Regulatory Sandboxes

The Future of Regulation in Technologically-Progressive Jurisdictions

Entrepreneur and investor Balaji Srinivasan wondered which jurisdictions are friendly to technological progressives in a recent article.

But what makes a tech-friendly country? Digital nomads may look at the internet speed, weather, Bitcoin taxes, or safety. You can play with these criteria—and many others—on NomadList to find the best location for you. Unfortunately, the number of sunshine hours doesn’t determine how technologically progressive a place is…

So should we look at the number of unicorns or the VC market? But these lagging metrics depend a lot on the size of the local market. Tech-friendly countries want to attract foreign talent, so we could consider foreign-born population or net migration rate. But many places welcome immigrants without being technologically progressive. Some governments offer programs to attract tech talents, investors, and remote workers; however, these countries are not necessarily technologically progressive themselves. For instance, no matter how good the Portuguese tech visa is, the local administration isn’t that tech-friendly (even though the situation has improved since Covid).

We could also look at the institutional framework: tax burden, rule of law, hiring and firing practices, absence of corruption, or how easy it is to start a business. These great indicators are gathered, with others, in global rankings such as Heritage’s Index of Economic Freedom, Fraser’s Economic Freedom of the World, the World Bank’s Ease of doing business index, the IMD World Competitiveness Ranking, the World Economic Forum’s Global Competitiveness Report, WIPO’s Global Innovation Index, or the US News Best Countries rankings.

These rankings are fantastic, but they were built for the twentieth century. For example, it may be easier to set up a new business—like a restaurant—in Denmark than in the US, but Uber and other ridesharing companies are banned in Denmark. And even though Germany is among the most innovative and competitive countries, startups can’t easily give stock options to employees there.

All countries want to make it easier to start and run businesses. All governments have been decreasing corporate tax rates over the last decades, and since the World Bank launched its “Ease of doing business index,” states have kept improving their score. Indeed, reducing the number of days to create a company is not controversial. Conversely, tech and new entrants may challenge entrenched interests and legacy companies. And there are also huge expectations from citizens and customers to maintain high-quality standards enforced by strict regulations. These regulations were often necessary and well-designed when enacted, but they may not be adapted to new technologies.

So how do countries deal with radical innovation in the most regulated industries?¹ They act like tech companies: they use a sandbox to test their code in a safe environment, iterate at a small scale, and then deploy.

To do so, governments create a “regulatory sandbox”: a “safe, testing space where participants can test their new business model, innovative products, services, and delivery mechanisms, without immediately incurring all the normal regulatory consequences of engaging in the activity in question.” According to Margaret Hagan, in such a safe playground: “the private sector can innovate without worrying about fines or liability, the regulatory agency can test regulations to see what works before going through the long process of creating new rules, and consumers have access to these services in a controlled environment. The goal is to relax or change existing regulation in a controlled and evaluated space to run real-world experiments. These experiences can be collected and inform evidence-based regulatory schemes. Sandboxes are used as an alternative to regulation that is based on speculation about what behaviors could result — and what risks and harms can emerge — from changing technologies or changing policies.” Sandboxes are similar to Chinese Special Economic Zones, established to experiment with the market economy, but they’re not limited to a specific location.

This revolutionary concept was invented by the UK’s Financial Conduct Authority (FCA) in 2015 to support fintech companies. Like all great ideas, regulatory sandboxes seem obvious in hindsight. That’s why Singapore (with its “never say no” approach), Hong Kong, Australia, Canada, Malaysia, Japan, and others quickly followed. In 2020, the World Bank identified 73 fintech sandboxes in 57 countries. More than half of them were created between 2018 and 2019, suggesting rapid growth:

World Bank, Global Experiences from Regulatory Sandboxes, Nov 2020

The UK FCA’s evaluation found that the sandbox had reduced the time and cost to market, facilitated funding, increased competitive pressure on incumbents, and led to high levels of innovation with new offerings (e.g., 40% of British sandbox firms were using distributed ledger technologies). Regulators also found there was little misbehavior on behalf of the participating firms. According to a 2020 paper, “the empirical analysis showed that the introduction of regulatory sandboxes had positive effects on venture investment in fintech industry.” And according to the Bank for International Settlements: “regulatory sandboxes improve access to funding by reducing information asymmetries and regulatory costs.”

The first sandboxes were limited to fintech, but the concept was then extended to other industries, such as:

• A healthcare sandbox for telemedicine in Singapore,

• A sandbox for drones and aviation in the UK,

• A sandbox for self-driving vehicles in Taiwan,

• A data protection safebox in the UK to reduce the uncertainty and ambiguity of GDPR related to new use cases of data,

• A sandbox for a digital ID in the UK,

• A legaltech sandbox in Utah to allow nonlawyers to provide certain traditional legal services, already joined by Rocket Lawyer,

• An energy sandbox in the UK allowing residents to buy and sell renewable energy from solar panels within their own apartment buildings. Typically, this is illegal in the UK, where generated power can only be used by the owner or sold to the grid.

Last month, the UK launched a “scale up” sandbox, as its programs initially primarily targeted startups.

All these regulatory sandboxes were focused on a specific industry. Japan and Korea went further with “all-inclusive” regulatory sandboxes that any tech company can join, from a shared kitchen service to an e-bike manufacturer.

Utah followed these examples and launched the first US all-inclusive regulatory sandbox in March 2021. Utah will create a new body, called the Office of Regulatory Relief (ORR). Any company can apply to the ORR, cite specific regulations that they wish to be waived, and explain how doing so will benefit consumers and promote innovation. The ORR then reviews the application with the relevant regulatory agency. Successful applicants receive regulatory exemptions for one year and may reapply for a one-year extension. A revolution!

What about other US states? Arizona was the first state to adopt a sandbox in 2018 (in fintech). Then followed by Wyoming, Nevada, West Virginia, Hawaii, and Florida in fintech; and Kentucky, South Dakota, and Vermont in insurtech. The State Bar of California’s approved a sandbox for legal services in 2020. Still, it hasn’t been implemented yet because some lawyers are afraid of increased competition and see this initiative as a “seismic event.” Tennessee is also considering an all-industry sandbox similar to the one in Utah. [Update: The Tennesse sandbox didn’t pass.] There are currently 28 initiatives for new sandboxes in state legislatures across the US.

Source: Libertas Institute. Note: the California legaltech sandbox has only been approved rather than implemented.

Unfortunately, there’s no US federal regulatory sandbox (as far as I know, please answer this post if I’m wrong). In 2018, the Treasury noted that: “replicating the regulatory sandbox approach in the United States is complicated by the fragmentation inherent in the US financial regulatory system: the sheer number of disparate regulatory actors, at both the federal and state levels, compounds the difficulty of implementing a single, comprehensive sandbox-style solution. A no-action letter or exemptive relief granted by one regulator does not bind others.” However, some agencies like the Consumer Financial Protection Bureau (CFPB) may issue “No Action” letters to startups, a partial alternative to sandboxes. On the other hand, the EU plans to create a pan-European sandbox for distributed ledger technology (DLT) and blockchain services by 2022.

Every month, a new sandbox program starts somewhere. Still, not all sandboxes are created equal. Some only provide limited exemptions, so startups don’t apply, as for the Ontario Energy Sandbox. Or the testing period may be too short. Building an excellent regulatory sandbox isn’t easy. And some researchers even criticize them because they “put pressure on the time and resources of already over-burdened regulatory authorities.” But that’s precisely why I think they’re fantastic! Jurisdictions with regulatory sandboxes accept transferring the regulation’s burden—time, money, and skills—from companies to governments. I believe this is the best evidence that a country is technologically progressive.

The UK and Singapore have arguably been first movers, and they continue to pioneer innovation in this space. But other jurisdictions, such as Utah, are catching up, and the competition will be fierce!

There’s probably a regulatory sandbox program tailor-made for you, depending on your needs. I listed more than 100 sandboxes in 63 jurisdictions here. Have a look, and go where you’re treated best! And if you find any mistake or if I forgot a program, please get in touch with me or comment on this post.

1. Such as energy, transportation (drones, ridesharing, autonomous vehicles, etc.), finance, insurance, healthcare, legal (legal advice, arbitration & ADR, etc.), defense, food, agriculture, fishing, or gambling.