FROM ISSUE NUMBER 28 ~ SUMMER 2016 GO TO TABLE OF CONTENTS

TEXT SIZE A A A

The Future of Work

ELI LEHRER

In recent years, more than a few Americans in public life have spoken of a revolutionary "gig economy" that, through websites and smart-phone apps, is providing new types of work opportunities and new ways of linking consumers with service providers. Some say this new economy — dominated by firms like Uber, Lyft, and TaskRabbit — has produced positive effects, such as newly flexible work arrangements, increased innovation, increased self-employment, and a dominance of small employers. Others, however, say it has produced mostly negative ones: increased temporary employment, decreased job security, and less full-time work for many Americans.

Republican senator Marco Rubio of Florida — until recently, a presidential candidate — declared in October of last year that "an important truth facing us in this election...is that the American economy is fundamentally transforming" in the direction of new employer-employee relationships. The center-left Markle Foundation largely agrees. "America is in the midst of the biggest economic transformation in over a hundred years," the foundation wrote in announcing its multi-year, multi-million-dollar Rework America project, which is intended to explore the new nature of work in America's digital economy.

While Rubio is enthusiastic about this new gig economy, and Markle generally optimistic, not everyone feels the same way. To Democratic presidential frontrunner Hillary Clinton, current trends "are polarizing our economy" and leaving many Americans to make "extra money renting out a small room, designing websites, selling products they design themselves at home, or even driving their own car." Clinton frets that this transformation is "raising hard questions about workplace protections and what a good job will look like in the future."

Conventional wisdom says that loyalty between employers and employees is declining rapidly, for both parties. More than 90% of millennials surveyed by the executive-development firm Future Workplace say they expect to remain in a job for less than three years. A widely publicized report commissioned by the Freelancers Union in 2014 found what appears to be strong evidence of a growing gig economy: About a third of the workforce, 53 million Americans, are now freelancers, the report says. Conventional wisdom also holds that their numbers will increase. The accounting- and tax-software firm Intuit projects that, by 2020, 40% of the workforce could be freelancers.

The market certainly appears to be convinced that the gig economy is a big thing. Household-name companies like Uber, Thumbtack, and Airbnb have achieved valuations in the billions of dollars on the premise that, by unlocking otherwise-trapped capital and skills ranging from un-driven cars to underutilized carpentry talents, they are uniquely positioned to match workers and clients by providing everything from surgery to valet parking.

The idea of a new, revolutionary shift in the way Americans work offers a compelling narrative for policymakers. Those on the left, like Clinton, use it to argue for a stronger social safety net, positing that only the government can provide the protections that once came from traditional employers. Those on the right, like Rubio, exalt the growing ranks of rugged individualists — small-business entrepreneurs and innovators who, if largely left alone in an environment undergirded by the rule of law, modest regulation, and low taxes, will deliver a better future for everyone.

For all its appeal across the political spectrum, the narrative of the gig economy has a major problem: There is very little broad statistical evidence that the American workplace has fundamentally transformed, become dominated by smaller businesses, become more flexible, become more mobile, or reduced opportunities for full-time work. Indeed, the preponderance of the evidence indicates the opposite has happened. And that's not necessarily a good thing.

Today's workers are actually increasingly likely to stay with the same employer for a decade or more. Self-employment is decreasing. New business formation is down. Very small businesses' share of all employment is shrinking, while very large businesses now employ a larger percentage of the workforce than they did in the past. Start-up activity has declined. Some widely publicized increases in certain classes of "contingent workers" who are "on call" or working on a project basis don't point toward more "gig" work as most people think of it, and may actually point away from it.

That's a shame, because a stronger, more vibrant gig economy would actually serve to remedy a number of the economic and social problems modern America faces. Lawmakers should consider public policies that would create a more robust gig economy and provide stable, sustainable business and employment models, along with new labor laws and structures for certain companies and the people who work with them. These policies would include steps to pare down occupational-licensing requirements that hold back the gig economy. These should be coupled with a voluntary re-imagining of the employer-employee relationship through a new class of "flexible workers," a new designation of some work providers as "platforms," and the creation of new, voluntary, worker-controlled benefits exchanges that offer a portable, individualized, and flexible job-related safety net.

THE ECONOMIC NON-TRANSFORMATION

By most important measures, there has not been an increase in the percentage of employed people working part time or doing gigs instead of regular work on a full-time basis. My colleague R. J. Lehmann uncovered much of the data in a 2015 white paper, published by the R Street Institute. Since 1985, the percentage of the workforce working part time has ranged from a hair under 15.8% in the boom year of 2000 to 18.5% in 2010, as the country struggled to emerge from the Great Recession. From 2014 to 2015, as Uber and Airbnb racked up multi-billion-dollar valuations, the percentage of people working part time fell from 17.8% to 16.5%. The percentage of part-time workers who would rather work full time is also decreasing as the economy improves. Most of the variation closely tracks the economic cycle: Full-time employment increases during economic booms, and part-time employment goes up during recessions. As of January 2016, the percentage of the workforce doing a job part time was 16.4%, almost exactly the average of the past 30 years. The percentage of workers working part time who want full-time work tracks the economic cycle as well and is also now near historical norms.

Overall, employment patterns have shifted — in the direction of increased employment by big firms and a declining role for small businesses and the self-employed. Since 1993, the earliest year for which there is comparable data, the percentage of workers employed by small firms (one to four employees) has fallen slowly, but fairly consistently, from 5.6% to a bit under 5%. Meanwhile, the percentage of the workforce employed by firms with 1,000 or more workers has risen from 35.6% to 39.2%. Average tenure with the same employer has also risen in recent years, going from 4.9 years in 2004 to 5.5 years in 2014. The percentage of workers over 25 who have been with their current employer for more than a decade has also risen consistently, from 30.6% in 2004 to 33.3% in 2014. The percentage of people who are self-employed has steadily and consistently declined over the past several decades, falling from a high of about 7.3% in 1991 to 5.3% in 2015.

Start-ups are also on the decline. According to the Kauffman Foundation, the number of start-ups peaked at a bit over 500,000 per year in the 1980s and again during the middle of the last decade, but fell during the Great Recession. It still hasn't recovered entirely, even as the job market has returned to historical norms. A majority of this decline does seem to result from a shrinking number of workers between the ages of 24 and 54, the prime age bracket to start a business, but a still-significant portion stems from factors unrelated to the size of the workforce.

To be sure, there have been changes in the economy, including changes in the nature of work. The most important is a small but meaningful recent increase in the number of people who are "contingent workers": those working on an "on call" or project basis, finding jobs through temp agencies, outsourcers, and consulting firms. The most recent evidence comes from a March 2016 National Bureau of Economic Research paper by Lawrence Katz of Harvard and Alan Krueger of Princeton. The two find that the percentage of workers who are in some sense "contingent" grew from about 10.1% to about 15.8% between 2005 and 2015. While this is a meaningful change of some economic significance, it doesn't reflect growth in what most people would think of as a gig economy. Many of the workers counted as contingent are full-time employees who receive Form W-2s and might remain with their employers for years. Among the people counted as "contingent workers" are all workers who were "on call" at any point during the week (a common situation in retail and restaurant settings, even for full-time employees) or who were contracted out to other firms. Thus, contingent workers would range from part-time Walmart employees, who might be "on call" on Saturday in case the store is packed, to nearly everyone at the business-consulting firm Bain & Company. In fact, the largest and most notable increase found by Katz and Krueger was in exactly the type of work that firms like Bain do — consultants hired by one firm to work at another. But even if they are different from most conventional full-time employment, such jobs are actually more stable and less "gig-like" than the jobs at many very small businesses or start-ups, much part-time work, and all self-employment. It's reasonable to posit that many people who would have previously worked more or less for themselves as freelancers or start-up entrepreneurs are now working for consulting firms, outsourcers, and temp agencies.

In fact, the most important changes in the nature of work that have happened in recent decades have nothing to do with the gig economy. For example, the percentage of all workers who belong to labor unions has fallen from a high of a little more than 28% in 1954 to about 11% today, and the percentage of all workers employed by manufacturing enterprises has fallen from about a quarter in 1950 to 8% in 2015. But both the decline in unions and the not-coincidental decline in manufacturing that began in the 1950s have occurred at a consistent rate and have nothing to do with gig work. Furthermore, women, who made up less than a third of the workforce in 1948 (the earliest year for which there is data), now comprise almost half of all workers. Men's workforce-participation rates have fallen fairly consistently, in good economic times as well as in bad ones. On the other hand, overall workforce-participation rates for adults (about 63% as of 2015, and trending upward as the economy improves) roughly match their post-World War II average, though they are lower than they were for much of the 1980s and 1990s.

In sum, workforce data reveal increasing stability: Part-time work and self-employment are becoming less common; more workers have health benefits than before; and employees are more loyal to their employers and vice versa. More men are out of the workforce in larger numbers, but women have largely picked up the slack. While some new work arrangements are emerging, only a very small percentage of them involve what people think of as gig work, and many new types of employment relationships differing from those in the past may actually signal a decline in gig work. Overall, America's workforce is more stable and less geographically mobile than it has been in the past.

So from where do the statistics and perception of a huge and growing gig workforce come? Some parts of the perception do have a basis in reality: Perhaps because better scheduling and forecasting makes it easier for large employers to do it, more workers are "on call" and have to be available when an employer asks rather than having predictable hours. This might make some jobs in service settings feel less secure than they were in the past. But, since average tenure with the same employer is increasing, this doesn't actually seem to have weakened the employer-employee bond on a widespread basis, or it has at least been offset by other trends. The number of people employed by one firm and working at another has also increased by any measure, and there is evidence that "project-based" employment is up. The quick growth of firms like Uber and Thumbtack means that we may have a larger gig economy soon, though barely one American income earner out of 200 got any money from an app-based employer in 2015, according to a major study of JPMorgan Chase deposits.

But estimates saying that half of the workforce will be contract workers or freelancers within a decade don't really appear to reflect anything other than the authors' opinions. The much-cited Freelancers Union figure of 53 million freelancers is a current estimate (as of 2014) that doesn't even claim to represent an increase over the past. It includes those with full-time jobs who also "moonlight," the heads of employer-businesses who nonetheless count themselves as freelancers, and temporary workers in seasonal businesses who don't meet any common definition of "freelancer." A full-time tenured college professor who serves as a part-time consultant to an insurance company may well get the same 1099 tax form provided to freelancers — but he almost certainly doesn't think of himself that way. Counting people like this as freelancers (as the Freelancers Union does) stretches the definition to the breaking point.

In any case, some of the highest-profile "gig" workers may not remain as such. In April 2016, Uber agreed to a settlement worth as much as $100 million over some class-action lawsuits brought in California and Massachusetts that could have reclassified its workers as employees, rather than independent contractors. But this only postpones the day of reckoning for the company. Indeed, cities like Seattle continue to move toward changing the status of these workers, and even requiring such companies to recognize new forms of forced unionism. Trial lawyers and labor organizations are sure to bring new lawsuits, and only a handful of relatively conservative and non-urban states like Mississippi and South Dakota actually have laws that make the status of workers clear. And even these state laws are almost certain to come under scrutiny by federal officials and courts. Before long, it's almost inevitable that some courts will rule that the companies most associated with the gig economy are legally required to act as full-time employers, whether or not workers want it. And, while Uber's multi-billion-dollar valuation has, for now, allowed it to buy a reprieve and establish some innovative new labor structures, less well-capitalized companies may not have this option. Politicians, lawyers, and labor unions are simply looking for the next target.

The gig economy isn't threatening to take over the world. The preponderance of the evidence seems to indicate that it isn't growing at all, and may well be shrinking by most common definitions. And that's a problem.

THE BENEFITS OF A GIG ECONOMY

A vibrant and growing gig economy of the sort that doesn't currently exist could bring several benefits to U.S. labor markets. It would provide new employment opportunities, particularly for those who have a difficult time working a full-time job. It would open new frontiers for small businesses and help reignite the geographic mobility that has declined in recent years. Finally, it has the potential to reduce social-welfare spending significantly.

An expansion of part-time labor opportunities would accrue mostly to those who can't work full time. The most obvious advantages of gig work are intrinsic to its very nature: Lacking a full-time job with regular hours opens opportunities to spend more time on other worthy and beneficial pursuits, like caring for family, taking classes, working on a start-up venture, or pursuing an artistic endeavor. The largest social benefits, however, are likely to accrue to those who simply have a hard time holding down full-time work. In an ideal world, every adult of prime working age would have the drive, education, habits, and physical and mental health to work a full-time job, go to school, or care for family. But some adults lack one or even all of the things needed to be a fully productive member of society: They may be physically disabled, mentally ill, afflicted with substance-abuse problems, or products of a home or school environment that utterly failed to equip them with the discipline and habits needed for full-time work, school, or family responsibilities. A realistic assessment of the employment landscape reveals that not everyone who ought to work full time will do so, and many who would like to work lack the skills or habits to do so full time.

The problem of labor market non-participation is particularly severe among men. As Charles Murray points out in his book Coming Apart, the percentage of prime-working-age men who do not participate in the labor force has more than tripled since the 1950s. Likewise, in his important 1994 book, The Homeless, Christopher Jencks of Harvard University convincingly describes how the decline of day-labor markets (along with other changes, like the near-elimination of single-room-occupancy hotels) contributed to the rise of visible homelessness. The gig economy could help bring back employment opportunities for people currently stuck at the bottom.

Many of the problems that result in people — especially men — being unable or unwilling to work full-time jobs will, of course, require other public-policy interventions. But enabling a gig economy with more opportunities for part-time work is far easier than figuring out how to shrink the growing ranks of those who subsist on disability-insurance payments, or teaching adults with bad work habits to establish good ones. In many cases, the most effective way to improve the work behavior of marginalized populations is simply to encourage them to work. Gig work fits marginal workers perfectly and could provide them with the dignity and purpose that comes with having a job, rather than giving them nothing but public benefits.

A stronger gig economy could also help reverse the decline in U.S. business formation. Start-ups are important to economic progress, although not exactly for the reasons many politicians and business journalists assume. As a group of researchers found in a major 2010 NBER working paper, start-ups represent just 3% of total employment, but account for almost one-fifth of all gross jobs created. A robust gig economy could help amplify these contributions by better rewarding start-ups in general. The emergence of gig-economy platforms provides a potential model for entrepreneurs and offers some an opportunity to earn supplemental income while pursuing another venture.

Additionally, a gig economy could help reinvigorate Americans' geographic mobility. As Lori Sanders and I previously discussed in these pages (see "Moving to Work" in the Winter 2014 issue), declining geographic mobility in the United States has, over the past 60 years, been both a cause and a result of declining income mobility. Workers who could simply switch from gig to gig would find it far easier to move than those tied to a job and an employer. It's already the case that a Lyft driver in St. Louis would face fewer obstacles in moving to Chicago than an accountant would. The driver would not need to change his "job" or his "employer," and might not even need the company's permission (although he might well need to meet various regulatory requirements). Highly skilled workers can almost always find employment opportunities wherever they go; a thriving gig economy would extend similar flexibility to less-skilled workers.

Finally, a strong gig economy could vastly reduce the need to provide social support (publicly or privately) to those capable of work. While the circumstances that lead people into poverty are invariably complicated, at least some of those who require social assistance, at any given time, are simply victims of labor-market friction. They can work, wish to do so, and have the skills and habits necessary for work, but they simply can't find the person or firm who needs them immediately. The ability to be matched instantly with a willing buyer of one's skills would go a long way toward solving this problem. And gig-economy companies do exactly that.

For all of these reasons, it would be beneficial to build a stronger gig economy and encourage further growth in gig work. Rather than approve proposals that more or less outlaw gig work on platforms — as many states and cities led by liberals are considering — governments should instead enact policies that promote a gig economy.

CREATING A GIG ECONOMY

Four major public-policy changes would help create a stronger, more vibrant gig economy and reverse the trends that have led to a decline in gig work: a rollback in professional-licensing mandates; the creation of a new, purely voluntary employment status called the "flexible worker"; recognition of a new class of employer called a "job platform"; and a largely private social safety-net system made up of worker-controlled benefits exchanges to provide a minimum level of social protection to these flexible workers.

The single largest government-erected barrier to the widespread sale of useful skills in the gig economy and elsewhere is America's rapidly growing professional-licensing regime. The Institute for Justice's 2012 License to Work study shows that, in 1950, only 5% of all jobs required licenses to practice. Today, almost a third do. Nearly all of these licenses require a degree of formal professional training, which in turn requires time and money (often a lot of both). Even the Obama administration, hardly a paragon of limited government, has taken on the cause of professional-licensing reform, issuing a July 2015 report that excoriates current licensing regimes at the state level and calls for significant reform.

That said, efforts calling for the outright abolition of government-mandated professional licensing (sometimes framed as a limited statutory "right to earn a living") have gained little traction and face significant hurdles. Even more modest efforts to replace licenses with bonding requirements or test-based certifications have likewise seen little progress. A 2015 study in the Monthly Labor Review found a grand total of eight successful de-licensing efforts at the state level, most of which were immediately followed by efforts to re-impose the license. Once a licensed cartel, trade guild, or certified union exists, it's very hard to eliminate. In this context, many of the middle-ground efforts suggested by the Obama administration's report deserve consideration. Since existing licenses are so difficult to eliminate, states should enact the suggested comprehensive "sunrise" reviews that impose significant barriers to enacting new professional-licensing requirements and that subject them to rigorous benefit-cost analysis.

The federal government, exercising the original meaning of its power to "regulate" commerce, could also take assertive action to make licenses more portable and regular across state lines. A handful of interstate compacts (some of which are not yet actually in force) already exist for professions as diverse as nurses and insurance brokers. As the Obama administration suggests, these compacts deserve vigorous expansion and encouragement. Federal law could further be used to enable a wider range of services to be offered via the internet. As the economy changes and evolves, such policies would reduce the need for professional licensing and augment individuals' ability to earn income from gig work without directly confronting any of the powerful licensing cartels.

But simply ending barriers to employment isn't enough to ensure the creation of a gig economy. The overall decline in the availability of gig work, as well as the legal challenges (via state and perhaps federal labor-law decisions) to treat existing sharing-economy companies as per se employers, suggest a need for new labor classifications: "flexible workers" and "job platforms." Flexible workers would have a legal status somewhere between those of full-time workers and contractors (what some experts are calling a "third status"), and would interact with job platforms that provide them work. Examples of such workers might include drivers for Uber and Lyft and "taskers" for TaskRabbit, while Uber, Lyft, and TaskRabbit would legally become "job platforms" or "platforms." (Some well-known sharing-economy companies, like Airbnb and Turo, largely exist as platforms for unlocking capital rather than providing "gigs" and face no significant challenges as a result of current labor law; they probably wouldn't have any "flexible workers" or need changes in labor law to continue operating as they do.)

Like full-time employees, flexible workers would have basic civil-rights protections; platforms couldn't deny them work opportunities on the basis of characteristics like race, religion, or sex. Flexible workers would also have access to state and federal worker protections to ensure that platforms pay them as promised for work, so they would no longer have to go to court like independent contractors currently do. They would also gain an important new right: a near-limitless right to work. Platforms couldn't impose non-compete clauses on flexible workers or take action against them for undertaking any otherwise legal employment, even work for a direct competitor. Likewise, neither platforms nor labor organizations could bind flexible workers to "closed shop" laws or force them into collective-bargaining arrangements. Furthermore, flexible workers, unlike employees, would have control over their own hours and conditions of work. Platforms could impose deadlines and performance standards but couldn't closely direct their work, tell them exactly where to do it, or set particular hour requirements.

Flexible workers would not receive most of the fringe benefits generally provided to those who work for traditional employers. Platforms would not have the obligation to pay a portion of their payroll taxes, and they would not face penalties under the Affordable Care Act for not providing their flexible workers with health insurance. The very nature of flexible work, furthermore, means that some other benefits could not exist for flexible workers, at least in the way that most currently understand them. Paid time away from work, for vacation and sick leave, assumes a regular job with set hours and can't exist for someone who has no formal "job," per se, but simply works gigs on a number of platforms. Since flexible workers aren't ever "hired" in the conventional sense — they simply register to work or get a contract — they couldn't be "fired" or "laid off" and thus couldn't collect unemployment insurance either.

In addition to the obvious cost savings of not providing benefits or owing payroll taxes, platforms willing to partner with flexible workers would have another very significant benefit: a reversal of the long-standing labor-law assumption that anyone who performs work for a company is an employee, unless the company can demonstrate that he is not. Platforms willing to accept the flexible-worker designation could still be compelled by courts or regulators to treat workers like employees if they provide minute-by-minute instructions, impose non-compete clauses, or mandate specific hours of work, but the burden of proof would lie with the workers to prove these things. A new status would also enable firms to provide flexible workers with job-related tools and supplies, at no charge, without triggering investigations into whether they are "employers" who consequently owe thousands (or even millions) of dollars in back taxes.

A thornier question centers on whether such platforms could or should be subject to state laws governing workers' compensation. Under the laws of 48 states, all employers are strictly liable for on-the-job and closely job-related injuries and illnesses suffered by their employees, with compensation paid on a no-fault basis, generally according to a schedule of capitated damages. Platforms may wish to join the workers' comp system and enjoy immunity from claims of job-related torts. Current law would generally make that impossible unless the workers were classified as "employees." A flexible-worker status should open the door for firms to opt in to workers' comp voluntarily, free from the concern that they would automatically be regarded as employers.

Indeed, the entire system, with its costs and protections for job providers and workers alike, would be optional. Job providers could avail themselves of existing labor law and not contract with any flexible workers at all. In any case, nearly all platforms that did partner with flexible workers would continue to employ some full-time employees (technical, legal, and marketing staff, for example) and would have to follow all of the legal rules that apply to these workers. People who work for all platforms combined below a certain hour threshold (perhaps 20 hours in a month) would remain independent contractors regardless of who they worked for.

For both political and practical reasons, it's desirable that flexible workers have some of the protections and safety-net benefits afforded to those who currently work full time. This isn't entirely because the benefits are in enormous demand (although some workers certainly do want them). In fact, internal worker surveys conducted by transportation-network companies like Lyft and Uber show that drivers are much more interested in discounted or free oil changes and gas than they are in unemployment or even health benefits. Furthermore, some benefits that were once almost always work related, like pensions and health care, are now more personal than they were in the past.

Instead, there are three reasons for a flexible safety net that are both practical and political. The first, lack of interest in a safety net for gig-economy workers may partly reflect the difficulty of accessing many benefits. Right now, gig-economy workers have no access to public unemployment insurance, have no choice but to navigate an individual health-insurance market and its complicated exchanges, and rarely have private short-term disability coverage which, largely because of adverse selection bias, is vastly more expensive than similar coverage provided through employers. Second, even a minor setback of the kind that unemployment insurance and health insurance are supposed to cover can leave some gig workers reliant on safety-net programs like TANF, SNAP, and Medicaid, rather than remaining self-reliant like a regular salaried employee could during a temporary setback or short period of joblessness.

Third, the creation of some sort of safety net is probably necessary to get political buy-in for flexible-worker arrangements. Many Republicans and perhaps a few market-oriented Democrats might support an arrangement that simply creates a "third status," without any new conditions placed on employers besides the obligation to respect worker independence (and even that might be a hurdle in states without right-to-work laws). But state laws to facilitate this new status will still have to pass muster in the liberal states and very liberal cities where gig work is already the most common and in demand. Some sort of benefits platform is going to be a political necessity, even if it's possible (and perhaps likely) that a purely market-based benefits system could emerge without any government intervention. Both risk aversion and interest-group politics make it difficult to see how legislative action creating a new flexible-worker status could take place without providing some type of new safety net for such workers.

This idea has political traction already. In November of 2015, a number of prominent individuals, including the CEOs of large gig-economy companies like Lyft and Care.com, former SEIU leader Andy Stern, and former CBO director Douglas Holtz-Eakin, signed a joint open letter calling for a new type of benefits platform for gig-economy workers. Senator Mark Warner has also called for legislative action to create such a benefits system and will likely introduce specific proposals in the near future. Bills are also being drafted on the topic in at least three states and companies like Uber and Instacart have begun to line up behind them. Unless gig-economy platforms are squashed outright — as many cities and states are threatening to do — something will emerge to deal with these workers, and it would be best to ensure that it is as flexible and versatile as possible.

WORKER-CONTROLLED BENEFITS EXCHANGES

In particular, a "safety net" for gig-economy workers should consist of new structures called worker-controlled benefits exchanges (WCBEs) that combine some of the features of an SEP-IRA (the most versatile retirement-savings vehicle for the self-employed) and a broker for job-related benefits. Platforms that opt into the flexible-worker classification would finance these WCBEs through mandatory contributions, made on behalf of flexible workers. Mandatory-minimum levels of these contributions would be set according to worker earnings (perhaps with a certain maximum), and flexible workers and the self-employed could also make their own voluntary, pre-tax contributions.

WCBEs would serve as a partial, but more limited, replacement for the components of the basic safety net that most full-time workers get through a job: unemployment insurance-like coverage, paid leave, some aspects of workers' comp insurance (for gig-economy platforms not taking part in the workers' comp system), health insurance, and other benefits. They could assume a variety of legal structures: profit-seeking stock companies, taxable mutual enterprises, tax-exempt fraternal societies, or non-profits with elected officers. (The latter two would likely be required to maintain "common bonds" based on profession or belief.) Some might be explicitly chartered as insurance companies that directly underwrite some of the coverage they offer. A more common approach, at least at first, would be for the exchanges to place benefits with existing insurers and investment firms, some of which would likely craft discounted "affinity" programs specifically for the purpose of attracting business and providing customized benefits. The WCBEs would also serve as brokers to allow individuals to access small-group or individual health coverage and other benefits available elsewhere. The existing San Francisco-based firm Peers.org — which began as a nonprofit advocating for workers in the sharing economy and now serves largely as an online-benefits broker for gig-economy workers — provides one living example of what a freestanding WCBE might look like. Another comes in the form of the Independent Drivers Guild, a new entity affiliated with the International Association of Machinists and Aerospace Workers that represents Uber drivers in New York but isn't officially a union.

The benefits extended by or purchased through WCBEs would be worker controlled, in the sense that flexible workers would have an absolute right to determine which exchange or exchanges received mandatory-contribution dollars flowing from platforms. Platforms could not steer flexible workers to particular WCBEs, forbid workers from joining any particular WCBE, or have any ownership of them. Fully self-employed individuals could also contribute to WCBEs, but those holding full-time jobs and working for others generally could not.

The fundamentally new product provided through WCBEs would be a "replacement-income stream," a type of insurance that would replace unemployment insurance, sick leave, and some aspects of workers' compensation — the latter only for platforms that remain outside of the workers' comp system. Like unemployment insurance and most private disability insurance, the amount paid out by replacement-income streams would be pegged to a percentage of current earnings.

Unlike public unemployment insurance, replacement-income streams would be totally portable, entirely individualized, and wholly privately funded — although still regulated by the government as insurance products. Most likely, they would be regulated in a manner similar to the existing "credit insurance" typically used to pay mortgages, credit cards, and other loans for those who become involuntarily unemployed or temporarily disabled. To flexible workers, they would look something like the new product of "private supplemental unemployment insurance" that is already offered by companies like Assura Group of NY, Ltd. One would generally expect these insurance products to offer benefits for a shorter period of time than our public unemployment-insurance system, perhaps for eight weeks by default, rather than six months or more. Base-benefit levels would likely be quite modest to keep them affordable, and some WCBEs might well provide a minimal replacement-income benefit as part of basic membership. Flexible workers could add enhanced coverage if they chose. This would allow such products to be more flexible in structure than unemployment insurance, traditional workers' comp, and disability products. For example, if a platform that was contributing a sizeable share of a flexible worker's income ceased operations entirely or withdrew from a geographical area, replacement-income-stream policies would offer benefits to this worker. Benefits could also be claimed in the event of a documentable illness or injury that made a flexible worker temporarily unable to work. Replacement-income streams might also be tapped for "qualifying life events" like the birth of a child, possibly after a waiting period.

Assuming that the overall structure of the Affordable Care Act remains intact — by no means a certainty — WCBEs would also serve to allow flexible workers access to the small-group health-insurance market, and perhaps large-group or self-insured (ERISA) plans that larger exchanges could administer themselves, rather than the more costly and less flexible individual market. Ideally, WCBEs should also allow flexible workers to access catastrophic health-insurance policies that provide comprehensive coverage but require high deductibles (currently $6,850 per person) before paying most benefits. Changes would need to be made to the ACA to make this last option feasible since, under current law, for most Americans over age 30, catastrophic plans do not satisfy the law's individual mandate and do not qualify for subsidies. Even if such catastrophic plans aren't made available to all (and they should be), there's a strong argument to be made that flexible workers — who have made a voluntary decision to forgo other benefits of a full-time job — should be allowed to opt for these more limited health-insurance policies.

Flexible workers who don't see a need for additional benefits beyond whatever health coverage (if any) is mandated by law could opt to receive excess employer contributions as taxable income. (Flexible workers wishing to waive health coverage could, of course, do so and could use employer contributions to pay the fines currently in statute.) Some states might also choose to mandate the purchase of replacement-income streams as an alternative to unemployment insurance. Workers who want to purchase additional benefits could designate pre-tax dollars to buy more expensive health coverage, secure bigger replacement-income streams, finance forced savings to be used for "vacation" days, save for retirement (a topic Oren Litwin describes at length in a March 2016 paper for the R Street Institute), buy disability-insurance policies, or provide any other benefit that employers currently provide to their employees or that employees purchase with pre-tax dollars. Almost all of these benefits already exist in the individual market, of course, but if they achieved economies of scale, they would likely be more affordable and easier to administer through WCBEs.

The exact forms and leading functions of WCBEs would largely be determined by the workers themselves. Some would look much more like retirement plans that provide some guidance in navigating the insurance marketplace and would likely be run by insurance or investment firms. As WCBEs would be private entities, those that wanted to might engage in advocacy on behalf of their members' interests and might well come to look something like labor unions. (Vincent Vernuccio of the Mackinac Center has outlined some possibilities for what this could look like in his paper "Unionization for the 21st Century.") The newly recognized drivers' guild in New York, indeed, is already selling itself to drivers on the basis of the benefits they can now purchase. Since labor unions have been a first mover, then, many WCBEs are likely to look something like labor unions and might be run as subsidiaries of existing labor organizations. Peers.org, one prototype for a WCBE, started as an advocacy organization and continues some advocacy activities. But policymakers would be well-advised to let the exact forms of WCBEs evolve naturally, rather than trying to steer them into any particular model. In fact, as with any new program or policy, it would be advisable to place a sunset on any WCBE-enabling legislation to ensure that it's overhauled after a few years to correct any obvious flaws that emerge.

REVIVING THE SPIRIT OF OPPORTUNITY

The evidence of a growing, surging gig economy is scattered and limited. By most measures, the United States has not seen an explosion of gig work. A few successful firms may be disrupting certain industries, but there is no evidence that the nature of work has changed fundamentally. At most, high-profile firms like Lyft and TaskRabbit are only blunting long-term trends and policies that have put more people in full-time jobs, decreased workplace flexibility, and increased average job tenure. American workers today are more loyal to employers and work for larger businesses than at any time in the past for which we have data.

While some benefit from this state of affairs, many are left behind. The American spirit of enterprise, geographic mobility, and work opportunities for those at the bottom of the income ladder have been diminished, as a result of the overall trend away from gig work in most sectors of the economy. We need more gig-providing enterprises on the model of the high-profile internet-based businesses that have garnered so much attention in recent years, and we should have a legal structure to support them. America needs assertive public policies that will provide more opportunity and flexibility for workers and work providers alike.

Eli Lehrer is the president of the R Street Institute.