Thursday, February 09, 2017
Predicting the Future of Legal Services
Thanks to Dan Rodriguez for organizing this symposium and inviting me to participate. As a long-time reader of Prawfsblawg, I’m especially delighted to join the conversation.
It wasn’t easy to decide how to contribute to the discussion about two terrific books. They both raise so many important issues and make so many provocative points that it is hard to engage with the ideas adequately in a few posts. I ultimately decided to focus my comments primarily on Richard and Daniel Susskind’s The Future of the Professions: How Technology Will Transform the Work of Human Experts. In this admittedly lengthy post, I explain why we should take the Susskinds’ predictions seriously. In two subsequent posts, I’ll explore the implications of the predictions for legal education and legal services regulation, when I also hope to touch on Gillian Hadfield’s book, Rules for a Flat World: Why Humans Invented Law and How to Reinvent It for a Complex Global Economy.
My views on the future of legal services have been informed by Richard Susskind’s earlier books and my own work on projects where “futures” discussions have been front and center. For example, I recently completed service as the vice chair of the ABA Commission on the Future of Legal Services, which produced a report describing many “legal futures” issues in great detail. And I am now the chair of the governing council of the just-launched ABA Center for Innovation. Before becoming a dean, I was fortunate to serve as Suffolk Law’s inaugural director of our Institute on Law Practice Technology & Innovation and the related concentration in the area.
These experiences lead me to believe that we are going to see more significant changes over the next couple of decades than we have seen over the last twenty years (and those recent changes already have been considerable). That’s a long way of disclosing that I read the Susskinds' book expecting to find myself largely in agreement with their predictions. I wasn’t disappointed.
The Essence of the Susskinds’ Forecast
The Susskinds’ predictions turn in no small part on an important narrative about how people have developed and shared their expertise during different periods of human history. Namely, we have seen an evolution from strictly oral communications, to written work, to modern printing, and (most recently) to a digital age where knowledge is acquired and shared with great ease (pp. 147-53).
The Susskinds observe that, before the current digital age, information was difficult to obtain, giving professionals an important role and advantage. People could not easily find the information they needed about a topic, such as medicine, law, or accounting, so the public had little choice but to consult experts (e.g., doctors, lawyers, accountants, etc.) to answer even routine questions.
The Internet is now leveling that playing field, making the dissemination of expert knowledge considerably easier. (This online symposium is an example.) When combined with technological advances that have facilitated the automated delivery of that knowledge and related services – think the Mayo Clinic or WebMD (for medicine), LegalZoom (for legal services), TurboTax (for accounting), and Khan Academy (for education) – the Susskinds argue that we are beginning to see Clayton Christenson-like changes to the professions (though the Susskinds prefer to avoid the language of “disruption”). (pp. 109-10). The Internet is not only making it easier for non-experts to gain access to the information they need; it is driving a gradual expansion of automation from low-cost, routine professional services to more bespoke services, especially as artificial intelligence (AI) becomes more sophisticated (e.g., IBM Watson’s applications to the healthcare and legal industries).
The Susskinds do not argue that human-based bespoke services will necessarily disappear (pp. 192, 199), but they contend that we will see a continued transformation of how professional services are delivered and related changes to the basic skillset that future professionals will need. In a nutshell, they conclude that professionals will have to partner with the “machines” rather than try to beat them.
This is a greatly simplified account of a nuanced and well-researched book, but for those of you who have not read it, this is the gist of it.
My experience has been that some lawyers either do not agree with this forecast or have not given the subject a great deal of thought, so in the remainder of this post, I’m going to explain why I believe that we need to take the Susskinds’ forecasts seriously. (If you’re already convinced that the Susskinds’ have painted a reasonably accurate picture of the future of the professions, you can safely stop reading here.)
Is the Prediction Right?
The quotable baseball manager Yogi Berra once said that “[i]t's tough to make predictions, especially about the future.” Tough, indeed, but not impossible. Richard Susskind’s predictions about the future of legal services have been prescient for decades. Although Richard hasn’t always been on the mark, he’s been right often enough that his ideas deserve careful attention.
One of the primary reasons that I’m convinced that the Susskinds’ forecasts are reasonably accurate is that the predicted future is already taking shape. As with just about any prediction, the odds of getting it right improve considerably with additional data. Take, for example, a prediction that autonomous cars are going to transform transportation within our lifetimes. Twenty years ago, that prediction would have been quite speculative, because the technology needed for such a development was not yet available. But given the technology that now exists, we can predict with a much higher degree of confidence that driverless cars will become ubiquitous and transformative within a couple of decades or so, perhaps considerably sooner.
The analogy between self-driving cars and legal services innovation is not perfect, but it is instructive. Like predictions about the coming ubiquity of self-driving cars, predictions about the likely transformation of professional services are drawn from existing data and technology, not a speculative forecast about future capabilities.
Consider some recent developments in the legal industry. In the context of dispute resolution, online platforms (online dispute resolution, or ODR), led by Modria, now resolve approximately as many disputes as the entire U.S. court system combined. The technology is now moving into courthouses. Court-annexed ODR platforms are emerging, and more are likely to appear in the near future. Even when disputes are resolved in courts using traditional procedures, technology is changing the landscape. When I started practicing 20 years ago, discovery required associates (and paralegals) to engage in page-by-page reviews of paper documents; today, technology-driven ediscovery is performed by outside vendors. And technology is informing how lawyers do their work. For example, companies like Lex Machina (now owned by LexisNexis) and Premonition are using data analytics to give lawyers valuable insights about opposing counsel, courts, etc. to craft better arguments and assess the value of cases. One company offers to scan your opponent's briefs and suggest possible cases to cite in response.
In the context of transactions, LegalZoom now automates the creation of a wide range of basic legal documents and has served millions of consumers. Numerous other companies and organizations offer similar services, either directly to the public or for law firms, legal departments, courts, and legal services organizations. Due diligence and contract management are often outsourced to legal process outsourcers (LPOs), as are a range of other services.
Law firms and in-house legal departments are responding to these changes in various ways. In addition to making greater use of document automation, they are using expert system tools, creating legal project management departments, hiring legal solutions architects to design new ways of delivering legal services, and establishing research and development departments (e.g., Dentons’ NextLaw Labs, Davis Wright Tremaine’s De Novo, Seyfarth Shaw’s Seyfarth Lean, and Littler Mendelson’s Service Solutions). Law firms are also diversifying their revenue sources by creating ancillary businesses, such as e-discovery services or data analytics. In-house counsel are placing a greater emphasis on legal operations (e.g., the Corporate Legal Operations Consortium (CLOC)), and they are making greater use of LPOs and unconventional lawyer-staffing solutions that include new technology solutions as well.
Bar associations are responding too. The ABA recently issued a report on these developments and has established a Center for Innovation. State bar associations are examining futures issues, and other associations around the world are engaging in similar efforts (e.g., the U.K.’s Law Society and the Canadian Bar Association).
This is all just a sampling. New legal tech and innovation startups are appearing everywhere and delivering a growing range of services. Venture capitalists are taking notice as well and increasing their investments into innovative solutions for the legal industry.
At the same time, the available tools are getting more sophisticated, especially as AI itself becomes more capable. There are increasing efforts to apply AI to law (e.g., ROSS), a development that fits nicely into the Susskinds’ predictions that the changes already underway will continue to transform legal services, even at the most sophisticated levels of the industry.
A Caveat: “The future is already here — it's just not very evenly distributed.”
This quote is often attributed to author William Gibson, and it offers a pretty good summary of what is happening with professional services. Innovations are transforming those services, but the changes are (and likely will continue to be) uneven.
To be sure, the Susskinds concede this point. They believe that there will not be a “big-bang revolution.” Rather they predict “incremental transformation,” or a “staggered series of steps and bounds.” Nevertheless, the Susskinds conclude that “the eventual impact will be radical and pervasive.” (p. 231)
Here is where I may part ways just a bit from the Susskinds. I agree that technology-driven changes to the delivery of professional services will take the form of “incremental transformation,” but I think that the extent and especially the timing of the changes are likely to vary much more than the Susskinds imply. The transformational changes that they predict will vary from one profession to the next (and vary among professional practice areas) by considerable periods of time, perhaps decades. Some practice areas within some professions, including the legal profession, are sufficiently bespoke that the AI necessary to replicate the services or substantially transform how they are delivered is still entirely speculative (much like self-driving cars were twenty years ago). That’s not to say that transformational changes will not eventually emerge in these resistant fields; it is just that the technology necessary to bring about those changes does not yet exist and may not for quite a bit more time.
This is a variant of the objection that the Susskinds confront in their book: “this may be true of everyone else’s practice area, but not mine.” (p. 232) My objection is a little bit different. I’m not suggesting that any particular profession (or practice area within a profession) is immune from these changes. I’m arguing that the changes are likely to be more uneven and difficult to predict in terms of scope and timing than the Susskinds suggest. In other words, nobody is immune from the changes, but some are likely to be more resistant than others.
Consider a recent McKinsey report (related New York Times story here), which concludes that automation of industries (including the professions) will vary depending on technical feasibility, the cost of developing and deploying the solutions, labor market dynamics, economic benefits, and regulatory and social acceptance. Assuming these factors are the right ones (and they seem right to me), the pace of change is likely to vary depending on the industry and the specialty. Indeed, the report suggests that the “professions” themselves have among the lowest automation potential of the types of employment surveyed. (See Exhibit E4.) Although I think that claim may underestimate the automation potential in many parts of the legal profession, the point is that the extent and timing of the transformation of professional services is not easy to predict. The report explains that “[o]ur scenarios suggest that half of today’s work activities could be automated by 2055, but this could happen up to 20 years earlier or later depending on the various factors, in addition to other wider economic conditions.” That’s quite a broad window, and it reflects the uncertainties that necessarily exist when venturing predictions across industries, professions, and specialties.
To be clear, I’m not suggesting that the Susskinds’ forecast is wrong. I’m just a bit less confident in the general applicability of their forecast than I am about (say) predicting the coming ubiquity of self-driving cars.
Assuming you’re convinced (as I am) that the Susskinds’ vision of the future is roughly close to the mark, what should we do about it? In my next two posts, I’ll explore what this all means for legal education and legal services regulation.
Posted by Andrew Perlman on February 9, 2017 at 08:15 AM | Permalink
The comments to this entry are closed.