Papers on Public Policy: Trade off Between Protecting Investors and Promoting Entrepreneurial Activity

Do investors need “protecting” from making angel investments? What actually happens when those protections are made?

Today we will examine two papers on SSRN that cover a seemingly-innocuous change to accreditation rules, so see what effect changes to who can invest have on the companies seeking funding.

The first paper Is There a Trade-off Between Protecting Investors and Promoting Entrepreneurial Activity? Evidence From Angel Financing by Jiajie Xu, an excerpt from a dissertation from the University of Iowa. Available here (and here too).

This paper looks at the impact of the change in 2011 to the definition of accredited investor, where the SEC removed the value of a primary residence from the wealth definition. This sounds like an innocuous and minor change to the rules governing who can invest in startups.

Well, this apparently-innocuous change had a significant impact! Removing home equity possibly removed over 20% of previously-eligible households (p.3), because a significant proportion of household wealth for single-digit millionaires would be their home’s equity.

Even more dramatically, the rule change had a clear, material impact on angel funding, and a whole cascade of downstream impacts. It “had a significantly negative impact on local angel financing” (p.3), with a decrease in angel funding as “marginal angel investors” had to leave the market; fewer follow-on rounds; fewer exits (p.4); fewer patents, reduced revenue, and fewer jobs; greater borrowing from SBA loans and more entrepreneurs taking out second mortgages. A whole litany of costs on entrepreneurs and society that significantly exceeded the benefit (from estimated lost investments avoided) to the supposedly-protected investors. Yikes!

How did the paper reach those conclusions? The paper looked at the ratio of home value (from Zillow) to individual net worth (from SIPP and IRS data) to create a “HV/NW ratio” for each city. The higher the ratio, the more impact the rule would have, as a higher proportion of high-net-worth people would be removed from the market. It looked at the differential impact in subsequent activity in several different areas, and whether the impacts were worse in areas with higher HV/NW ratios. The results were dramatic (and statistically-significant).

How plausible is this? As always, we can complain about the data — Crunchbase, Form Ds, etc., give incomplete and sometimes entirely wrong information. We could object to some of the concepts and assumptions, in particular that as accreditation by income was unchanged it’s not obvious that just changing the wealth calculation would have impacted angels (who also generally self-certify, so may not have read the rules or necessarily followed them truthfully). And we always question if “ability to raise a next round” (p.28) is very illustrative. (And we’ll leave for you to opine on the benefits and concerns of difference-in-difference analyses more generally.)

Particularly powerful for us, as investors in “local markets” was that the negative impact was particularly acute outside of the traditional hubs of San Francisco , NY, and Boston (p.26). It’s also pleasing (while also being upsetting) that removing the “marginal” angels did not just prevent “marginal” companies being funded, as the impact on jobs, sales, etc., were significant, indicating high quality companies were hurt too. Local market angels do invest in good companies (when they’re allowed to!).

Overall, the paper is disturbingly effective in showing the impact of one small regulation change — a net negative of (at best) $6.32 billion (p.41) (though with major caveats about second-order effects here).

If this is repeated with new innocuous rule changes, or potentially some dramatic changes like indexing accreditation to inflation, there could be big negative impacts on early stage funding as angels are removed from the market. All the work going on to increase entrepreneurship could be undercut in one easy rule change.

Academics love a good change in the rules, as they often create clear boundaries that show how things really worked. The second article covering the same rule change is Angels, Entrepreneurship, and Employment Dynamics: Evidence from Investor Accreditation Rules by Laura Lindsey from Arizona State University and Luke Stein from Babson College, first uploaded to SSRN in 2017 (get it here).

The paper explores the same rule change; it also uses some of the same datasets and difference-in-difference methodology. But the focus this time is the impact on new business formation and related effects (like job creation and wage effects). Can you guess the impact Lindsey and Stein found?

Of course you can: fewer businesses were started. “We find a negative and statistically significant reduction in new businesses of about 2% on average.” (p.3) That’s an “economically meaningful and plausible” effect (p.23). And that, of course, leads to fewer new jobs (this is pretty common knowledge — new businesses create all net new jobs — but here is more supporting evidence) and to lower wages for employees overall.

The particular nuances here are that it is smaller angel investors (people investing a few thousand dollars in total), not high profile celebrities or super angels, that got hit most — and it is smaller businesses that also got most hurt. They weren’t bad investors and bad businesses: the paper shows once again the good/bad news that these “marginal investors” were not just funding businesses of lower quality. Good businesses didn’t get funded that otherwise would have.

We might not be enough of a conspiracy theorist to say this impact was deliberate, but it’s hard at this point to see these consequences as “unexpected.”

A couple of fun nuggets:

- Like Jiajie Xu, the authors think this rule change reduced the number of accredited households by “almost 20%” (p.11). Not every household that is allowed to make angel investments does, but still that remains a surprising proportion.

- “for West Virginia… the SIPP sample included no potentially accredited investors.” (p.15). This doesn’t quite mean there are no angel investors in WV, but it does mean the Country Roads Angel Network needs extra appreciation!

“In the U.S., many angel transactions require no disclosure, and where disclosure requirements do exist, enforcement is lax and compliance levels may be low.” I don’t know what US this is referring to, but the one I’m in does not sound like this!

Hopefully you found those papers though provoking. As legislators and regulators propose new or changed rules (like the HB2799 discussed here) we hope they do too.