Don't Shoot The Message Board: A Data Driven Look At The Impact Of Section 230 On Innovation And The Economy
from the these-things-matter dept
We’ve obviously been talking a lot about Section 230 of the Communications Decency Act over the past few years — and it is often credited as being the most important law for the internet. Jeff Kosseff’s recent book calls it “the 26 words that created the internet,” while David Post once declared that Section 230 probably “created a trillion dollars or so of value.” We’ve talked a lot about how the real benefits of Section 230 are not to the internet companies themselves, but to the public’s free speech rights, but over the last few years it’s bugged me that there wasn’t a better attempt to measure the actual economic impact of Section 230 and other intermediary liability regimes.
Today, in partnership with NetChoice, we’re launching our new report: Don’t Shoot the Message Board, that attempts to explore what the data shows concerning the economic benefits of Section 230. We chose the name because it’s perfectly fitting. Section 230 was, literally, written and pushed by (then) Reps. Chris Cox and Ron Wyden in response to the awful ruling in the Stratton Oakmont case, which suggested that any company hosting a message board could be found liable for any of the content on that message board. Similarly, the common phrase is “don’t shoot the messenger,” which is very much about not blaming the party merely delivering the message, as opposed to creating or causing the message. Putting liability on intermediaries is very much about blaming the messenger for actions of someone else.
To provide some actual data for this debate, the report first compares a few different intermediary liability regimes to see what we can parse out. It starts with looking at the US vs. the EU. In the US, we have CDA 230 for most platforms, and DMCA 512 for copyright-related platforms. In the EU, they have the E-Commerce Directive (and soon they’ll have the implementations of the Copyright Directive, but this report looks at the state of things before that). In many ways the intermediary liability in the E-Commerce Directive is much more similar to the DMCA in the US than CDA 230. Using a variety of datasets and comparison points, we found that CDA 230 appears to have resulted in significantly higher investment in US internet companies who rely on CDA 230. Our data suggests CDA 230 alone is probably responsible for two to three times more investment in the US than the EU. It also drove much higher levels of investment as companies in our sample were five times as likely to raise over $10 million in the US and nearly ten times as likely to raise over $100 million than their counterparts in the EU.
Tellingly, when comparing copyright-focused platforms, where the liability standards are similar (or, at least were similar, prior to the Copyright Directive), we found EU companies did much better compared to their American counterparts. In other words, it appears that different choices for liability regimes can have a major impact on the types of investment and how much is invested. One reason why many of the big music platforms may have come out of the EU, rather than the US, is that the US’s decisions on intermediary liability no longer gave the US an advantage for those kinds of platforms. In short: the decision to offer fewer liability protections in the US drove those investment dollars elsewhere.
The paper also compares the DMCA and the CDA in the US alone, to see if there’s a major difference in spurring investment — and we found clear evidence that having these two different regimes resulted in much more investment focused on platforms that rely on CDA 230 (social media, communications platforms) as compared to content based platforms where the DMCA is of greater importance. There is obviously overlap, as many platforms rely on both laws, but when breaking out music companies vs. cloud computing, cloud storage, e-commerce and social media companies, we found over and over again that the latter received more investment, were more successful in the long run, more likely to have a successful exit and less likely to shut down. In other words, having the strong protections of CDA 230 seemed to help lead to more successful companies, and thus, more innovation.
To back up these findings, we looked at a variety of countries where a major change — either from a key court ruling or changes in the law — created a sudden shift in intermediary liability protections, and then did a before-and-after analysis of the impact on investment and startups in those markets. Once again, the findings were more or less what we expected. When a country strengthened the protections for intermediaries, investment went up, the number of startups increased and there was greater innovation. When a country removed or weakened such protections, investment dropped. Noticeably.
I should note that there were some exceptions to this rule — and that happened mostly where there wasn’t an existing strong startup ecosystem (such as Argentina). There, when intermediary liability laws were strengthened, there was little evidence of a sudden influx of investment. So, that suggests that strong intermediary liability protections are important, but not the only important thing, in driving greater innovation and investment.
As we note in the paper, this is not meant to be the definitive look at these issues. There are many, many different factors and variables that play into the startup and innovation ecosystems. However, there was so little data on the direct impact of things like Section 230 that we thought it would be helpful in furthering the debate to at least have some data-driven research into the impact.
At a time when politicians around the globe are suddenly increasingly interested in weakening intermediary liability protections, they should at least consider what that might do to investment, innovation and jobs. Our new report suggests weakening such protections may very well be “shooting the message board.”