Should we be scandalized by IPO pops?

So, I’m late to this. There’s a big link parade at the end of the post.

LinkedIn had an IPO on May 19, priced at $45 per share. The stock briefly sold in the $120s that day, and closed at $94.25. In the lingo, there was a big “pop”.

From a certain perspective, IPO pops are puzzling, even scandalous, events. Here is the theory of the outraged:

  • Shares are assets with real economic value that professional investment bankers, after communicating with “the investor community”, are capable of discerning.
  • The price at which those shares will trade in public markets, on the first day and thereafter, is a reasonable approximation of real economic value, because stock markets are efficient.
  • Therefore, if investment banks (who, in consultation with the issuer, set the IPO price) sell the shares for substantially less than the price at which the shares trade on the open market, they have screwed their client. The issuing firm and its original investors have “left money on the table” by failing to extract the full value their shares. Meanwhile, someone, some flipper, will have purchased shares at the IPO price and resold them after the pop, taking profits that might have gone to the issuer.
  • So issuers really ought to be upset about IPO pops (even though they mostly aren’t). “I don’t understand why competitive forces don’t drive this kind of egregious underpricing out of the system,” a finance professor tells the Financial Times.

Puzzle, puzzle, toil and trouble. Here’s a tidbit to taunt the good professor: In the 1990s and especially during the tech bubble, pops tended to be larger for IPOs led by “top tier“, high market-share investment banks than when shares were offered via midlevel underwriters. It’s a topsy turvy world, apparently. During hot IPO cycles, when underpricing is especially pronounced, “competitive forces” seem actively to favor underpricers.

Pretty much every premise of the case against IPO pops is false. Shares of all but the most staid firms do not have known, predictable economic values that highly trained professionals can predict ex ante. Further, share prices are autocorrelated, which is a fancy way of saying that if a stock trades for $100 today, it is more likely to trade above $100 three months from now than a stock that trades today for $50. There are lots of ways to interpret share price autocorrelation. Perhaps markets are efficient, so that a high price today is indicative of durable economic value. Perhaps markets are not so efficient, but investors nevertheless use yesterday’s price to determine the price at which shares will trade today.

Regardless of which story you believe, consider the situation of insiders and early investors in IPO firms. These investors face a “lock-up” period of three to six months after the IPO, during which they cannot sell. (This is intended as a kind of guarantee to new buyers that the shares are not total lemons.) If you hold a share of stock that you cannot sell for several months, you are better off, in a statistical sense, if the shares that you hold trade for $100 today than if they trade for $50 today. Sure, even after a pop, share prices could wither to worthlessness by the time the lock-up period ends. And sometimes that happens. But overall, if you are a preexisting investor in an IPO firm, the expected future value of your shares is substantially higher if your shares trade at $100 now than if they trade at $50.

So, if you are an early investor in a firm now making its debut, an IPO pop is mixed news. On the one hand, discovering that the shares you continue to own are very valuable is good news. On the other hand, if it is true that you could have sold the shares that you did sell for this much higher price, you’ve been screwed. On balance, how should a rational shareholder evaluate these conflicting signals? Should she be glad or disappointed? Should she fire her shoddy investment bank or celebrate its success?

If you are sure that stock markets are completely efficient and so share prices are independent of all the schmoozing and marketing done by your underwriter, then you should be outraged. You would have learned the same good news had you gone with a different investment bank, and your underwriter, if it were more competent or less corrupt, could have set a higher price and made you a great deal more money. But if you think that the stuff investment banks do when they underwrite an IPO actually does affect the price at which shares eventually trade, you might not be so angry. You might consider the “money left on the table” to be part of the fee you pay in order to be made as rich as possible.

IPOs are not all alike. In the lingo, they are sometimes “financing events” and they are sometimes “pricing events”. When IPOs are financing events, insiders are selling substantial fractions of their firms, trying to to divest their holdings or raise large sums for corporate purposes. When they are “pricing events” insiders are selling a small fraction of their shares in order to gain various benefits that come with being a public firm. In a “financing event”, when insiders are selling a lot of stock, the money left on the table from an IPO pop might amount to a substantial fraction of total equity value, too much money to be treated as a transaction cost. But in a “pricing event”, the money left on the table in a pop — the “opportunity cost of issuance” — may not be so large.

A very good predictor of how much an IPO will pop is “overhang”, the ratio of shares retained by insiders to shares sold during an IPO. IPOs with high overhang — that is, IPOs where insiders are selling only a small fraction of the firm — are much more likely to pop than IPOs in which investors are selling a lot of their shares. (This is true even controlling for the absolute value of shares sold, so it is unlikely to be just an artifact of scarcity.) To my mind, the explanation for this regularity is simple. Investment banks behave differently for high overhang IPOs (“pricing events”) than for low overhang IPOs (“financing events”).

For low overhang IPOs, in which much of the firm is being sold, underwriters go for accuracy. Investment banks do not want clients to spread word that they lost half the value of their firm to flippers on the big day. So bankers work to keep the IPO price and the immediate market price aligned. They try to set a reasonable price in the first place, they place shares with investors likely to sell pops and buy dips, they stabilize prices directly via their own activity in the market. (Underwriters have a partial exemption from market manipulation rules that allows them to “stabilize” new issues.)

But for high overhang IPOs, investment banks, in consultation with their clients, go for broke. The “book-building” process, often described as an anodyne sounding of investor interest, becomes an occasion to market the hell out of the issue. Investment banker activity, proxied by changes (even downward changes) to the planned issue price are predictors of IPO pops. For high overhang IPOs, underwriters and their clients agree that everything that can be done should be done to get the shares trading at the highest valuation possible, despite a necessarily conservative issue price.

When only a small fraction of a firm is being sold, issuers quite rationally permit investment banks to underprice their IPOs, because doing so aligns underwriters interests with their own. Issuers want their firms to be highly valued. An issuer who makes it clear that she will hold her underwriter accountable for underpricing is behaving foolishly, threatening to punish an outcome she desires. A smart issuer understands perfectly well that money left on the table will be used as kickbacks to favored clients of the investment bank. But why should she mind? She views that money as performance pay, a transaction expense. Given the small fraction of shares sold, it represents a modest cost. Further, she understands that the reason she chose her market-leading, high reputation underwriter is precisely because of the bank’s relationships with institutional investors, the bank’s ability to persuade people in its rolodex to take up and hold (not flip) new issues. If the issuer is not naive, she knows that the underwriter’s ability to place shares comes from plum deals the bank frequently offers the people in its rolodex. With the money she leaves on the table, the issuer is paying for exactly what she is trying to buy.

Efficient markets proponents will blanche at this whole scenario. How can underwriters affect share values? Surely, investment banks can’t “fool” the market over a six month lock-up period?! But nobody is fooling anybody, exactly. Nobody — not Warren Buffett, not the firm’s CEO, not even your psychic friend at Goldman Sachs — knows the “true value” of a speculative firm. A small rejiggering of earnings growth assumptions or the appropriate discount rate can double or halve estimates of “economic value”. The dirty little secret of fundamental analysis is that it can never tell you the correct price of a stock. Fundamental analysis can indicate that a price is wrong, that it is deeply below or outrageously above any reasonable valuation. But an independent analysis (one that ignores the market and estimates value based on a discount rate and expected cash flows) will very rarely approximate actual share prices (unless the analyst cheats, and reverse engineers the market). What issuers believe a good investment bank can do, with its marketing and its reputation, is get the shares trading on the optimistic end of the range of reasonable valuations. And that, to preexisting shareholders, can be much, much more valuable than a bit of money left on the table from underpricing.

So, is there any scandal here at all? I think so, but it’s not about investment banks screwing underwriting clients. On the contrary, I think investment banks usually serve both their underwriting clients and their favored investors pretty well. The scandal, I think, is that the IPO process offers issuers, underwriters, and favored investors too much and the rest of us too little. After the first day pop, IPOs tend to underperform other issues over the long term. Not by enough to reverse the first-day pop over the lock-up period. On average, new IPOs don’t underperform the market very much in the first six months after the pop. (During the 2000s, IPOs did perform poorly even in the first six months, but that is probably because the tech bubble crashed within 6 months of many IPOs.) IPOs get optimistically priced on their first day, and whoever winds up holding the shares from the end of the lock-up period and out several years pays for that. In general, buying IPOs at the issue price is a great deal, while buying IPOs on the secondary market is hazardous even a year or two after the offering. The IPO process ends up being a boon to insiders (the issuer, its underwriter, and favored investors), which is paid for over time by less connected investors who fail to demand a sufficient premium to hold recently IPO-ed shares.

In the scheme of things, this is pretty small beans. Caveat emptor and all of that. Still, a practice that taxes investors broadly in order to reward people for systematically mispricing securities does deserve some tut-tutting.

P.S. The academic literature on IPO underpricing is all about kickbacks. I prefer my conspiracy theories to be fringe, but this is all pretty mainstream. Famous explanations describe underpricing as a kickback necessary to induce uninformed investors to participate, or to induce informed investors to reveal what they know during the book-building process. For a real conspiracy theory, check out spinning, which has investment banks offering kickbacks to managers so that they’ll tolerate underpricing that screws their own firms’ shareholders.

Update History:

  • 4-June-2011, 4:25 p.m. EDT: Changed an “and” to a “so”, so it’d read better. Also change “And sometimes that does happen” to “”And sometimes that happens”.
  • 11-July-2011, 11:30 p.m. EDT: Replaced a “they” with “shares”, the intended but missing antecedent. Removed an ungrammatical “to”. No substantive changes.

13 Responses to “Should we be scandalized by IPO pops?”

  1. Nathaniel Graham writes:

    I had initially read Joe Nocera’s column at the NYTimes, laughed, and figured there was (still) no point in reading popular coverage of finance. I see he’s not the only silly columnist out there.

    Chris Clifford has pointed out that while there are a number of explanations for IPO underpricing, none of them are especially compelling. What strikes me is that I have never heard a coherent argument for a bias towards overpricing–more debates in finance aren’t so unidirectional.

    Given that the primary market is significantly different from the secondary, there is considerable (more than for established corporations) uncertainty regarding the ‘true’ valuation, and it is difficult to short IPOs, while we may not a single strong explanation for underpricing, I don’t see why we ought to be surprised by it. Frankly, if someone ever does completely explain it, I wouldn’t be shocked if the market did become efficient with respect to IPOs.

  2. Nemo writes:

    I wonder where LNKD would have priced had they offered the shares to everybody in a reverse Dutch auction (a la Google).

  3. David Merkel writes:

    Let the private shareholders sue over such matters — there is no public interest here.

    Small offerings of a hot stock tend to produce pops. Who should care? Selling shareholders should care, but they have the option of not selling. If they don’t like the price range, they can refuse to sell. There are no forced sellers.

    Given that they sell only a small portion of the shares in most IPOs of newly public companies, private shareholders are willing to give up something for price discovery. The first sale is small and later sales will be larger, at prices validated by the market.

    Beyond that, it is common practice to price the average IPO low — more so toward the beginning of an IPO wave, because it establishes a positive buzz about IPOs, making them easier to sell, until liquidity runs out.

    I don’t see much to complain about here on a public policy basis. Let the selling shareholders press a case that they were misled by the investment banks, and watch them lose.

  4. Max writes:

    Some IPOs are consistently overpriced: closed end funds. That is a real scandal, unlike this tempest in a teapot.

  5. Jacob writes:

    I like your take on the whole kerfuffle. To recast Blodget’s analogy, what you’re describing is a situation where your real estate agent helps you sell one of your 100 identical apartments for $1 million, and then the next day he resells it for $2 million. HOWEVER, the apartment wasn’t really worth $2 million. Yes, some schmuck was willing to pay that much using optimistic forecasts after being goaded by an agent, but a more reasonable estimate would be close to what you sold it for…$1 million.

    Therefore you’ve converted a portfolio worth $100 million into one with a market value of $200 million, at a cost of only $1 million in lost sales revenue…pretty clever.

    BUT WAIT…you’ve just dismissed market efficiency in order to justify the $2 million sales price. That’s quite right, markets aren’t efficient. But isn’t it then unreasonable to take the $2 million market price and extrapolate it to the rest of the portfolio? Aren’t we implicitly assuming efficiency if we take the $2 million market price and extrapolate a portfolio value of $200 million?

    Yes, prices exhibit autocorrelation, and yes, given that the market once inefficiently priced the asset at $2, it’s a good (but not unbiased) prediction to think that it will continue to do so. But if the best estimate of the portfolio’s fair value is closer to $100 million than $200 million, then it’s not at all reasonable to assume that the “pricing event” has in fact revealed the value of the whole portfolio.

    Insiders will benefit in the short term after they ditch their holdings at an inflated valuation, but if we sold off the rest of the overhang, smart money would be on a long-run lower price than is currently being observed.

  6. […] The last word on the IPO pricing “scandal.”  (Interfluidity) […]

  7. dave writes:

    It depends a great deal on the IPO, but at least for some of these mega deals that are turning a few tech nerds from ramen noodles to super rich I really don’t think they are overly concerned with getting maximum value. They just want things to go smoothly and to start being rich already, they are willing to overpay for that.

  8. Steve writes:

    Another reason that issuers may allow underwriters to “underprice” on offering is that most of the insiders’ shares were purchased at prices much lower than even the issuing price in the IPO. The insiders get a “pop” no matter what, assuming their shares are registered as part of the IPO or a follow-on registration.

  9. Former Capital Markets Guy writes:

    So an understanding of the economics of an investment bank points to another reason for underpricing. A typical IPO fee is in the range of 7%, but a big chunk of that gets paid to co-managers, lawyers and all the other fee-earners. So on a $100M IPO, the lead manager walks away with maybe $3.5-$4.0M in fees.

    But let’s say the deal pops 50% on the first day. The lucky hedge funds and other institutional investors who get that $50M on the first day are asked to (nothing in writing of course) kick back 30% of those first day profits back to the trading desk of the lead manager, in the form of trading commissions. So the lead manager of the IPO can take in up to $15M (30% of the $50M first day profit) in the form of trading commissions over the next few weeks. That’s how the lead manager really makes money on an IPO. (And that’s why you want to be the lead manager, rather than just a co-manager on the deal.)

    That’s how it really works my friends!

  10. […] Recessions and Recoveries: Job Changes (Economix) • Should we be scandalized by IPO pops? (Interfluidity) • Could Fast Food Automation Replace Low Wage Workers? (Econ Future) • Apologies – we need a […]

  11. erkie78 writes:

    So, I

  12. beowulf writes:

    Between the IPO pop and the 7% I-bank commissions (nice to see the travel agent business model is alive and well), what a tremendously inefficient way to allocate investment capital!
    So how do you disintermediate (bearing in mind Larry Lessig’s point that nothing will be reformed until after campaign finance is reformed)? Why not require Dutch auctions for IPO allocation? Hell, precede that with an auction where I-banks bid down their commissions to win underwriting contract.

    Its not really an economic problem, its a political one, the FIRE sector owns Congress. Its not an accident that Members of Congress are such ace investors (Here’s my Social Security reform plan, index benefits to market performance of investments held by congressmen and senators).

  13. […] Should We Be Scandalized By IPO Pops? […]