8 ms·
Securities laws, in general, have never been absolute in the United States. When companies offer stock or other securities to purchasers, the broad rule is that
by grellas 15y ago
Securities laws, in general, have never been absolute in the United States. When companies offer stock or other securities to purchasers, the broad rule is that "you can offer anything you want, even something junky, so long as you disclose all material elements associated with the offering such that a reasonable investor can make an informed decision in deciding to purchase it." In other words, there is no all-seeing, all-knowing authority supervising the process who declares that the offering is substantively sound. It may or may not be. All that is required is that the issuer meet the requirements imposed by securities laws for disclosing all material facts relating to the investment. That is what the registration statement does in a public offering. And that is all it does. If that is done, and junk is offered, and the public wants to buy junk, the securities laws permit this.
Even respecting disclosure, though, the U.S. securities laws have always tempered the burdens associated with making detailed disclosures of the type required in a registration statement with important rules saying, in effect, "we as regulators realize that requiring companies to go through a multi-million process just to offer their securities to investors is too much and therefore we will exempt a broad number of categories from the registration and detailed disclosure requirements to enable small companies to offer their stock for sale as well." That premise underlies a whole range of securities law "exemptions" that permit small offerings, etc. so that companies can grow and develop without choking on process. The ultimate exemption under federal law is Section 4(2) of the Securities Law of 1933, which basically exempts private placements from the burdens of going through the registration process. Section 4(2) has been around forever and has no formal requirements. It simply provides that anything that is a true private placement, as opposed to a public offering, is exempt. Because the assessment of what is a private versus a public offering turned on detailed facts and circumstances, and this in turn led to substantial uncertainty and lots of litigation (is an offering made to 20 people "public"? how about if you don't know them? how about if you advertise the offering to get them interested? how about if they are small, unsophisticated investors?), the ultimate exemption - or, more accurately, "safe harbor" that assured an issuer that an offering would be exempt - was Regulation D, adopted by the SEC in 1982 and widely used by startups ever since that date to make tons of private placements that have been streamlined, simple, and cost-efficient ways of offering their stock to investors.
As is apparent from the above, the securities laws have always sought to strike a balance between imposing regulatory requirements (and burdens) that are aimed at protecting investors, on the one hand, and moderating the burdens so imposed to facilitate capital formation for situations where it makes no sense to impose needlessly burdensome requirements on small issuers and where less burdensome, fallback protections can be used instead of the full panoply of protections that apply to larger-scale offerings. By definition, this means that U.S. securities laws have always recognized a trade-off between having strong regulatory requirements aimed at investor protection, on the one hand, and lessening such requirements for some situations so as to give practical routes for capital formation for companies unable to meet the rigorous requirements. At many points along the way, the legislators who pass such laws and the regulators who administer them make multiple social policy judgments saying, in effect, "this is a situation that calls for the maximum protections but this one will leave investors fairly protected with more minimal protections in place." That is why it costs many millions of dollars to do the legal and accounting work to take a company public but only a couple of thousand to issue stock in a new corporation and only a few tens of thousands to raise a few million in a Series A private placement. The law is designed to accommodate the practical needs of companies that want to raise capital. Securities laws don't vanish in the private placement context. They simply impose far fewer requirements aimed at investor protection and all the more so when investors are presumed to have a strong ability to protect their interests (this is why offerings are often limited to "accredited investors," i.e., high net-worth or high income individuals, among others).
The JOBS Act is a piece of legislation that takes the rather burdensome accounting requirements first imposed by Sarbanes-Oxley on all publicly traded companies - and adding $1M+ in annual costs to even the smallest issuer in order to attain regulatory compliance under those rules - and exempts a set of relatively smaller publicly-traded companies from having to comply with those requirements for a 5-year ramp-up period after first going public. This part of the Act says, in effect, "we realize that the IPO market has been moribund ever since SOX was enacted and, because part of the reason is the heavy regulatory burdens imposed by SOX, we will seek to encourage more IPOs by giving issuers more incentive to go public without having to face huge expenses right out the gate." Now, this social policy judgment made by Congress may or may not be sound. But it is a policy judgment declaring that the SOX rules are just too much for relatively small companies just going public and therefore should be relaxed for such companies in order to enable them to realize their practical goals of going public, building momentum, and only later having to comply with the full SOX rules. One can question this judgment but one cannot question that it falls squarely within the pattern and practice of U.S. securities laws as implemented for decades. It is always a trade-off between optimum investor protections and practical limitations on such protections in the name of letting legitimate capital formation get done. Will this "legalize fraud," as suggested in this piece? I doubt it. The SOX rules have a short history and securities laws go back to the 1930s, more or less ably protecting investors during their long tenure before SOX took effect. Such protections will continue to exist for offerings made by these small issuers who will get some interim relief from SOX requirements. One can argue that it is bad policy to afford such relief. But to suggest that it "legalizes fraud" is to absurdly overstate the case.
The JOBS Act similarly loosens requirements for crowd-funding, for enabling private companies to have larger numbers of shareholders before having to register as publicly-traded companies, and for other contexts as well. On balance, it is aimed at promoting more effective capital formation by loosening otherwise strict SEC rules when new conditions warrant. This, to me, is very good for startups and the Act as a whole should, in my judgment, lead to many excellent results. That is why it has received almost uniform and very strong support from pretty much the entire startup community. It does not legalize fraud. It strikes a classic balance between formal investor protections and real-world practicalities. If the balance proves wrong, nothing will stop Congress from pulling back. In the meantime, let's see if crowd-funding can be used to give us new ways of raising capital and if the IPO market can't be rejuvenated after a long dead spell. The Act stands to benefit startups in major ways and, though not exempt from criticism, is by no means some radical departure away from investor protection under U.S. securities laws. On the contrary, it stands squarely within the traditions of those laws and is a good example of precisely how such laws have been implemented for many decades.
- ricardobeat 15y agoThanks for the explanation. You should publish this.
- ghshephard 15y agoI would politely suggest that he just did. To the most important and selective audience imaginable.
- CodeMage 15y agoI imagine that the GP's point was that publishing here doesn't reach the same audience Rolling Stone does. If so, I agree.
- ghshephard 15y agoI would politely suggest that he just did. To the most important and selective audience imaginable.
- excuse-me 15y agoTurning it from a skilled game of picking stock into a skilled game of avoiding disclosure. I didn't know that "Florida Swamp Land Inc" was a bad buy. Although it was promoted by a wholly owned subsidirary company that I actually work for from the same desk in the same office - but our internal rules mean that I didn't know that I knew that when I was working for the selling company. In case you are concerned here is a report from an independent ratings agency that we pay to say this - and coincidentally I also own and work for.
- tptacek 15y agoSo, this is obviously a great comment, but two responses: * The SOX requirements that capped the first Internet bubble clearly retarded tech company IPOs, so that only companies with (say) more than 60MM/yr cash-flow-positive revs could consider IPO'ing. The net effect of that for startups is negative: it virtually eliminates one previously available path to liquidity. But the net effect of that to society has been to transfer a lot of risks that were previously borne by the public markets to private equity and VC. Is that a bad thing? In other words: yes, way fewer IPOs. But also much higher quality IPOs. Even Groupon is superior in many ways to companies that managed to IPO towards the end of the first bubble. Also: companies without a clear path to acquisition are forced to adopt business models that distribute profits to owners (else why run the company). The end result of that might be pretty positive. * The Crowdfunding provision in the act clearly doesn't "legalize fraud". That's a hyperbolic claim. But is it hyperbolic to say that it simplifies and eases fraud? Especially if the Crowdfunding disclosure and capital restriction rules place Crowdfunding in a "worst of both worlds" situation in which shady (or just incompetent) companies find it productive to raise from crowds, but valuable companies get fed to insiders at VC funds.