We are entering a Platform Economy—one in which tools and frameworks based on the power of the Internet will frame and channel our economic and social lives. The algorithmic revolution, in which computable algorithms are applied to activities from consumption and leisure to services and manufacturing, is the foundation of this digital transformation.1 Algorithms now live in the cloud and form the basis of digital “platforms.” For our purposes, platforms are “frameworks that permit collaborators—users, peers, providers—to undertake a range of activities, often creating de facto standards, and forming entire ecosystems for value creation and capture.”2
The cloud is at once infrastructure, marketplace, and ecosystem.3 The variety of platforms nearly defies categorization. To illustrate, Google and Facebook are digital platforms that provide search and social media, but also are platforms on which other platforms are built. Amazon is a marketplace, as are Etsy and eBay. Amazon Web Services provides infrastructure and tools with which others can build, while Airbnb and Uber are forcing deep change on quite different businesses. These diverse, cloud-residing platforms however we categorize them, are provoking a profound reorganization of markets, work arrangements, and, fundamentally, of value creation in the contemporary economy.
Our basic premise is that the emergence of a platform-based economic reorganization will not dictate our future, although it will and already is beginning to frame the choices we make.4 How we choose to deploy platform-based tools will reflect corporate strategy and public policy, and will condition the society we are building.
Knowledge to Shape the Future
Will the Platform Economy—and the reorganization of markets, enterprises, and social organization it portends—catalyze economic growth and a surge in productivity driven by a new generation of entrepreneurs? Or will the reorganization concentrate gains in the hands of those who generate the platforms, and possibly stifle future entrepreneurs? Will it spark a wave of entrepreneurial possibilities, or an avalanche of dispossessed workers trying to make their way with gigs and temporary contracts? Ultimately, what do we need to know and understand to shape this future?
How pervasive will the platform effect be? The word of the day is disruption—the sense that many traditional business models, organizations, and forms of value creation will be swept aside or radically transformed. Although control or ownership of platforms is a separate matter, groups of peers that coordinate activities and transactions on platforms challenge existing business models. For example:
- Taxi businesses are threatened by Uber.
- The music industry is challenged by iTunes, Pandora, and Spotify.
- The camera industry is disrupted by GoPro, which, through its website, is trying to organize itself as a platform firm, even as YouTube and app stores emerge.
We must consider the impact of platform strategies on competition in diverse segments of the economy, and what competitive strategies will be introduced. Long ago, online stores emerged in the retail sector. Consignment businesses have grown dramatically, and have opened the way for an array of newly minted entrepreneurs. And we know that many products have become parts of systems, the classic example being the iPhone, which provides access to a platform store that has a previously unimaginable variety of virtual products. The effects certainly will vary across sectors in very different ways.
While some will question whether productivity and growth will be accelerated, the more profound questions may be whether economic and social life will be transformed, and whether the outcome will lead to a very different distribution of wealth and power in global society.
The platform is likely to effectively define the digital era, with the algorithm, Internet and cloud as the building blocks. We contend that we are seeing the beginning of a digital transformation, which will extend to the Internet of Things and beyond, and is likely to release both enormous creativity and wicked management problems.
What will the Platform Economy do to work, entrepreneurship, income, and inequality? Platforms are creating new opportunities for earning income, and are generating an array of entrepreneurial opportunities … but of what sort? Will we have an increasing number of laundromat entrepreneur equivalents who are viable in their own right but unlikely to generate sustained productivity and growth, thus reflecting the patterns of growth rather than generating them? Will platforms induce a set of new businesses that, in their turn, drive employment and growth? Or, put differently, will the platform transformation create a community of incipient entrepreneurs or vulnerable workers?
Work is being reformatted. For many, traditional employment—a single organization providing long-term engagement, usually with some form of social benefits—is giving way to gig and contract arrangements. Not surprisingly, business strategies shape job quality. Even in low-margin, low-price businesses, there are “better” job strategies that can provide workers higher wages and benefits, and help strengthen a firm’s competitive position. In the aggregate, the shifting place and character of entrepreneurship, and the reorganization of work, may powerfully alter the distribution of wealth and income in societies.
Addressing the Elements of a Platform Economy
Assuming we are moving to a Platform Economy, we must address four elements. At first glance, each is evident. But a closer look reveals that what to achieve and how to achieve it in each case are not as evident.
1. Infrastructure: We obviously need appropriate infrastructure—but what kind? And how do we acquire or build it? Does it simply require ever-faster broadband access for the community as a whole? Or does effective infrastructure require tools and training for firms and consumers? If so, should they be provided as a public utility or by the market?
2. Training and skills: Does the Platform Economy require a new set of skills, or only a recasting of emphasis? Needs certainly include widespread comfort with using platforms and apps. But does that require a heavier investment in STEM (science, technology, engineering and manufacturing), or in design and art? The answer is less obvious than it might appear.
3. Social protections: Will entrepreneurs and contract work arrangements be facilitated and encouraged by broadening social protections? Or will those protections inhibit the flexibility required of the economy?
- 4. Regulatory transitions: Rules and regulations for the marketplace and labor markets will not adapt themselves to the needs and logic of the Platform Economy. Moreover, this transition will lead to debates and fights about the adaptation of those rules. There will be struggles about protections for communities, clients, workers, and the market itself.
- As communities, the evident instance is whether Airbnb represents a change in land use—do I want a virtual hotel in my neighborhood? And should Airbnb hosts be able to discriminate against people they do not want to welcome, whereas a hotel is legally proscribed from discriminating?
As clients, do we want assurance of health and safety in our Uber cars or Airbnb rentals?
As workers, what risks should we bear, and what risks should be shouldered by the platform owner? Indeed, the matter of when workers are contractors and when they are employees is being adjudicated now in a number of states and nations.
As the market itself, what protections and policies are required to preserve competition? The European Union’s efforts to regulate Apple, Google, Facebook, and other globally dominant, Silicon Valley-domiciled platform leaders is an example.5 In fact, what are the competition issues when dissatisfaction with Google Search, for example, can lead to a costless, seamless, immediate switch to Microsoft Bing or the independent DuckDuckGo?
The Question of U.S. Leadership
Finally, we should consider whether U.S. entrepreneurs or, more specifically, Silicon Valley or Silicon Valley-inspired entrepreneurs, will retain their early and currently obvious dominance in a platform era. Throughout the digital era and this current wave of change, American policy initiatives and firms have led the way. Silicon Valley’s enormous success with finding and funding new products, firms, and disruptions makes it seem that the digital transformation is an American prerogative, and that others can only follow and whine.
While generations of digital change have issued forth from Silicon Valley, inevitability is truth only until things change. Henry Ford, we are often told, launched the era of mass production that represented American innovation and contributed to an era of American predominance. Indeed, policymakers and scholars alike referred to Fordism as an economic revolution.
Seventy years later, Toyota arrived, and American advantages became obstacles. As Cohen and Zysman observed in 1988, the U.S. faced decades of manufacturing decline. Comebacks were slow, and our manufacturing base and domestic skill sets eroded.6 Silicon Valley breakthroughs did not help the industrial heartland, as Florida and Kenney argued in 1990.7 Can this happen again? Let us not too quickly point gleefully to the mistaken overenthusiasm about Japan’s consumer-electronics and semiconductor successes, which faded as Silicon Valley-driven U.S. entrepreneurial firms reasserted themselves through success in software-driven electronics. Rather than dismiss the question, we should ask what might happen if a surge in platform leadership emerged from outside the United States.
Two possibilities establish the reality of this question:
- The huge size of the Chinese market suggests that platforms bred in that largely protected market could have the scale and financial sustainability to move abroad and establish standards. With the possible exception of Alibaba, there currently is little evidence that Chinese platforms are gaining traction, even in East Asia. But will this change?
- German manufacturing firms are concerned about maintaining position in what is labeled Industrie 4.0, which is what they term the Internet of Things. They want to assure that their small, midsized, and major manufacturers capture the advantages of platforms and next-generation cloud-based computing, and define that trajectory globally.
Asked generally, will European firms’ manufacturing, industrial and service strengths— as well as the Chinese operating behind protectionist barriers—create application domains that can outcompete U.S. entrepreneurial firms?
For now, as a beginning, we return to our basic premise. The Platform Economy is upon us. What it means for our economy and society will be a choice, not an inevitable unfolding of the technology.
How we label this transformation matters. Labels specify targets for policies, strategies, and studies.
The new, digitally founded economy has been given a variety of names based on its perceived attributes. Each name lends itself to the study of different outcomes and activities over others. At various times, it has been called the creative economy8 and, by contrast, the gig economy/the precariat/1099 economy,9 focusing on the impact this emerging economy is having on work.
Recently, “Sharing Economy” has been a popular label10. However, much activity labeled sharing—such as Uber and Airbnb—is far from the stated visions of:
- Wikipedia—the shared construction of a knowledge tool.
- Napster—sharing music, legally or not.
- Open-source software—shared among all.
Despite the attractive label and entrepreneurial successes, there are drawbacks to calling this a Sharing Economy. Uber and Airbnb are entrepreneurial initiatives that facilitate the conversion of consumption goods, such as automobiles and apartments, into commercial offerings. Airbnb is closer to nineteenth-century, working-class women taking in boarders. Uber facilitates gig work for drivers providing their personal vehicles, or buying vehicles to join the game.
This ”sharing” can and often does resemble a putting-out economy from the early industrialization period before factories, but the putting out of work to individuals now is facilitated by digital platforms.
Simultaneously, the rapidly growing mobile-app and user-generated-content firms, such as YouTube or Instagram, are structured as digital consignment industries, borrowing from the compensation scheme used by artists working for galleries, for example. Airbnb also has some consignment aspects.
The “Platform Economy” is our preferred term for this emerging organizational model. Why? Digital platforms are the base upon which an increasing number of connection-based activities—marketplace, social, and political—are organized. If the Industrial Revolution was organized around the factory, today’s changes are organized around platforms that apply algorithms to enormous databases running in the cloud. The salience of these digital platforms suggests that we are in the midst of an economic reorganization in which platform firms are developing power roughly equivalent to that of Ford, General Motors, and General Electric during earlier eras.
Debating the Future
There is a ferocious debate about whether the future of the Platform Economy will be a utopia of abundance, or a dystopia of limited employment and stunning inequality.
The early pioneers in the industry—particularly Bob Noyce, Steve Jobs, and Bill Gates on the West Coast—truly believed they were creating the future, and opening the world’s possibilities and prospects.11 And optimists still abound: San Francisco, for example, now has been called the new Hollywood, as visions of profitable disruptions mobilize entrepreneurs and data scientists.12
The question for investors, inherently optimists who search for profitable activities, is: How is value created and captured in the platform era? To answer, many point to the benefits of emerging platforms. Zipcar, by reducing the need for individual auto ownership —or at least potentially increasing access to auto transport for those who do not own vehicles—stands as a commercial vision of sharing, although it actually is a particular form of rental,13 alongside true sharing activities, such as Wikipedia.
However, as already noted, the optimistic, utopian vision of contract workers as simply proto-entrepreneurs who treasure their flexible schedules often collides with claims that they simply are a new precariat—dependent contractors in precarious roles within a modern putting out. Similarly, the utopian vision argues that platforms such as Uber and Lyft can unlock the commercial value in underused personal goods—apartment rooms or cars can become investment goods in commercial markets without consequences for the community. This is a large assumption. Similarly, the platform businesses that match workers and tasks may make labor markets more efficient but, at the same time, can generate fragmented work schedules, and increasing levels of part-time work without employment-related benefits that previously characterized much full-time work.
Will, then, the tools we have built turn on society? Even as the digital era unfolded in its utopian phase, there were skeptics, and perhaps most prescient was novelist Kurt Vonnegut. In his first novel, Player Piano,14 Vonnegut envisioned a digital future of abundance—albeit a digital future of machines built with tubes rather than semiconductors—with radical social division between a creatively employed and highly credentialed elite, and an underclass. His dystopian vision now is finding full expression in the fear that digital machines, artificial intelligence, robots, and the like will displace work for a vast swath of the population.
Bill Davidow, once at Intel and then at his own Silicon Valley venture capital company, expressed this in the business literature, “What Happens to Society When Robots Replace Workers?”15 An outpouring of popular books and more formal articles from the economics profession argue that jobs will be displaced by digital automation and robotics. The best known and most popular economic expression of this belief is The Second Machine Age.16
The question really is about what balance will exist between jobs created as the digital wave flows through our economy and society, and what jobs will be displaced. It is feasible to catalogue existing work, particularly routine work that can be fully characterized, as likely dislocated by digital tools, and perhaps estimate the numbers of such existing jobs that may become obsolete.17 More difficult to catalogue, and open to speculation, are new kinds of work being created now and that will be created in the future. Some early indicators can be enumerated, but certainly not exhaustively counted. Algorithms and databases are automating work but, as this occurs, new work is being created. There will be new products—goods and services, as well new production processes—that are likely to require intensive design, creativity, and technology.
Moreover, the character of some existing work—much or little we cannot know—will be reframed, but not eliminated by digital technology. Uber, TaskRabbit, Handy, and other platform firms are transforming industries by connecting "workers" with customers in new ways. In some cases, this is displacing or threatening existing, often regulated, service providers, such as taxis and hotels. In other cases, it is formalizing previously less organized or locally organized work. Still other platforms, such as app stores and YouTube, are creating entirely new occupations or occupational branches. Finally, existing organizations are creating digital and social media marketing departments and jobs. The question in these cases is not whether there will be jobs, but what system of control and value capture is in place. Having written this, our sense is that, across the board, "employment" appears to be more precarious than ever.
These changes are not likely to result in the "workerless" society. Rather, we risk a society within which the preponderance of the work and value creation is more dispersed than ever before, even as platform owners centralize transactions and capture value from activities on their platforms. Importantly, we can only speculate about the balance and character of firms and jobs destroyed, created, and transformed, and about the character of the work and organizations generated.18
Indeed, we would note that there is a classic dilemma in the use of digital automation: anything that can be characterized sufficiently to become computable can be copied.19 At that point, another round of innovation and imagination is required.20
Can automation innovate itself? Or will teams of people and digital tools be required to be competitive? Note that the Turing Test might establish that a digital machine can imitate intelligence. But the test does not establish, or purport to establish, consciousness, nor consider whether human consciousness differs in fundamental ways from current algorithmic tools.21 The debate about jobs created or destroyed cannot be resolved. Importantly, we can only speculate about the balance and character of firms and jobs destroyed, created, and transformed.22 We can, we emphasize, only examine indicators and traces.
We are confident, though, that the outcomes—jobs created, and jobs evaporated or transformed—will depend not on the technology itself but on how the technology is deployed. Choices about deployment will turn on entrepreneurial initiative, corporate strategies, and public policies. We know, for example, that the consequences of deploying radio-frequency identification in retail differed dramatically in Denmark, France, and the United States. The outcomes, interestingly, were not a product of labor-management fights, but of conflicts between producers and distributors.23
Similarly, in the discussion of the Internet of Things or the digitally based reorganization of manufacturing, we find significant differences among national emphasis and investments.24 Which communities, this leads us to ask, are most likely to be the sources and beneficiaries of the emerging Platform Economy? Which are most likely to be discomfited? The strategies for deployment are, of course, precisely the substance of our choices for a future in the Platform Economy.
The Algorithmic Revolution25 and Clouds:26
Technical Foundations of the Platform Economy
The algorithmic revolution and cloud computing are the foundations of the Platform Economy. Computing power in itself is only the beginning of the story. Computing power is converted into economic tools by algorithms operating on raw data. When aspects of activities can be converted into processes, which can be formalized and codified with clearly defined rules for their execution, they can be reduced to computable algorithms.
The software layer that stretches across and is interwoven with the economy is a fabric of algorithms.27 That software layer—that algorithmic fabric—encompasses manufacturing. It is the Internet of Things/Internet of Everything/Industrial Internet, with its implied webs of sensor networks. This includes services, which often employ those sensor networks, and covers other diverse social, political and economic activities. The software layer extends the availability and lowers the cost of access to digital tools, and to traditional tools accessed by and controlled by digital processes. Costs sometimes drop through open-source software—the race to zero in cloud computing—and/or by the ability to collectively provide tools through online platforms or physical commercial sites, such as TechShops.
Cloud computing is about how computing is done. It is much less about geography, or where it is done.28 It rests on virtualization and abstraction of computing processes.31 While the details of how it works do not matter to our discussion, the consequences do.
We should note that the major providers of cloud services remain, at least for now, large American firms that developed the cloud paradigms and systems for their own internal use. For provision, scale does matter. For users—individuals, small- and medium-sized enterprises, startups, and corporations alike—the consequence is a radical reduction in the cost of computing resources, and information- and communication-technology tools. As important, as is now widely recognized, the terms of access to computing resources change, as well. Users can “rent” resources in units rather than having to own or build entire computing systems. Computing, and the applications and platforms it facilitates, are then available as an operating expense rather than a capital expense.
Let us link the story of algorithms and cloud computing to the emerging Platform Economy.
Algorithms live in the cloud and dramatically ease the creation of platforms.29 Digital platforms are, then, computing frameworks upon which users can undertake a range of activities, often forming entire ecosystems for value creation. Many of the current Internet platform firms use Amazon Web Services. So, indeed, platforms can grow on platforms as an array of applications.
Many of those platforms-on-platforms are what we would call complementors. Complementors include emerging actors such as:
- AppAnnie, which ranks revenue generated by apps.
- Advertising “agencies,” which analyze YouTube advert buying.
- TubeMogul, which classifies YouTube “stars” and measures their reach.
- Various agencies that cultivate new YouTubers.
These complementors are powerful allies in building and maintaining the lock-in for the master platform. Platforms are, to put this differently, algorithm-enabled cyberplaces where constituents—people or machines—can act or transact. Of course, building a platform is work, but platforms themselves then generate or organize the work of others by providing digital locations for connections that organize work and other activities.
Making Sense of the Diversity of Platforms in the Platform Economy
Digital platforms, based on algorithms and databases, are restructuring ever-greater parts of the global economy. In many cases, they have disrupted the existing organization of economic activity by resetting entry barriers, changing the logic of value creation and value capture, repackaging work, and/or repositioning power in the economic and network system.
Speculations aside, we currently have no real theory of the effect of these diverse platforms on the overall economy, nor a particularly dominant approach to categorizing platforms. The intent here is to begin to structure a discussion about how an economy increasingly organized to operate on platforms affects the organization and practice of competition, work, and entrepreneurship.
Traditionally, we have categorized by sector, but sectors are blurring. A smartphone is at once a communications device, a camera, and a music system, to start a list. We might ask whether the consequences of platforms differ by the original sector being transformed. But, ultimately, increasingly blurred sectors are not good organizing categories for understanding the Platform Economy.
Perhaps we should categorize platforms by function or business model. Here again, categories blur and overlap. As a result, an initial listing is partial and choppy, an awkward cut into the complexity.
- Platforms for platforms. In a sense, the Internet itself is the foundation of the Platform Economy, but there are a series of businesses that then provide the infrastructure and tools for the rest. For example, Amazon Web Services facilitates the construction of cloud services, the tools with which other platforms are built.
- Platforms mediating work. In some platforms, this resembles the function of electronic headhunters or human resource departments. In other cases, it can be seen as a modern form of the putting out system of nineteenth-century industrialization. Mediating work itself has many versions, which include:
a. Globally biddable work. Examples include Odesk/eLance (since renamed Upwork), Innocentives, and Amazon Mechanical Turk.
b. Occasional informal work. Facilitated by apps and, therefore, cyberformalized, Task Rabbit, Handy, and Homejoy are perfect examples.
- Platforms making tools become available online. Github is becoming the repository of all kinds of open-source software available to anyone. This dramatically reduces the cost of software tools and building blocks.
a. Automated HR, Job Rooster, and Wonolo provide diverse HR functions.
b. Zenefits provides an online marketplace of HR tools free to small businesses. In the process, it makes mediation by local insurance brokers unnecessary.
- Electronic goods markets for retail and business. These run an entire gamut.
a. Virtual markets for physical goods. Etsy and eBay are distinct versions.
b. Retail sales platforms, such as Amazon or hundreds of company-specific apps.
c. Apple and Android stores, which are platforms that facilitate diffusion of the apps.
d. Virtual consignment platforms, such as YouTube, Amazon self-publishing, and many others.
- Platforms intending to transform service industries. Airbnb and Uber are examples of platforms intending to convert consumer goods into investment goods. For example, rather than sharing, Uber connects drivers with customers algorithmically. In this case, drivers are treated as contractors, which puts them in a more precarious position.
- Shifting the place of intermediaries in finance.
a. Platforms, such as Kickstarter or Indiegogo for project funding, can replace traditional intermediaries.
b. Finance platforms displace traditional financial institutions, e.g., AngelsList for venture capital, or Zopa or Rate Setter for peer-to-peer lending support.
- Facilitating social and political organizations, including worker organizations.
Does beginning with the platform categories above provide insight into the character and ability to scale the entrepreneurial opportunity? Diverse as they are, all these platforms have created business ecosystems, repackaged work relationships, and/or transformed terms of competition.
These platforms represent a multiplicity of business models and functions, and raise questions that suggest wildly varied answers. As a place to start a discussion, we might ask these questions about each platform, or type of platform.
- How is value created? The Platform Economy itself is a distinctly new set of economic relations resting on the Internet. The ecosystem created by each platform is a source of value, and sets the terms on which owners and platform users can participate.
- Who captures the value? There are a variety of mechanisms with varied implications for gains distribution.
- Owners of some platforms “tax” all transactions, while others monetize their services through advertising.
- Contractors, consigners, or quid pro quo workers do work previously reserved for traditional employees, or take on entirely new categories of work created by platforms.
- Venture laborers, so named by Gina Neff, earn high wages working in platform firms. More significantly, if the firm is successful, the value of the platform is capitalized in the stock markets, resulting in remarkable wealth for the firm’s direct employees and entrepreneurs.30 If the firm falters or fails, these individuals must find new employment.
- Mini-entrepreneurs or consignment workers—alternately positive and negative terms for the same individuals—provide platforms such as app stores, YouTube or Amazon self-publishing with goods that are, usually, but not necessarily, “virtual.” Many will be unsuccessful or marginally profitable, but some can be fabulously successful. While as yet unmeasured, it seems likely that this is creating many more opportunities for entrepreneurship. In certain cases, particularly in apps, products in the consignment economy may become so successful that venture capitalists will invest in the entrepreneur/firm, and the employees will become venture labor. Some of these apps can become platforms themselves. Put differently, the consignment level has significant upside for participants, but the upside is accompanied by high risk.
- Who owns or controls the platform? The answer differs and makes a difference. For users and producers, for example, there are differences between Wikipedia (a network managed by consensus rules), the Danish agricultural cooperative platform (a consortium in which participant owners know each other, and set clear boundaries between themselves and others) and Uber (a tightly held firm funded by venture capitalists, whose value eventually will be capitalized by sale of control, either through acquisition or stock offering). Power may be centralized (controlled transaction and communication systems separate buyers/users from providers, and separate providers from buyers), or decentralized (such as how Wikipedia diffuses power over content).
- How is work packaged and value created? Some workers—including those employed by Microsoft, Google, LinkedIn, and Facebook—retain traditional employment relationships. These firms expect long but relatively flexible working hours, and offer free food, drinks, transportation, and a myriad of other benefits, which make them seem almost like corporate paradises. Those working on gigs, consignments, or contracts have radically different relations, though the hours are flexible and largely self-controlled.31 One important question, then, is what percentage of work now is organized in these radically new ways?
- What is the distribution of risks and rewards for those in these various ecosystems?
The Consequences of Platforms for Entrepreneurship and Work
Entrepreneurship and the packaging of work are tightly interwoven issues.
Consider entrepreneurship. Media attention and much talk in the venture community focuses on “disruptions,” which appear to be where massive opportunities exist. Uber disrupts taxi companies. Airbnb challenges hotels. Zenefits threatens local insurance brokers.
But how many instances of disruption are there? Do these disruptions create a flood of viable entrepreneurial possibilities, or destroy the security of employment relations? Do they operate to create new sources of income and reasonably compensated work throughout the society?
It is evident that platforms open an array of entrepreneurial opportunities. Some entrepreneurs, such as Zipcar’s Robin Chase, envisioned not only an alternative economic model, but also an alternative social model: Own a car, or access occasional use of one through the Zipcar platform. If that model spreads widely, it would result in a drop in overall demand for auto production. This may or may not disrupt Hertz (Zipcar was sold to Avis).32 But could it dramatically affect automakers if fewer persons buy automobiles.33 In other words, such “sharing” solutions could have unforeseen ripple effects on entire market ecosystems, as encyclopedia producers discovered to their dismay.
Many platforms, by their very nature, prove to be winner-take-all models in which the owners of only one or two surviving platforms appropriate a portion of the entire value created by all the platform participants. More importantly, power is centralized to the platform owner who, after winning the initial competition, becomes a monopolist who can make decisions to maximize his/her own welfare. At the same time, the monopolist platform owner squeezes the platform community—the drivers on Uber, the content providers, the consigners—who are instrumental in producing the value in the first place.
Consider, by contrast, how platforms affect work. In these business models, what happens to the organizational forms of work, and to the form in which work is packaged?
Conceptually, if not literally, Uber converts taxi company employees or former medallion owners into contractors who access income through the Uber platform. Are these contractors, mini-entrepreneurs, or extremely precarious workers relabeled as contractors? Are some Airbnb offerings not just another form of rentals that transform apartments from long-term residences to short-stay offerings?
Is this entrepreneurship in any significant way? Moreover, how do we understand the business models of the winners among those who produce apps or YouTube videos, or self-publish books on Amazon? For these individuals, there is a law of returns—a few big winners are remunerated by advertising, product placement fees, and personal appearances. And there is a very long tail of producers who are creating the vast bulk of consigned content without compensation.
Considering disruptions, mini-entrepreneurs, contractors, and gig workers leads us to ask: Does the Platform Economy point to an even more unequal society? Does the answer depend on the character of platforms, or on the policies and politics of the Platform Economy?
Policy and Politics
The policy objectives and concerns in a platform era seem evident. The late nineteenth century saw the emergence of the corporate organization as a means of orchestrating economic activity and organizing markets.34 In the twenty-first century, we might speculate, the platform in the cloud takes on a variety of these functions.
Hence take Google, the Platform Economy giant. It is now a large firm, and yet has only 50,000 employees. Uber has only about 1,500 employees, and already is a global business. What policy and political issues emerge when the orchestrators of economic activity are relatively small firms rather than organizations as large as Ford Motor Company, General Electric or, the behemoth of them all, General Motors were in their heyday?
From the American standpoint, we must ask how policy will influence entrepreneurship and work in a cloud-founded Platform Economy. Let us set aside, in this brief essay essential questions about whether cloud technologies, and the platform-driven economic reorganization they prompt, will drive productivity growth, and whether the reorganization, on balance, will destroy jobs or reduce the skill levels required.
Instead, we ask:
- On balance, are the array of entrepreneurial, innovative opportunities widespread? Or will they hover around a few big winners, and an array of small–scale, highly vulnerable players?
- Do we create a new source of productivity or a new form of putting out?
- Can Uber drivers be self-supporting contractors in a 1099 economy, rather than stable workers in an employment economy? Or are they extremely vulnerable gig workers?
- Do we generate labor market flexibility or a “precariat,” as some believe, that resembles a cyberized Downton Abbey, replete with a new and sizeable underclass.
The policies we adopt may determine the balances achieved in the platform era. If we want an entrepreneurial spirit to infuse the platform world, then we want risk-taking entrepreneurs who form platforms or seek advantage as contractors/consigners within it. What encourages risk? Is it a certainty that, if a gamble fails, one can always play again?
Similarly, if we want workers to accept the new flexibility, how do we assure them that, if they accept the flexibility, they will be beneficiaries, not victims, of the greater social value and wealth that is being created? As victims, they will resist. As beneficiaries, they may help facilitate the shift. Stated simply, public policies will shape the gains, risks, and responsibilities both for work and entrepreneurship.
How then can we make the Platform Economy a vibrant source of growth? We view two public policy domains as critical.
First, social policy, sometimes called welfare, shapes the risks workers and entrepreneurs take, and affects their evaluation of whether to pursue or resist change. In the United States, benefits such as pensions and medical coverage, until the emergence of ObamaCare, have been tightly tied to employment. Lose your employment, lose the protections. The American debate often assumes that expanded welfare protections mute initiative,35 pointing to Europe as investing in social protections at the cost of economic dynamism. Regardless of whether this was, in fact, ever the case, the question is whether social protection will inherently mute initiative now. The real issue was never the protections themselves, but how they were organized. The Danish flexible security model provides evidence that social protections can lubricate the engines of change.
The Platform Economy, with expanding contract work and gig employment, should lead us to look again at the Nordic social policy model. Simply put, it attaches many social benefits to citizenship rights. The notion of flexible security gives employers extensive rights to adjust their work forces as needed while still providing workers social securities, or protections, in the form of training, job placement, and basic income. This is no panacea. Struggles to sustain employment endure, and there are continuing fights about these programs. But we must consider that, in this environment, addressing the downside risk of entrepreneurial efforts, while providing worker flexibility with broader social safety nets, are social rights that may make a Platform Economy a source of sustainable growth.
What assurances of social safety do we want to provide to risk takers, to encourage those risks? Could Uber get drivers in an environment with a greater social safety net and a reasonable basic standard of living guaranteed? Evidence from Europe suggests an emphatic “yes,” as governments are forced to control Uber drivers.
Second, what market rules are appropriate for a Platform Economy? There will be an array of political struggles about these rules, and we should welcome them as part of defining the market and society in a cloud/platform era. There will be political fights about how to protect communities, clients, and workers against market disruptions. Some of those fights will be about business models playing a game of policy arbitrage, while others may be about rules on consignment platforms. In this game, the advantage of platform-based companies often rests on an arbitrage between the practices they adopt versus established companies’ operating rules, which are intended to protect clients, communities, workers, and markets themselves.
These are classic issues, but it is worth working through examples and formally laying out the problem. Taxi companies pay insurance that protects clients and other drivers should their employees have accidents. What insurance protections should be required for Uber or Lyft, and who should pay for them? Airbnb effectively ignores, in many cases, land-use rules intended to preserve particular community values. Should that be permitted? A taxi driver is legally required to pick up anyone hailing the cab in any part of town, but an Uber driver can refuse. A hotel must lodge a boarder regardless of ethnicity, and cannot decide that a person’s religion or ethnicity is undesirable. Should the same apply to Airbnb hosts? With these examples in mind, policy arbitrage is not the ideal basis of a new competitive business model.
The list of policy domains that will have to be rejiggered in a platform era must include competition policy, taxation rules, service provision requirements, and intellectual property rules. We may want to review existing public policies, recognizing that platform entrants playing the arbitrage game likely will press for policy revisions.
Importantly, it is not possible, or even appropriate, to simply declare that old rules and values will apply in this new era. For one thing, it is not always possible to transport the values of one era to the next. Clearly, platforms raise new issues about market dominance and the ability to extend position from one market into another. We only have to look to Europe’s struggles with Google, and look back at the battles around Microsoft. As seemingly settled fights in all these policy domains are reopened, issues will be refought, and new outcomes can be expected.
Conclusion: Takeaways to Consider
Technologies—the cloud, big data, algorithms, and platforms—will not in themselves dictate our future. Rather, the future is ours to choose. How we deploy the technologies, and the rules set for their deployment and use, will be critical. When we look at cases such as electric utility grids. call centers or radio-frequency identification in retail, we find that the market and social outcomes of new technologies vary across countries. Of course, a chosen technology path frames choices.36 Larry Lessig, what seems like long ago, wrote that Code is Law37, and code is increasingly West Coast law. If not solely the technology, what then explains Platform Economy variances between and within countries?
Deployments differ with corporate strategies and public policies. Consider corporate strategies. Will companies view workers as costs to be contained, or as assets to be developed and promoted, even in an era of algorithms, robots and automation?38 And, as important, will worker assets be directly tied to the firm, and what entity should bear the costs of their conservation and improvement? In other words, the old question of the boundaries of the firm are being reposed.
Consider policies. What balance do we seek between flexibility/adaptability and social protection? Do we recognize that flexibility can come with protections against risk?
The consequence is that we will be making choices in an inherently fluid environment shaped, to some degree, by unpredictable technical changes, and by social reaction to these changes. Ultimately, all of this will depend on how we believe markets should be structured, what we socially value, and how we will channel the enormous value that these socio-technical changes are creating.
About the Authors and Acknowledgements
Martin Kenney is professor of Community and Regional Development at the University of California, Davis, and a member of the Berkeley Roundtable on the International Economy.
John Zysman is codirector of the Berkeley Roundtable on the International Economy, and also is a professor, Political Science, at the University of California, Berkeley.
Each author contributed equally to the formulation and development of the ideas in this paper. They extend thanks to Ruth Collier, Lilly Irani, Bryan Pon, and Anne Visser for their comments about earlier work, and their contributions to the discussion from which this paper emerged.
- Zysman, J. 2006. “The algorithmic revolution — the fourth service transformation,” Communications of the ACM. 49 (7).
- Direct wording is borrowed and slightly modified from Mattila, J., and T. Seppälä. 2015. “Machines in a Cloud – or a Cloud in Machines? Emerging New Trends of the Digital Platforms in Industry and Society”. The foundational book describing platforms and their use is Gawer, A., and M. A. Cusumano. 2002. Platform Leadership: How Intel, Microsoft, and Cisco Drive Industry Innovation (Boston, Harvard Business School Press). Also, see the more recent Gawer, A., and M. A. Cusumano. 2014. “Industry Platforms and Ecosystem Innovation,” Journal of Product Innovation Management 31(3), 417-433.
- Kushida, K. E., J. Murray, and J. Zysman. 2012. "The Gathering Storm: Analyzing the Cloud Computing Ecosystem and Implications for Public Policy," Communications and Strategies 85:63-85.
- For a theoretical conceptualization of platforms as private market regulators, see Boudreau, K. J., and A. Hagiu. 2008. Platform rules: Multi-sided platforms as regulators. available at SSRN 1269966.
- For an early discussion of the issues, see Ballon, P., and E. Van Heesvelde. 2001. “ICT platforms and regulatory concerns in Europe,” Telecommunications Policy, 35(8), 702-714.
- Cohen, S., and J. Zysman. 1988. Manufacturing Matters: The Myth of the Post-Industrial Economy (New York: Basic Books).
- Florida, R., and M. Kenney. 1990. The Breakthrough Illusion: Corporate America’s Failure to Move from Innovation to Mass Production (New York: Basic Books).
- Florida, R. 2002. The Rise of the Creative Class (New York: Basic Books).
- Friedman, G. 2014. “Workers without employers: Shadow corporations and the rise of the gig economy,” Review of Keynesian Economics, 2(2): 171-188.; Standing, G. 2011. The Precariat: The New Dangerous Class (London: A&C Black).
- Yochai Benkler is an early proponent of this framing. See, for example, Benkler, Y. 2006. The Wealth of Networks: How Social Production Transforms Markets and Freedom (New Haven: Yale University Press). For a critique of Benkler, see Dijck, J. V. 2013. The Culture of Connectivity: A Critical History of Social Media (Oxford: Oxford University Press).
- Robert Noyce founded Intel, Steve Jobs established Apple, and Bill Gates founded Microsoft. Others in that early cohort, particularly the semiconductor industry folks who were driving the revolution, included Jerry Sanders of AMD, Charlie Sporck of National Semiconductor, and Bill Davidow, who re-emerged as a pessimistic commentator on our digital futures following a career at Intel and in venture capital.
- Gapper, J. 2015. “Silicon Valley has become a dream factory,” Financial Times, May.
- Unsurprisingly, Zipcar was purchased by Avis in 2013. It is important to note that Zipcar is an asset-heavy business model, as it needs to own the vehicles.
- Vonnegut, K. 1952. Player Piano (New York: Charles Scribner).
- Davidow, W. H., and M. S. Malone. 2014. “What Happens to Society When Robots Replace Workers?” Harvard Business Review.
- Brynjolfsson, E., and A. McAfee. 2014. The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies (New York: Norton).
- See Frey, C. B., and M. A. Osborne. 2013. “The Future of Employment: How Susceptible Are Jobs to Computerization?,” September 17, 2013.
- From our own research, we would propose that today’s home-run firms create far fewer jobs than did GE, GM, or Ford in their era. And yet, today’s firm exert just as much organizing influence over the social world and over our thinking about the social world. Google employs fewer folks than Microsoft does, and maybe fewer than Microsoft did at a similar stage in its life. However, Google is far more omnipresent than Microsoft ever was.
- This point was introduced into our debates by Niels Christian Nielsen’s persistent arguments. See, for example, Nielsen, N. C., J. Murray, and J. Zysman. 2013. The Algorithmic Revolution and Empowered Human Value Creation (Danish Technology Institute Rosenthals/Book PartnerMedia).
- Turing, A.M. 1950. “Computing machinery and intelligence,” Mind, 59, 433-460. See also John Searle’s Chinese Room discussion. “John Searle in his construction of the Chinese Room problem distinguishes likewise between the imitation of consciousness, in this case capacity to understand and use a language, and the ability to imitate and translate without understanding.” For a simple depiction of the argument: John Searle, pages 21-24 (Continuum Books, Joshua Rust, 2009).
- This often will require case study research, and what Richard Nelson has termed “appreciative theorizing.” Appreciative theorizing is qualitative in nature as opposed to the formal modeling and quantitatively derived theorizing of economics.
- For example, Watson, B. C. 2013. “Platforms, Productivity, and Politics: Comparative Retail Services in a Digital Age,” The Third Globalization? Can Wealthy Nations Stay Rich in the Twenty-first Century?.; Breznitz, D., and J. Zysman, editors. 2013. (Oxford University Press). On call centers, Batt, R. 2002. “Managing customer services: Human resource practices, quit rates, and sales growth,” Academy of Management Journal, 45(3), 587-597.
- The German studies Industrie 4.0 and Smart Services Welt put a distinct emphasis on existing German strengths and how to preserve them. They are a very different flavor, in our view, from the American discussions.
- Zysman, J. 2006. “The algorithmic revolution — the fourth service transformation,” Communications of the ACM. 49 (7).
- Jonathan Murray, Kenji Kushida, and Patrick Scaglia have been essential to our understanding of these issues. See, for example, Kushida, K. E., J. Murray, and J. Zysman. 2011. "Diffusing the Fog: Cloud Computing and Public Policy." Journal of Industry, Competition and Trade 11(3):209-237.; Kushida, K. E., J. Murray, and J. Zysman. 2012. "The Gathering Storm: Analyzing the Cloud Computing Ecosystem and Implications for Public Policy." Communications and Strategies 85: 63-85.; Kushida, K. E., J. Murray, and J. Zysman. 2013. "Clouducopia: Into the Era of Abundance," CLSA Blue Book January.; Kushida, K. E., J. Murray, and J. Zysman. 2014. “The Next Epoch in Cloud Computing: Implications for Integrated Research and Innovation Strategy (BRIE Working Paper, 2014-4).; Kushida, K. E., J. Murray, and J. Zysman. 2015. "Cloud Computing: From Scarcity to Abundance," Journal of Industry, Competition and Trade 15(1): 5-19.
- Hatch, M. 2014. The Maker Movement Manifesto: Rules for Innovation in the New World of Crafters, Hackers, and Tinkerers (New York: McGraw Hill).
- Geography is not completely irrelevant. For many functions, speed matters; consider high-speed trading activities. Even search benefits from fast responses.
- It is important to note that fundamental insights regarding the importance of platforms come from work by Michael Cusumano and Annabelle Gawer that drew upon the history of Microsoft, Intel, and Cisco. Gawer, A., and M. A. Cusumano. 2002. Platform Leadership: How Intel, Microsoft, and Cisco drive industry innovation (Boston: Harvard Business School Press). There is now a large and fascinating body of literature about how firms should develop their platforms.
- Neff, G., 2012. Venture Labor: Work and the burden of risk in innovative industries (Cambridge: MIT Press).
- There are exceptions, as Uber and Lyft have quite stringent control over their service providers. Ultimately, the courts may decide that their drivers are, in fact, employees, thus challenging their current model. See, for example, the court ruling in San Francisco that a lawsuit about whether Uber drivers are employees could be tried by a jury. The lawsuit’s outcome will significantly affect the status of labor in a number of these firms. Wilson, M. 2015. “Juries to Decide Whether Uber, Lyft Drivers Are 'Employees” (Findlaw, March 12, 2015).
- Of interest is what happens to these “sharing” sites as they grow, and the owners receive venture capital and must monetize operations in preparation for an exit event.
- Today, the Prius seems to be the universal choice for Uber drivers.
- Fligstein, N. 1993. The transformation of corporate control (Cambridge: Harvard University Press).
- Many have noted that a number of young entrepreneurs who created platforms came from social backgrounds in which their parents were able to support them during their “hacking adventures.” If they failed, they could return to the university. The fact that these entrepreneurs were backstopped by family and social status does not diminish their accomplishments, but provides context.
- See, for example, Latour, B. 1990. “Technology is society made durable,” The Sociological Review, 38(S1), 103-131.
- Lessig, L. 1999. Code and Other Laws of Cyber Space, (New York: Basic Books).
- Ton, Z. 2014. The Good Jobs Strategy: How the smartest companies invest in employees to lower costs and boost profits (Amazon Publishing).