- or acquire a publicly traded company and do a reverse merger. Practically fee-free compared to an IPO. But doesn’t generate a lot of publicity.

Probably the companies are really after publicity more than the investment capital… so they can say “advised by Goldman Sachs” etc. etc.

Exactly. Collusion in the Madoff case, the mortgage securitization and MERS, and forclosuregate for 3 of thousands.

Posted by TinyOne: “Additionally, if you abolished the SEC and the registration process…”

That simplifying is how you get a MERS mess… so I doubt that would be the ticket to anything but more hell.

Thanks for asking hard questions Felix.

Additionally, if you abolished the SEC and the registration process, and let Companies freely issue shares in a manner they saw fit, issuance costs would drop.

High headline fees are inevitably popular with the banks as they do not like to advertise discounts. It may be that this approach is simply more common in the US – not sure how/if the authors in the study adjusted for this.

It reminds me very much of the real estate sales market. Realtors® charge 6% regardless of costs or conditions and they do not compete on price.

Recently, there have been a few cracks in the armor of real estate sales — low-commission realtors, limited-service, etc. But it’s not easy to break the 6% norm. The trick is that buyers care so much more about the ultimate price, which varies wildly — by much more than 6% or 7% — that realtors and banks doing IPOs can persuade their customers that they have more to lose by going to any emerging low-cost rival than they have to gain. Sure, your commission may be cut in half, but your price will be off by 10% or more. So why do it? (And that’s assuming there even _are_ low-cost rivals in the IPO market, which the chart suggests are very, very rare.)

These banks really are vampires, sucking blood from the rest of the economy. It is so disgusting.

You misunderstood that chart — the thing that appears to be a solid black line is in fact the 95% of US data points that are EXACTLY 7%.

There can be little doubt that IPOs are more expensive in the US. But to say that the US has a cartel whereas Europe does not, you would have to sow that the US distribution is significantly narrower than the European distribution.

True, the range is clearly greater for the European than for the US distribution; but there are so many more points in the European distribution that this can easily be misleading. Eyeballing the plot, I think the US distribution has a raw std deviation of about 1% and the European distribution has a raw std deviation of about 1.5%. These should be readily computable. Applying a statistical test (probably some sort of Chi-Sq test), my guess is that you would not find the two distributions to differ significantly in variance, and that the hypothesis that the US has a cartel whereas Europe does not would therefore not be borne out.

It would, of course, remain to be explained why the US is a more expensive place to do an IPO than Europe is, but that’s a very different question, best answered by asking someone who id an IPO in the US, “Why did you do your IPO in the US, when it’s so much cheaper to do it in Europe.” My guess is that most of the answers would be, “Well, our VC has a long relationship with bank X and he felt very comfortable going with them.” Comfort is expensive, as anyone who has purchased an automobile is well aware.

Theoretically, a company would cut their prices even when times are good. In practice, I just have not seen that happen. Companies do not generally do elaborate calculations of more volume for lower prices when they lower their price. They generally only do so when their volume starts to lag and they start losing money (mass-market retailers are an exception to this).

For IPO’s, you would think banks would have dramatically lowered their prices during the post-2000 IPO bust, but investment banks have some idiosyncrasies compared to companies that actually make stuff:

1. Their fixed costs are fairly low. They have office space and support staff, but for the IBD itself, these fixed costs are minimal compared to their revenue.

2. Most companies would consider compensation for their white-collar and skilled employees to be fairly fixed. Even today, many managers would rather cut prices than cut compensation costs. On the other hand, investment banks have more built-in variability in their compensation structure. Analyst bonuses may range from 10k in 2001 to perhaps 80+k in 1999 or 2006. There were stories in 2001 of IBD analysts working for less per hour than store managers at McDonald’s.

Banking managers also tend to have less reticence about layoffs during bad times. In this respect, their compensation costs are less fixed.

I would guess that in both respects, variable vs. fixed pay and willingness to lay off people in bad times, European banks’ compensation costs are more fixed. In bad times, they would rather lower prices to recoup compensation than lower compensation and keep the same costs.

I also wonder why the current IPO system has not been completely commoditized. My only guess is the role of gamemanship with IPO’s. In the regular underwriting system, Goldman Sachs sells to Fidelity who then probably gets a nice first-day gain, or “pop.” It would seem this first-day gain would hurt the capital raised by the end company, but in reality it helps publicize the stock for brokers to hawk it to small retail investors. With the publicity at their disposal, mutual funds can now capitalize on the pop through using an army of retail brokers.

To put it in more crude terms, the system is basically built to defraud the least possible rational investors, the retail shops. Eugene Fama may not approve, but that’s how the real world works.

The question then becomes why couldn’t mutual funds do the same thing with auctions instead of a 7% underwriting broker? The issue is that after the bank sells the IPO to its funds, the very next trade becomes the published price. The very next trade is whoever has the highest bid, or the least rational person, and therefore you get a big pop from the bank’s price to the “market’s” price, when the market is whoever the highest bidder is.

For auctions, that highest bidder is bidding alongside the funds. There isn’t the publicity of the funds buying the stock and then the “market” saying that it’s really much more valued. Without that publicity, the retail brokers do not have as good a story to tell. You would think they could tell the same story before an auction and the underlying company could get the value instead of the mutual funds, but it doesn’t work that way. Brokers just don’t have a good story for companies with no price discovery and that feeds down to a lesser price in the initial offering.

I was in business school 10 years ago and my professors railed against the IPO process as one in which all parties except the investment banks get screwed.

High profile companies often issue at a massive discount to the market open. (like Linkdin) So the company get’s screwed. Investors have no idea how many shares they will be allocated… just that if they get lots of shares the deal is probably a flop and if the deal sells like hotcakes the small investors get shut out completely.

10 years later and the process still sucks.

why can’t the NYSE or the Nasdaq just run a Google style dutch auction for new listings using SEC audited financials?

“There you go again” :-( Can you provide ANY argument, evidence, _other_ source than your firm personal conviction, to suggest that any participant in the IPO process has this even as as a secondary or tertiary goal, or even knows what it would mean? Bonus’s linked to price-discovery? Something in the prospectus or filed with teh SEC? Or maybe a revealed-preference argument showing that somehow IPOs are good at providing pricing (I would imagine this might involve an analysis of the size of the float, otherwise we could just list one share, right?)

I am happy to provide pricing, for the princely sum of $0.00. Here it is: $10/share. But of course, this is the _wrong_ pricing and an IPO (or just a properly run one??) provides the _correct_ (/better) pricing according to the following definitions and/or scholarly citation… [Felix, please help me out here?]

There doesn’t have to be a written agreement for collusion to exist.