly, however, economic insecurity also dramatically increased. An indicator of economic insecurity developed by the political scientist Jacob Hacker shows a steady rise in insecurity from 1986 to 2012, measured as the share of families that experienced at least a 25 percent drop in available family income year to year, whether due to job loss or a medical emergency.6 These economic disruptions may lead to home foreclosures or poor credit, and become increasingly difficult to recover from. In such unforgiving circumstances, the disappearance of the social safety net is especially cruel.

The discussion here is intended to serve as an introduction to the issues and policy options surrounding the patchwork economy, and the need for a new social contract.

This report will outline the range of ways that the safety net could be improved for people in the patchwork economy, with an eye toward identifying some of the strengths and weaknesses of different approaches—how they could fit together, affect each other, or preempt each other. The discussion here is intended to serve as an introduction to the issues and policy options surrounding the patchwork economy, and the need for a new social contract. Much work remains on each of these questions, but the choices we make over the next few years may shape the terms of work and family life for decades to come.

To start, this report briefly outlines the historical background for the American social contract and the critical difference in legal status between employees and independent contractors, which shapes workers’ entitlement to various benefits and protections. The report then presents some possible solutions—first by exploring some of the efforts that help patchwork economy workers have a stronger voice, including efforts to organize workers and alternative business models that ensure that workers have more input and better working conditions. Next, the report looks at legislative and legal solutions, including those that enforce existing laws and those that would propose alternatives to the existing classification of workers as employees or independent contractors. Finally, the report looks at efforts to provide support structures that are not tied to traditional full-time employment, often known as “portable benefits”—whether those benefits be delivered privately, through government, or other mechanisms. It considers the various questions involved in structuring portable benefits, as well as their role in the broader social policy debate for patchwork economy workers.

The American Social Contract and Employees versus Independent Contractors

The social contract that helped build a broad middle class, expand prosperity, and reduce inequality from the 1950s through the 1970s was shaped to some extent by the New Deal, but perhaps even more by what Fortune magazine in 1950 called the “Treaty of Detroit.” That year, the United Auto Workers ended a period of intense conflict with the “Big Three” automakers (General Motors, Ford, and Chrysler) and negotiated a generous agreement first with G.M. and then with the other two companies. The contracts included a pension plan; a provision by which productivity increases would directly lead to wage increases; and health insurance (a few years earlier, large employers had begun to offer health coverage as a way around wartime wage controls).

The Treaty of Detroit reflected two choices that shaped work over the next several decades: first, a recognition by business that the security and well-being of its workers was in its own interest; second, a decision by labor that it was better off obtaining benefits linked to a specific employer than waiting for government to act.

The Treaty of Detroit reflected two choices that shaped work over the next several decades: first, a recognition by business that the security and well-being of its workers was in its own interest; second, a decision by labor that it was better off obtaining benefits linked to a specific employer than waiting for government to act. Similar contracts followed in the steel industry. Even nonunionized firms and industry sectors would later follow the pattern set in Detroit in 1950, helping to shape an American social contract that is deeply linked to the workplace and built on a specific model of employer and employee.7

Under U.S. employment law, however, not all workers who perform a task for someone else are “employees”—workers may also be “independent contractors.” While the difference between these two categories has been explored at length elsewhere,8 the distinction is important enough to the state of the today’s social contract to warrant describing briefly here.

Legally, an “employee” (a worker who receives an IRS Form W-2) is engaged in a labor agreement with an employer that requires the employee to give up control over certain aspects of his or her work life (for example, control over the hours she works) in exchange for certain protections and benefits under the law. An employee, for example, is obligated by law to be paid at least the minimum wage; to be paid overtime (if the type of work she is doing is eligible); to have taxes for Social Security and Medicare withheld by her employer; and, if provided with a retirement plan through work, for that plan to have to fulfill certain conditions. She is also able to organize collectively with her peers to form a union (depending on the exact type of job she does) and to sue if she has been discriminated against by her employer.

An independent contractor (sometimes known as a 1099 worker, from the IRS Form 1099) maintains more control over her work life, and in exchange, receives fewer protections under the law. More specifically, independent contractors are not required to receive any of the benefits that employees receive by virtue of their employment status—minimum wage, overtime, and so on. In addition, many other aspects of the governmental safety net—for example, unemployment insurance—assume an employer-employee relationship. Independent contractors, as a result, have much less of a governmental or employer-provided safety net.

Employers9 have strong incentives to treat workers as independent contractors rather than employees. It gives employers greater flexibility with their workforce, avoiding difficult and costly firings or layoffs; it can reduce costs for benefits; it can deter or evade unionization; and it can burnish company image for investors, who often look for hiring freezes or reductions in full-time employees as a sign of cost cutting. A 2000 study commissioned by the Department of Labor found that the primary reason employers use independent contractors or misclassify workers is “savings in not paying workers’ compensation premiums and not being subject to workplace injury and disability-related disputes,”10 while another study estimates that misclassifying a worker as an independent contractor can save an employer 20 to 40 percent on labor costs.11 While there are many abuses of independent contractor status, there are other instances in which the status is perfectly appropriate.

Note that people can have both W-2 jobs as employees and 1099 jobs as independent contractors at the same time. This is one of the reasons that the category “self-employed” is not the same as the category “independent contractor.” The Bureau of Labor Statistics measures self-employment by asking people if they identify as self-employed; someone with a mix of income from different sources may not identify as such even if a substantial portion of their work life is as an independent contractor. However, since there is substantial overlap between the categories, understanding who is self-employed also helps to understand the range of people who are independent contractors (see Table 1).

Table 1. Industries with Many Self-Employed Individuals (partial list)

Agriculture, especially crop production and animal production/aquaculture



Transportation, especially trucks

Financial activities, including finance/insurance and real estate

Professional and technical services

Management/administrative/waste services, especially landscaping and services to buildings and dwellings

Health care and social assistance, including child care

Leisure and hospitality

Repair and maintenance services

Personal and laundry services, including beauty salons

Source: Bureau of Labor Statistics, 2015.

It is also worth observing that a temporary worker—someone whose job is not permanent—may be either a 1099 worker or a W-2 employee. Either way, temporary workers (otherwise known as “temp workers”) face somewhat similar safety net risks as full-time 1099 workers, given that their work is inherently less stable.

As the economy changes and fewer work arrangements fall into the traditional employer-employee model, the lack of a safety net for independent contractors and temp workers becomes even more of an urgent problem.

Solution: Increasing Worker Voice in the Patchwork Economy

While there are a number of ways to improve and modernize the safety net for patchwork economy workers, one option is to increase worker voice, either through organizing workers (in unions or outside of unions) or through alternative business models that provide workers with more input and better working conditions.

Fostering Unionization, Worker Centers, and Other Forms of Organizing

Despite the overall decline in union membership in the private sector, organizing workers for collective action—in unions or outside of unions—remains a means to improve work in the patchwork economy. Organizing is a process as well as an outcome—it increases the power of workers relative to their employers or clients, making other good outcomes for workers possible (including potentially increasing the plausibility of some of the other policy ideas discussed in other sections of this paper).

Traditional union organizing by employees is protected by a variety of federal statutes, including the National Labor Relations Act (NLRA), but those protections are not available to independent contractors. Nonetheless, that has not prevented organizing in certain corners of the patchwork economy. Perhaps the highest profile example of such organizing, at least recently, was the work of the App-Based Drivers Association (ABDA), affiliated with the Teamsters union, to secure a law in Seattle enabling drivers for ridesharing companies such as Uber and Lyft to form unions despite their status as independent contractors.12 The law was passed by the Seattle City Council unanimously in December 2015; while critics claim that the law may be preempted by federal law and/or may violate antitrust laws, that determination will be made by the courts. Some legislators in California are trying a similar approach; the 1099 Self-Organizing Act, recently introduced in the California State Assembly, would let workers at gig economy platforms with ten or more independent contractors organize.13

Note that other unions have taken different approaches—for example, the International Brotherhood of Electrical Workers (IBEW) argues that Uber drivers are employees and has filed a petition with the National Labor Relations Board (NLRB) to represent 600 Uber drivers who serve New York’s LaGuardia Airport, an approach that will require the NLRB to rule on the question of whether Uber drivers are employees.14

Importantly, the history of organizing patchwork economy workers predates the ridesharing apps and other gig economy companies. In the absence of NLRA protection, some of these organizing groups have taken advantage of completely unrelated laws for leverage in bargaining. The New York Taxi Workers Alliance, for example, founded in 1988 and chartered as the first ever AFL-CIO union of independent contractors,15 takes advantage of the taxi medallion system and the structure of the taxi market to negotiate de facto contracts for taxi workers without the complicated checks and balances of NLRA bargaining. The union has played a major role in several big taxi fights of the past decade, including in 2012 winning “a livable income raise, first-time regulations of taxi companies, and a Health and Disability Fund for drivers, the first for taxi drivers nationwide and one of the first for independent contractors.”16 Other organizing groups, like the Freelancers Union, attract some members first with their services—for example, health insurance, dental insurance, liability, or disability insurance, for individuals who otherwise would be buying expensive individual plans or going uninsured—and then also go on to do advocacy.17 Still other organizing groups, like the National Domestic Workers Alliance, have run campaigns to pass “domestic workers bills of rights” in various states while building their organization of workers in the process. (See Table 2 for a broader list of groups organizing patchwork economy workers.)

Table 2. Examples of Groups Organizing Patchwork Economy Workers

App-Based Drivers Association (Seattle)

California App-Based Drivers Association

Freelancers Union

National Day Laborer Organizing Network (NDLON)

National Domestic Workers Alliance

National Guestworker Alliance

New York Taxi Workers Alliance

Transunion Car Service (New Jersey)

Source: Compiled by author.

Organizing is not easy in the twenty-first century, in any kind of workplace. But the history of the social contract suggests that organizing can make benefits and stability available to all workers, regardless of their status under employment law. In the patchwork economy, organizers have shown tremendous creativity in finding ways to provide tangible benefits to workers, win campaigns, and gain leverage. Apart from that, these organizing efforts help patchwork economy workers see themselves as part of a larger community, with shared interests. That said, it remains to be seen whether organizing groups can successfully make job stability a significant priority given the centrality of flexibility in the business models of patchwork economy companies.

Alternative Business Models

A second way that workers can obtain more of a voice as well as better working conditions within their companies is if the companies themselves consciously choose a structure that makes that happen. While there has been much talk about the importance of the flexibility inherent in using 1099 workers in small startup companies (and, perhaps, pressure from investors to maintain that flexibility), some high-profile companies in the gig economy, like Instacart, have converted at least some of their independent contractors to W-2 employees.18 Other high-profile startups in the gig economy have opted to hire only W-2 employees, including Hello Alfred, the personal butler service that won TechCrunch Disrupt SF 2014, and—newly as of this year19—Honor, the home health care company that won best new startup at the ninth annual #Crunchies.20 Honor, explaining its decision to shift its business model to W-2 employees and make workers eligible for benefits, suggested that this was a response to worker turnover, a desire for consistent training of employees, and the importance of long-term relationships between seniors and caregivers.21

Other companies have gone further, pledging to abide by the “Good Work Code,” a set of eight values inherent in “good work,” developed by the National Domestic Workers Alliance, including safety; stability and flexibility; transparency; shared prosperity; a livable wage; inclusion and input; support and connection; and growth and development.22 Obviously, the limitation of any “code of conduct” is that it is voluntary, and that only a minority of companies are likely to agree to it. That said, when companies that abide by such principles succeed by Silicon Valley standards, it helps to buttress the idea that rapid business growth and good working conditions are compatible.

That said, when companies that abide by such principles succeed by Silicon Valley standards, it helps to buttress the idea that rapid business growth and good working conditions are compatible.

The issues raised by the changing economy may go well beyond such basic questions as whether workers should be classified as employees, or companies should abide by certain standards. Some economists predict that technology, including but not limited to robotics, will continue to ensure that the returns to labor will fall, while the returns to capital—invested in automation—will rise. (This phenomenon extends well beyond the gig economy.) While a number of thinkers have suggested tax policy solutions to address these changes in the economy,23 labor economist Richard Freeman has argued that “to spread the benefits of robotization widely and prevent an inegalitarian nightmare, economic policy should seek to turn workers into the capitalist owners of the robots.”24

One limited way to foster employee ownership is to create worker cooperatives, such as the taxi cooperatives Green Taxi in Denver25 or Union Cab in Madison, Wisconsin26; Loconomics, a cooperative alternative to TaskRabbit;27 or Green Worker Co-ops in the South Bronx, which incubates worker-owned green businesses.28 Co-ops have the advantage of building equity for workers and giving workers voice in the clearest way possible—they own the business. The co-op movement has not seen much growth in recent years, but perhaps the increased growth of the patchwork economy will be a reason for that to shift.

Another incremental set of ways to broaden ownership of capital would be to create more mechanisms for employee ownership in traditional corporate structures, such as employees “owning shares held by an employee ownership trust; . . . receiving stock options as part of their pay; . . . having part of their pay come in the form of profit- sharing or other forms of group incentive pay; . . . being able to buy shares at low prices via employee stock purchase plans,”29 among other options. Employee Stock Ownership Plans (ESOPs) are the best known model of worker ownership, and enjoyed a renewed popularity after the 2008 crash.30 But ESOPs may also spread even more economic risk to workers, who are already vulnerable to downturns in their companies’ fortunes. Broadening the ownership of capital is a massive task that does not invite easy answers, but may require creative thinking about corporate law and corporate governance.

Solution: Legislative and Legal Solutions

Another set of solutions to the problems of the patchwork economy are legal and legislative—ranging from lawsuits and new legal frameworks about the proper classification of employees and independent contractors, to laws designed to provide non-employees with benefits or protections.

Enforcement (and Non-Enforcement) of Employment Laws

The rising number of people with nontraditional work relationships has tested the boundaries of existing employment law, in ways that can create risk for workers. In some cases, workers who are not legally “employees” have no legal protections against certain kinds of risk; in other cases, there are legal protections available but those protections are hard to enforce. Workers may also operate in legal netherworlds, where it is unclear or at least undetermined whether they are legally “employees” or just independent contractors.

Some of the risks inherent in being an independent contractor can be addressed through legislation targeting those risks in particular. While wage theft is a problem in many workplaces, for example, some freelancers—especially those who regularly work without contracts—have found it difficult to ensure that they get paid afterward. In response, the Freelancers Union has worked with the New York City Council to develop the Freelance Isn’t Free Act, which would require contracts with payment deadlines for certain freelance work and would set penalties for employers that do not pay, including, potentially, jail time.31

Other enforcement efforts focus on whether workers are in fact employees in the first place. The AFL-CIO recently declared that gig economy workers should be considered employees, not independent contractors.32 Uber and Lyft drivers in a number of locations have sued their companies, alleging that they have been misclassified as independent contractors, and that they are actually employees, given the amount of control that the companies exercise over their working terms and conditions. In turn, the companies have generally argued that the drivers choose this job in no small part because of the freedom it offers them—including the ability to choose their hours (and, implicitly, the freedom to work for both Lyft and Uber at the same time)—and that this flexibility would be lost if drivers were forced into a traditional employment relationship. In at least three individual cases, various California agencies have ruled that an Uber driver was an employee, but these cases apply only to the workers in question and have no precedential value.33 Class action lawsuits continue to work their way through courts, although California Lyft drivers recently settled with the company, agreeing that they will be classified as independent contractors but that the company will pay $12.25 million to those suing and amend the terms of service for drivers to make them more aligned with the definition of independent contractor.34 (The settlement has not yet been approved by a court.) In some places, including North Carolina, Arkansas, and Indiana, companies like Uber and Lyft are proactively dealing with such lawsuits by asking the states to pass laws requiring that workers for “transportation network companies” be designated as independent contractors.35

This report will not weigh in on whether certain workers should be employees rather than independent contractors, but starts from the premise that it is both inevitable and appropriate that—in the patchwork economy as a whole—there will always be at least some jobs performed by independent contractors and temporary workers. (Likewise, there will also always be some workers misclassified as independent contractors who deserve to be treated appropriately as employees.) The important question is how to provide workers who find themselves outside the traditional employer-employee relationship with an appropriate safety net.

The important question is how to provide workers who find themselves outside the traditional employer-employee relationship with an appropriate safety net.

Again, it is important to reiterate here that potential misclassification is not only a gig economy issue—it is a much broader patchwork economy issue. Indeed, as the AFL-CIO notes in its statement about gig economy workers, the debate about whether gig economy workers are employees is similar to the debate about FedEx workers; FedEx has historically classified its drivers as independent contractors (illegally, in the opinion of the AFL-CIO), while its competitor UPS has drivers that are unionized W-2 employees.36 The IRS estimates that millions of Americans are misclassified as independent contractors,37 while the National Employment Law Project has argued that “[i]ndependent contractor misclassification occurs with an alarming frequency in: construction, day labor, janitorial and building services, home health care, child care, agriculture, poultry and meat processing, high-tech, delivery, trucking, home-based work, and the public sectors.”38 (A guest piece in Forbes identifies a similar but slightly different list of industries by culling through statements from government officials about misclassification: janitorial services; construction; nursing; staffing; Internet services; transportation and trucking; cable companies; security; catering services; hotel/motel; oil and gas; landscaping; and car service/limousines.39) Litigation, however, is less than an ideal solution for either workers or employers, since it is ad hoc and leaves uncertainty about how any given situation would be resolved.

Indeed, some employers, for their part, have been frustrated with the ambiguity in employment law, fearing that they cannot provide benefits, education, and training for their workforce without increasing their litigation risk in misclassification lawsuits.40 Some companies have developed unusual partnerships to try to mitigate this problem; Lyft, for example, has a partnership with the retirement investing platform Honest Dollar to help its drivers get access to retirement plans—but is very clear in its materials that this is a partnership with another company, not Lyft offering benefits.41 In an attempt to resolve some of the employment law ambiguity, the Department of Labor recently released guidance to clarify who is an employee under the Fair Labor Standards Act,42 but questions remain, and there are several other legal definitions of “employee,” including those under the Internal Revenue Code. As a result, a number of patchwork economy companies have expressed a desire for a safe harbor or other mechanism that would let them provide certain benefits to their workers without any effect on classification lawsuits.43

Rethinking 1099/W-2 Worker Classification

One ambitious method for addressing the potential legal netherworld created by nontraditional work relationships is to modify the legal structure categorizing those relationships. As already mentioned, the boundary between “employee” and “independent contractor” has huge implications for benefits and worker protections: employees (but not independent contractors) are entitled to the minimum wage, overtime, employer contributions to payroll taxes, unemployment if they lose their jobs, workers’ compensation, the right to organize collectively under the protection of the law, protection from discrimination in the workplace, and other protections.

One way to address that is to use legislation to create a new category of workers—not quite 1099 independent contractors but also not W-2 employees. Last year, for example, the CEO of Stride Health wrote an article in Newsweek, suggesting that legislators create an “On-Demand Contractor” category.44 Several others have latched onto the idea of a “dependent contractor,” a category that exists in Canada and Germany but not in the United States.45 (Of course, the fact that it exists elsewhere is also a drawback, because—as Harvard Law professor Ben Sachs points out—the category has a specific meaning in other countries that may not make as much sense in the United States.46)

More recently, Seth Harris and Alan Krueger—the former acting secretary of labor and head of the Council of Economic Advisers in the Obama administration, respectively—have proposed one of the more comprehensive conceptions of a third category, which they call “independent worker.”47 This proposal, which would require amendments to a number of statutes, would allow independent workers to organize and bargain collectively (through an amendment to antitrust laws rather than the National Labor Relations Act), to be covered by civil rights laws, and to have wages withheld from their paycheck for tax purposes, among other changes to current law. Independent workers would have the ability to pool with other independent contractors to receive certain benefits such as disability insurance, retirement accounts, liability insurance, and others, but would not be covered by minimum wage, overtime, or workers’ compensation (workplace injuries would generally be addressed through the tort system).

People have continued to debate what kinds of protections should be included for a third category of worker48 and whether the gig economy is still evolving so rapidly that it would be premature to legislate in this area.49 Others, like Ben Sachs, researcher Alex Rosenblat, and the Economic Policy Institute, question whether the business model of companies like Uber and Lyft is really new enough to justify a third category of worker, given that this kind of third category would be less protective of workers than classification as an employee.50 Yet given the scope of legislative changes needed to make a third category of worker function appropriately and the fact that many of those changes would need to be done at the federal level, it seems clear that such legislative change—if even needed and/or advisable—is a project for the longer term, not the immediate future.

It is, of course, worth noting that even without creating a new, third category of worker, legislators could theoretically require that independent contractors (or certain independent contractors) be provided with specific benefits or protections (for example, tax withholding or pre-tax benefit contributions). Arguably, the bills that let independent contractors organize do just that. That said, workers who work for multiple employers would have a hard time taking advantage of many of those benefits if the benefits are not portable between employers—a problem that brings us to the next section of this report.

Solution: Creating Portable Benefits for Workers

The benefits provided to workers under the assumptions of the Treaty of Detroit, such as health insurance and pensions, traditionally have been tied closely to a particular employer and a particular job. Portable benefits would be the opposite: benefits that are available to all workers, that continue from job to job, and across multiple jobs at once. Portable benefits, however, can still take multiple forms—benefits can be provided through programs that are not connected to work or the employer at all; or through programs that involve employers but establish benefits that can be provided across employers.

Portable Benefits Not Connected to Employment

The most familiar form of benefit that is portable and not directly connected to employment is provided through universal government programs of various kinds. Perhaps the best example of this is in health care, where the Affordable Care Act (ACA), after many decades in which access to health insurance was bound up with employment, finally ensured that individuals and families, regardless of their current employment status, have access to affordable health coverage. Labor market trends suggest that the ACA is enabling greater voluntary part-time work by people who were previously bound to full-time jobs for the sake of health coverage.51

Other kinds of benefits could be offered through universal or broad-based programs. For example, Gene Sperling, the former director of the National Economic Council under both President Bill Clinton and President Barack Obama, has proposed creating a universal 401(k), which would give everyone—regardless of their employment—access to a retirement program and some government matching funds to incentivize saving.52 (The MyRA recently created by the Obama administration offers at least the framework for a universal retirement fund;53 Social Security also provides a base on which a fuller universal retirement program could be built.) Benefits ranging from wage insurance to disability insurance could also be designed universally; there is no reason our social safety net needs to be tied to specific employers at all. But it depends on the public having the political appetite for government to play a large role—essentially reversing the decision made in the Treaty of Detroit to pursue benefits through employers rather than a reluctant political system.

Benefits ranging from wage insurance to disability insurance could also be designed universally; there is no reason our social safety net needs to be tied to specific employers at all. But it depends on the public having the political appetite for government to play a large role—essentially reversing the decision made in the Treaty of Detroit to pursue benefits through employers rather than a reluctant political system.

A quintessential example of a universal benefit that is not tied to employment and could provide stability for people in complex job situations is the universal basic income (UBI) or its variants, such as the negative income tax. (The Earned Income Tax Credit is a mild version of the negative income tax, but is closely tied to employment.) These ideas, largely out of favor since the 1970s, have recently been revived by some in Silicon Valley as well as a few libertarians, as a way to, with minimal bureaucracy, compensate for a growing economy that still fails to provide stable employment and benefits to a significant portion of the labor force. Some see it as a necessary step in an economy in which automation leads to a long-term reduction in employment.54 While interest in the UBI is limited for now, it suggests the beginning of a conversation about the extent to which social policy should be tied to employment at all in the twenty-first century.

A second form of portable benefit not tied to work is one in which workers purchase benefits on their own, often from organizations that cater to people doing freelance, temporary, or independent contractor work (organizations like the Freelancers Union, Peers, or MBO Partners). Some of these organizations function as the “employer of record” for workers who would otherwise not have access to employer-based benefits, enabling them to sign up for benefits like tax withholding, group plans for vision and dental insurance, and access to employer-based 401Ks.

An often-overlooked form of benefit that can be either independent of specific work or connected to a person’s job is skills training. Skills are inherently portable; once acquired, skills are non-excludable from any future position that a worker may hold. But skill training also often requires a large up-front investment in both time and money. As a result, skill training is particularly at risk in an economy in which people are no longer tethered to one employer for long periods of time. Moreover, as University of California, San Diego professor John Ahlquist asks, “Who will bear the risks of investing in skills that may become obsolete?”55 While many arrangements provide skills to workers in a particular community, often through community colleges, those programs are often geared to, and directed by, the largest and most established employers in a community.56 The White House has put a renewed focus on apprenticeships in recent years,57 but apprenticeships may not work as well for workers in fields dominated by short-term gigs. (Of course, some traditional methods to finance education, like federal student aid and tax credits, can also be viewed as a way to gain skills training.) It remains to be seen how skills training can be best provided in an increasingly mobile and changing economy.

Portable Benefits Funded through Work Arrangements

Portable benefits can also be structured in ways that include employers, by allowing, encouraging, or requiring them—or their customers—to pay a portion of the cost of benefits for patchwork economy workers. Several variations on this kind of portable benefits model already exist, especially in industries where labor unions serve as intermediaries. Many unionized construction workers, for example, have multi-employer plans with an “hour bank” system, where employers pay a pro-rated amount into the fund based on the number of hours that the worker worked for that employer; the rate that employers pay into the fund is determined through the employer’s collective bargaining agreement (CBA) with the union. Similarly, the SAG-AFTRA, the union that represents approximately 160,000 actors, singers, news editors, and other media professionals,58 has two sets of pension and health plans for its members.59 The two pension plans are funded solely by the employers, at a rate determined by the CBAs negotiated with the unions, while the health plans are funded by both employer and participant contributions.60 A somewhat different funding model is the Black Car Fund, which by statute provides workers’ compensation coverage for black car drivers in the state of New York (including drivers for Uber and Lyft), funded by a customer surcharge on the cost of the trips.61

A number of people have proposed the brief outlines of portable benefits that could serve a broader set of workers. In “Common Ground for Independent Workers,”62 an open letter published on Medium, a group of approximately forty people including economists and people from start-ups and labor groups (including former SEIU President Andy Stern; the two founders of Lyft; and the CEO of Handy), laid out a commitment to both stability and flexibility, and supported the creation of portable benefits that are independent of income source; flexible and pro-rated; portable; universal, regardless of employment status; and supportive of innovation.63

The proposals thus far seem to agree on the basic outline for benefits: a package of portable and universal benefits, likely administered through payroll deduction and likely pro-rated in some way based on the amount of work done for a given employer.

Several other people have tried to flesh out models for work-related portable benefits in somewhat more depth. The proposals thus far seem to agree on the basic outline for benefits: a package of portable and universal benefits, likely administered through payroll deduction and likely pro-rated in some way based on the amount of work done for a given employer. The proposals differ in other elements, however—for example, New America senior fellow Steven Hill’s version (which he calls “Individual Security Accounts”)64 goes into much more detail on how much the accounts would cost, while David Rolf and Nick Hanauer (president of SEIU 775 and entrepreneur/venture capitalist, respectively), writing in Democracy, are much more focused on which benefits should be included in their “Shared Security Accounts” (“Mandatory accrued benefits should include a minimum of five days a year of paid sick leave, 15 days a year of paid vacation leave, a matching 401[k] contribution, and the same health insurance premium contribution as currently required under the Affordable Care Act. . . . Mandatory insurance benefits should include unemployment, workers’ compensation, paid maternity, paternity, family, and medical leave.”).65 While both the Hill and Rolf/Hanauer proposals are agnostic about who administers the portable benefits, Sara Horowitz, the founder of the Freelancers Union, has called for benefits to be “administered by unions, nonprofits, faith-based groups and other community organizations that would collect payments and distribute benefits when freelancers needed them.”66 And Senator Mark Warner, who has been supportive of the basic idea of portable benefits, was one of the few to suggest that consumers, in addition to employers, could help fund benefits for workers.67

None of the proposals is fully concrete yet, and many options for the structuring of portable benefits still remain (see Table 3). All of these options have pluses and minuses associated with them. Programs that let people opt into receiving benefits may be politically easiest because they may cost less and “preserve the most optionality,” but they also risk creating moral hazard in ways that may increase the costs of the benefits overall. Similarly, programs that let companies and employers opt out of participation may be politically easiest at a time when few companies offer benefits to patchwork economy workers, but if all employers in a given sector were required to participate, none would be at a competitive disadvantage. Using external partners to provide funding might help to decrease the costs of the benefits to workers and employers, but would likely be less stable for the long term. Creating portable benefits at the local or state level would allow for experimentation and would substantially increase the chances of implementing such benefits in the short term, but risks setting up different systems and different regulations in various places across the country.

Table 3. Key Questions in the Design of Portable Benefits
What benefits or protections are encompassed? Examples: health care, retirement, unemployment insurance, workers’ compensation, paid leave, overtime, tax withholding, liability insurance, disability insurance, paid sick days or vacation days, skills training, etc.
Who will fund the benefits?


Employers/companies/other wage providers


External parties (for example, foundations or worker organizations)

customers (for example, Uber passengers)

How will the funding be structured?




Who will administer the benefits?

Private sector third party (for example, insurance company)

Nonprofit third party (for example, nonprofit insurance provider)

Worker organization (for example, union or worker center)


Who would be eligible for portable benefits?

All workers (universal)

All 1099 workers

Only a subsection of 1099 workers (for example, those in a particular field)

Who will receive the benefits?

All eligible workers

Only those who opt-in/opt-out

At what level of government will benefits be mandated and/or regulated?




These questions draw on some thinking and writing by Libby Reder, Greg Nelson, and Natalie Foster, as part of the Aspen Institute’s Future of Work Initiative.

One of the biggest questions remains what kind of entity will administer the benefits. Horowitz, for example, has been quite clear that the reason she wants a nonprofit to administer the benefits is so that nonprofits will be able to benefit financially from these arrangements: “The organization providing portable benefits should have a social purpose, not simply a profit mission. It needs to be administered by a nonprofit, so that revenues get recycled back into the enterprise to sustain it.”68 Others in the labor movement have speculated that organizations administering portable benefits could ultimately become a way to organize those workers, although that seems by no means guaranteed (as one organizer commented, bureaucratic Taft-Hartley plans have not exactly been good organizing venues). The difference between having for-profit companies and nonprofit companies administer portable benefits may seem almost ideological—some will believe that for-profit companies will be more efficient and also may allow for more competition in the marketplace, while others will believe that nonprofits will put more of their money into their clients’ benefits. Whatever entity administers the benefits, it would certainly need to be regulated by the government in some way to ensure that benefits are reliable and serve the best interests of the workers.

Table 4. Summary of Safety Net Mechanisms
Mechanism Safety net elements Level of organization Who makes this possible? How quickly can this be done? Community benefited
Organizing Increases the power of workers relative to employers/clients, enabling workers to negotiate other things as well Local to national Workers Can start soon but takes time to build power Depends on who organizes
Alternative business models Workers get more voice and better working conditions within their company Business Business owners Quickly Workers for that specific business
Enforcement of employment laws (including through new legislation) Limited to specific law in question Local, state, or Federal Legislators (in conjunction with advocates, researchers, etc.); agency enforcement officials Agency enforcement can move quickly;  legislation depends on level of government; local and state legislation can likely move more quickly than federal People whose rights under that specific employment law were being violated
Enforcement of employment laws: Misclassification All the safety net elements associated with being an employee Courts; local, state, or federal government Courts and lawyers (in conjunction with advocates, etc.); legislators (in conjunction with advocates, researchers, etc.) Lengthy; court fights can take years Whoever was being misclassified
Non-enforcement of employment laws: safe harbor or other similar options May allow for additional benefits/training (but potentially also means workers do not get classified as employees when they otherwise would be) Complicated; depends on how the safe harbor is designed Legislatures and/or agencies deciding not to enforce Potentially lengthy; depends on how the safe harbor is designed Employer and potentially workers for that specific business
Shift in 1099/W-2 worker classification Depends on how classification designed; would include at least some of the protections of an employee Federal Federal government (in conjunction with advocates, researchers, etc.) Not quickly; federal legislation moves slowly People who fall into the new employment law classification and would have otherwise been designated independent contractors
Portable benefits Depends on benefits included— health care; retirement; unemployment insurance; workers’ compensation; paid leave; overtime; tax withholding; liability insurance; disability insurance; paid sick days or vacation days; skills; etc. Federal, state, or local Legislators (in conjunction with advocates, researchers, etc.) Relatively soon if state or local; not quickly if federal All eligible workers


The rise of the patchwork economy—including the gig economy, however many workers it actually encompasses—forces a long-overdue conversation about the nature of the American social contract and the future of work. As work has been transformed, over decades, from the industrial model of long-term employment in which the benefits and stability provided by the employer redound to the company’s benefit, toward one in which work itself is disaggregated, the American political system has largely ignored the questions raised: Is traditional union organizing under the National Labor Relations Act still a meaningful way to gain security? Do traditional employment law classifications still make sense? Can we provide a strong platform of economic and personal security that rests with the individual rather than the employment relationship?

As this report shows, there are multiple answers to each of these questions. That is a hopeful sign because it means that there are many ways to construct a viable political coalition that could develop solutions, drawing on the menu of options above or even combining them in various different ways. All that said, it is important to recognize that there is both a substantive and political path dependency that gets created by whatever option or options are chosen first. Legislation on classification could make lawsuits less necessary, for example—but such legislation could also make portable benefits politically more difficult, by appeasing part of the coalition that would otherwise support portable benefits. On the other hand, portable benefits—since they could be set up at the state or local level—might actually increase the political impetus for classification changes (or at least a federally created portable benefits package), since employers would risk being subject to different regulations depending on where they were located.

The various options also affect segments of the patchwork economy differentially. As we have seen, the problems of the patchwork economy span a range of industries, from drivers for tech platforms to domestic workers. But some policy options are more relevant for certain industries than for others. Sometimes this is self-evident; a bill requiring that workers for “transportation network companies” be designated independent contractors obviously will not affect misclassified construction workers. Other times, it is more implicit; while paid sick days are useful for everyone, they are perhaps more essential for a low-income janitor than for a higher-income professional consultant (and, conversely, the availability of liability insurance may be more useful to the self-employed lawyer than to a temp-agency worker).

Work in America has changed dramatically, and it will change further, in ways that we cannot foresee. The challenge for today’s policymakers is to build a social contract that works as well for this economy as the Treaty of Detroit structure did for its time—but more importantly, that is ready for the changes ahead of us.


Many thanks to Lee Drutman, Natalie Foster, Greg Nelson, Libby Reder, Andrew Stettner, and others for very helpful conversations and comments on this paper. This paper was produced in partnership with New America.


  1. This language draws heavily on David Weil’s concept of the fissured workplace. See David Weil, The Fissured Workplace: Why Work Became Bad for So Many and What Can Be Done To Improve It (Cambridge: Harvard University Press, 2014).
  2. See, for example, Lawrence Mishel, “Uber Is Not the Future of Work,” Atlantic, November 16, 2015,
  3. The McKinsey Global Institute estimates that self-employed independent contractors are approximately four percent of the labor force and temporary workers are another four percent. By contrast, a 2015 Freelancers Union survey estimated that 53.7 million Americans (about 34 percent of the workforce) do at least some freelance work; an Aspen Institute/Penn-Schoen-Berland survey estimates that 22 percent of the adult population has worked or offered services in the gig economy; and the Pew Research Center estimates that 10 percent of the national workforce is self-employed. Much of the disparity between these estimates comes from measuring slightly different populations, and from the treatment of what the Freelancers’ Union calls “moonlighters”—the 25 percent of workers who hold a primary job but earn other income as well, which ranges from essential to close the gap between poorly paid work and sufficient income, to activities that participants view as a hobby. There are also questions about whether independent contractors who work for only a single client for a long period of time should properly be classified as freelancers. Each of the reports handles these questions somewhat differently.   McKinsey Global Institute, “Connecting Talent with Opportunity in the Digital Age,” June 2015; Edelman Berland, “Freelancing in America: 2015,” Freelancers Union & Upwork, 2015,; “Forty-Five Million Americans Say They Have Worked in the On-Demand Economy, while 86.5 Million Have Used It, According to New Survey,”, January 6, 2016,; Pew Research Center, Three-in-Ten U.S. Jobs Are Held by the Self-Employed and the Workers They Hire (2015),
  4. Lawrence F. Katz and Alan B. Krueger, “The Rise and Nature of Alternative Work Arrangements in the United States, 1995–2015,” March 29, 2016,;  Anna Louie Sussman and Josh Zumbrun, “Contract Workforce Outpaces Growth in Silicon-Valley Style ‘Gig’ Jobs,” Wall Street Journal, March 25, 2016,
  5. Emmanuel Saez, “Striking It Richer: The Evolution of Top Incomes in the United States,” (updated with 2014 preliminary estimates), June 25, 2015,
  6. What Is the ESI?” Economic Security Index,; “Explore the Index,” Economic Security Index,
  7. Nelson Lichtenstein, Walter Reuther: The Most Dangerous Man in Detroit (Champaign, Ill.: University of Illinois Press, 1995); Frank Levy and Peter Temin, “Inequality and Institutions in 20th Century America,” NBER Working Paper, 2007,
  8. Seth D. Harris and Alan B. Krueger, “A Proposal for Modernizing Labor Laws for Twenty-First-Century Work: The Independent Worker,” The Hamilton Project,, and Richard A. Lazar, “Modernizing Employment Policies to Unleash a New Economy,” Techolicy, See also Robert N. Covington, Employment Law in a Nutshell, 3rd ed. (St. Paul, MN: West Academic Publishing, 2009).
  9. For simplicity, this report will use the term “employer” throughout when referring to the organization or company that pays an independent contractor; that terminology, however, is not meant to suggest that the relationship between the employer and the independent contractor is legally an employment relationship.
  10. “Independent Contractors: Prevalence and Implications for Unemployment Insurance Programs,” Planmatics, Inc., 2000,
  11.  Dale L. Belman and Richard Block, “The Social and Economic Costs of Employee Misclassification in the Michigan Construction Industry,” Institute for Public Policy and Social Research, Michigan State University, 2008, 11,
  12. Nick Winfield and Mike Isaac, “Seattle Will Allow Uber and Lyft Drivers to Form Unions,” New York Times, December 14, 2015,
  13. Jennifer Van Grove, “California Bill Would Let Gig Workers Organize,” San Diego Union-Tribune,
  14. Conor Skelding, “Union Files to Represent Uber Drivers Serving LaGuardia,” Politico New York, February 3, 2015,
  15. “Mission and History,” New York Taxi Workers Alliance,
  16. Ibid.
  17. Note that this is similar to the way the National Rifle Association works.  See, e.g., Peter Murray, “The Secret of Scale,” Stanford Social Innovation Review (2013),
  18. Davey Alba, “Instacart Shoppers Can Now Choose to Be Real Employees,” Wired, June 22, 2015,
  19. Elizabeth Weise, “No More Independent Contractors for this Gig Economy Startup,” USA Today, January 21, 2016,
  20. TechCrunch, Twitter Post, February 9, 2016, 12:13 a.m.,
  21. Weise, “No More Independent Contractors for this Gig Economy Startup,” USA Today,; Adi Kamdar, “Why Some Gig Economy Startups Are Reclassifying Workers as Employees,” On Labor, February 19, 2016,; “Caring for Our Care Pros: Honor Announces a Huge Change,” Honor, January 19, 2016,
  22. “Good Work Code,”
  23. Andrew McAfee and Erik Brynjolfsson, The Second Machine Age (New York: W. W. Norton & Company, 2014); Martin Ford, Rise of the Robots: Technology and the Threat of a Jobless Future (New York: Basic Books, 2015).
  24. Richard Freeman, “The Future of Work: Who Owns the Robot in Your Future Work Life?, Pacific Standard Magazine,
  25. Nathan Schneider, “The Future of Work: Owing What We Share,” Pacific Standard Magazine, September 1, 2015,; see also Trebor Scholz, “The Future of Work: The People’s Uber,” Pacific Standard Magazine, September 23, 2015,
  26. Alex Marshall, “An Old Idea for the New App-Based Economy,” Governing,
  27. Schneider, “The Future of Work: Owing What We Share”; see also Scholz, “The Future of Work: The People’s Uber.”
  28. Oscar Perry Abello, “NYC Set to Triple Number of Worker Cooperatives,” Next City (2016),
  29. Richard Freeman, “The Future of Work: Who Owns the Robot in Your Future Work Life?” Pacific Standard Magazine, August 17, 2015,
  30. Karen E. Klein, “ESOPs on the Rise Among Small Businesses,” BusinessWeek (2010),
  31. Rosa Goldensohn, “Wage-Theft Bill Threatens Employers with Jail, Requires Contracts for Freelancers,” The Insider (blog), December 7, 2015,
  32. AFL-CIO Executive Council, “The Policy Choices We Make Now Will Help Determine the Future of Work” (2016),
  33. Chris Roberts, “Updated: Another Uber Driver Awarded Unemployment Benefits,” SF Weekly, March 4, 2016,; Douglas MacMillan, “Another Uber Driver in California Ruled an Employee, not Contractor,” Digits (blog), September 10, 2015,; Mike Isaac and Natasha Singer, “California Says Uber Driver is Employee, Not a Contractor,” The New York Times, June 18, 2015,
  34. Isaac, “Lyft Agrees to Settle Class-Action Lawsuit with California Drivers,” Bits (blog), January 27, 2016,
  35. Heather Somerville and Dan Levine, “Exclusive: Uber Winning Make or Break Legal Battles across America,” Reuters, December 11, 2015,
  36. “The Policy Choices We Make Now Will Help Determine the Future of Work.”
  37. Treasury Inspector General for Tax Administration, “Employers Do Not Always Follow Internal Revenue Service Worker Determination Rulings” (2013),
  38. Catherine K. Ruckelshaus, “Testimony of Catherine K. Ruckelshaus of the National Employment Law Project Before the U.S. Senate Committee on Health, Education, Labor & Pensions,” National Employment Law Project, (June 2010)
  39. Daniel Fisher, “Is Your Company on the Independent Contractor Hit List?” Forbes, June 16, 2015,
  40. Carolyn Said, “Rise of Gig Economy Spurs Calls for New Workers’ Protection,” Government Technology, November 16, 2015,
  41. “Lyft x Honest Dollar: Introducing Savings and Retirement Solutions for Lyft Drivers,”
  42. David Weil, “Administrator’s Interpretation No. 2015-1,” United States Department of Labor,
  43.  Carolyn Said, “Rise of Gig Economy Spurs Call for New Workers’ Protection,” GovernmentTechnology (2015),; Lydia DePillis, “Tech Companies, Labor Advocates, and Think Tankers of all Stripes Call for Sweeping Reforms to the Social Safety Net,” The Washington Post, November 12, 2015,
  44. Noah Lang, “Employee or Contractor? Online Businesses Like Uber Need a New Category,” Newsweek, June 21, 2016,
  45. Lauren Weber, “What if There Were a New Type of Worker? Dependent Contractor,” The Wall Street Journal, January 28, 2015,; DePillis, “No, Driver Lawsuits Won’t Destroy the ‘Uber for X’ Business Model,” The Washington Post, March 13, 2015,; Caroline O’Donovan, “What a New Class of Worker Could Mean for the Future of Labor,” BuzzFeed News,
  46. Benjamin Sachs, “A New Category of Worker for the On-Demand Economy?, On Labor, 2015,
  47. Seth D. Harris and Alan B. Krueger, “A Proposal for Modernizing Labor Laws for Twenty-First-Century Work: The Independent Worker,” The Hamilton Project,
  48. See, e.g., the discussion at the event announcing the Harris-Krueger paper: “Modernizing Labor Laws for 21st Century Work,” (presentation, Modernizing Labor Laws in the Online Gig Economy, Washington, D.C., December 9, 2015),
  49. Harris, “Modernizing Labor Laws for 21st Century Work.”
  50. See, e.g., Ben Sachs’s commentary on the Harris-Krueger paper: Sachs, “Do We Need an Independent Worker Category?” On Labor, December 8, 2015, and Seth Harris’s response at: Harris, “A Response to Sachs on Independent Workers,” HarrisBlog (blog), See also Ross Eisenbrey and Lawrence Mishel, “Uber business model does not justify a new ‘independent worker’ category,” Economic Policy Institute, March 17, 2016,; Alex Rosenblat, “Uber’s Partner-Bosses (Drivers),” Medium, February 22, 2016,; Alex Rosenblat and Luke Stark, Uber’s Drivers: Information Asymmetries and Control in Dynamic Work (2015),
  51. Julie Verhage, “Dean Baker: There’s a Big Economic Benefit to Obamacare that isn’t Getting Much Attention,” Bloomberg Business, August 7, 2015,
  52. Gene B. Sperling, “A 401(k) for All,” The New York Times, July 23, 2014,
  53. “MyRA—A New Retirement Savings Account: Fact Sheet,”
  54. See Erik Brynjolfsson and Andrew McAfee, The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies (New York: W.W. Norton & Co., 2014). and Martin Ford, Rise of the Robots: Technology and the Threat of a Jobless Future (New York: Basic Books, 2015). The idea that government programs could provide stability for people in unstable job situations has even led some in the technology world to speculate that programs like universal basic income could compensate for an economy in which it might even benefit consumers more broadly to ensure that workers have multiple gigs on multiple platforms and remain independent contractors. See, for example, Steve R. Waldman, “1099 as Antitrust,” Interfluidity (blog), September 29, 2015,, and Ben Sachs’s response: Sachs, “1099 as Antitrust? On Labor,
  55. John Ahlquist, “The Future of Work: Risk Bearing and Risk Sharing,” Pacific Standard Magazine, September 3, 2015,
  56. Paul Fein, “Putting Community Colleges to Work,” Inside Higher Ed, April 17, 2014,
  57. Jeffrey Zients, “Expanding Apprenticeships to Invest in American Workers.” The White House (blog), September 9, 2015,
  58. “FAQs,” SAG-AFTRA,
  59. “About the Screen Actors Guild—Producers Pension and Health Plans,” Screen Actors Guild,; “About Us,” AFTRA Health & Retirement Funds,
  60. Ibid.
  61. “History,” The Black Car Fund,
  62. Byron Auguste et al., “Common Ground for Independent Workers—Principles for Delivering a Stable and Flexible Safety Net for All Types of Work,” Medium, November 9, 2015,
  63. Three of the organizers of this letter are continuing their work around portable benefits through the Aspen Future of Work Initiative.
  64. Steven Hill, “New Economy, New Social Contract,” New America (2015),,%20Social%20Contract_UpdatedFinal.34c973248e6946d0af17116fbd6bb79e.pdf; Hill, “Benefits for the Rest of US,” Washington Monthly, (January/February 2016):
  65. Nick Hanauer and David Rolf, “Shared Security, Shared Growth,” Democracy Journal 37 (2015),
  66. Sara Horowitz, “Help for the Way We Work Now,” New York Times, September 7, 2015,
  67. Mark R. Warner, “Asking Tough Questions About the Gig Economy,” Washington Post, June 18, 2105,
  68. Sara Horowitz, “Why Portable Benefits Should be a Priority in the New Economy,” Fast Company, December 8, 2015,