> compliance is so burdensome and expensive that firms don't want to enter the public markets

There's something bigger going on than SOX. That was the hypothesis the JOBS Act was predicated on [1]. Yet the trend continues uninterrupted.

My hypothesis: broadly held (i.e. by employees) companies taking long-term technology risks are fundamentally new. Moreso than reporting requirements, minute-by-minute pricing isn't helpful to a company challenging accepted wisdom over long time horizons. (Contrast: Tesla and SpaceX.) Being able to time large-scale valuation changes with technological milestones is a huge advantage no amount of public deregulation will compete with.

The trade-off between valuation control on one hand and liquidity on the other hand has multiple solutions. Deep private markets have an equilibrium between closely-held private businesses (e.g. Koch Industries) and widely-held public companies (e.g. Apple). (Multi-tier voting in public companies, as Snap and Facebook have, on the the other hand, seems untenable.)

[1] https://www.sec.gov/spotlight/jobs-act.shtml

This has less to do with Sarbanes-Oxley than you might think. The real shift is how much money is chasing very low returns. When 4% over inflation starts looking great companies don't need to go public to access billions in capital.

I doubt that investing in early-stage unicorns is the best way for the poor to get out of poverty.

No, but more folks in the middle class might be able to remain in the middle class if their modest portfolios had access to high-growth companies, ideally through index funds rather than individually-picked stocks

What weighting index of all IPO stocks during the sitcom era would produce a performance boosting return? Does AMZN fix all woes?

only in hindsight. but if they were part of an index the stocks would be even more expensive to buy into reducing gains overall. and remember that unicorns do not pay dividends.

I don't know if this is true, but if it is wouldn't there be a really obvious pattern of the market value of initial IPOs going up compared to before SOX was passed? Is there such a telltale pattern?

It is true

https://www.vox.com/2014/6/26/5837638/the-ipo-is-dying-marc-...

This article is just cherry-picking examples of specific tech companies and trying to make the point that SOX is onerous on that basis.

If instead, you look at NASDAQ's own website of recent IPOs[1] and look at the market cap of those companies there's plenty of smaller companies to be found.

E.g. BWB went public less than a month ago and has a market cap of ~300M, meanwhile that Vox article is claiming Netscape's IPO of 2B would be impossible today. AIHS is hovering at ~150M and DNJR at ~60M, both companies public in the last month.

Again, maybe it's true, but given all that I'm more willing to believe that the tech sector in particular just has more access to private equity than it did 20 years ago, and thus there's less pressure to go public. It's more of a hassle to be public, and that applied before SOX, at the very least you need to deal with nosy shareholders wondering how their money's being used.

1. https://www.nasdaq.com/markets/ipos/activity.aspx?tab=pricin...

> hurting them by putting growth opportunities out of their reach.

You're asserting that the "alpha" of the private opportunities is higher, even after you've accounted for risk. This should be amenable to statistical evidence? Including all the VC-funded duds and collapsed buyouts like Toys R Us.

I'm not sure that retail investors want this kind of risk profile; it seems to me that the demand is phrased in terms of interest rates. And to get back to higher interest rates we'd need more inflationary pressure from growth - which we don't have.

This is a false dichotomy. Just because the regulations we imposed in the past made things burdensome, doesn't mean that regulations we impose in the future have to be burdensome.

I'd want to understand why they're burdensome. Give me some concrete examples. Everybody talks about how regulations are burdensome but nobody actually mentions the specific things that are a burden.

I've never dealt with SOX compliance personally, but I did deal with PCI compliance. Now, PCI compliance isn't government-mandated, but since the penalty for non-compliance is that they take away your ability to take credit cards, it's treated as though it were a government mandate in a corporate setting. My observation about PCI compliance was the regulations themselves were clear, helpful, and well thought out. The burden was the months upon months spent arguing about who interpreted which regulation which way (and God forbid you just ask the regulators!), and an apparently certifiably insane executive board that would rather spend $1000 avoiding each individual regulation than spend $10 adhering to it.

You're required by to set up systems to ensure accounting information can not be tampered with, even if people tampering with accounting information isn't considered to be a high risk. To make it more concrete, it means you can't push to master any more in a git repository to fix some botched merge because, it happens to deal with revenue in some tiny corner of it's functionality and every change has to be approved by multiple people

In light of git's enforced immutability constraints and the fact that multiple approval for merges is SOP at many large companies, I suspect that isn't a very good analogy.

I thought they meant that literally, and not as an analogy.

At my company, developers can't do production deployments for, I've been told, "SOX compliance".