The term "pop" is sometimes used to describe the immediate post-IPO jump.
A small amount of "Pop" (no more than 10 percent) isn't a bad thing. It is a result of oversubscription, which every underwriter needs in order to assure they aren't left holding the bag. It also makes stabilization (the only legal form of market manipulation) less expensive for the underwriter.
You're going to see more extreme cases of "pop" in a market with a lot of uncertainty (like we have now). This is the underwriter being cautious. Their worst case scenario doesn't appear and the IPO turns out to be underpriced.
What we saw in the late 1990's was heinous. I worked at a startup investment bank (Epoch Partners) that was intended to take some of the pop out of IPOs (and make allocation more available to genuine retail investors).
Yes, but why care about underpricing at all? Can't you just have a formal auction, where everybody states (legally binding) how many stocks they want at which price level (e.g. for 3$ I'd buy 10 stocks, for 5$ I'd buy only 8 stock, and so on, basically giving your demand function) and then do a simple optimization that finds the highest price at which all shares sell. (Or alternatively, and perhaps better, the lowest price at which everybody who wants to pay at least this price, can buy.)
Why rely on guess work?
Edit: I saw on the linked Wikipedia article that some people have tried auctions. Google seemed a noteworthy example.
Google is of a size that requires distributed teams. It's hard to hire that many top engineers in one place, particularly given their laser focus on the top of the talent distribution. They pay a performance tax for having to distribute, but they get to marshall engineering resources most firms only dream of. And their culture is very consistent.
Did you notice that they were developed most powerful code-reviewing tools? That is an evolution of the pair-programming concept from an ortodox XP. ^_^
I think this is a great move on their part. My company uses oDesk. We have a large pool of contractors through them for human judgements, but also outsource some non-critical development to oDesk contractors.
I was thinking steganography too initially, but that's strictly information hiding. This "encrypts" it by XOR-ing it with random data. Without the "key" image, there is no information in the actual "data" image because it's literally just random data. In steganography the data is "discoverable", it's just hard to find; in this, it's impossible to derive the data without the key.
> But C is so hard to get right, that you will probably be able to use only the simplest algorithms in your C code
This is a gross exaggeration. It's not that hard to get C code right (C++ is a different story). I am unaware of any effort undertaken by skilled C programmers that failed because of limits C placed on algorithmic complexity. I am not arguing with your preference for higher level languages, just your statement that C is so difficult that it limits algorithmic expression.
C is substantially less compact and requires you to write code for things you get for free from other languages. Longer code takes more time to write and more time to read. Each feature or function point will, on average, take significantly longer to develop. On the other hand, a developer trying to write an OS in Python would also have some productivity challenges in other dimensions.
I am aware of the paradigmatic challenge C presents for many developers trained in the last 15 years. Trying to write in an OO style in C is neither fun nor advisable. Fortunately, most non-ui development is equally agreeable to other styles (although the developer may not be).
I'm not a C bigot and I like or love a number of high level languages (Python, Lisp, Haskell). I just don't think people should be afraid of C. Its closer-to-the-metal nature is an opportunity as well as a cost.
I agree. And I should have chosen different words. What you say is pretty much what I wanted to express.
The original comment said, that with Python you run into scalability problems earlier than with C.
And I wanted to add, that with C you run into (solvable but hard) `scalability' problems in terms of effort needed to cope with algorithmic complexity, much sooner. And more clever algorithms are often the key to solving scalability problems.
> And I wanted to add, that with C you run into (solvable but hard) `scalability' problems in terms of effort needed to cope with algorithmic complexity, much sooner.
I have certainly seen this effect. In retrospect, I wonder if this could be somewhat mitigated by real refactoring for C?
Perhaps. What also seems to work nice -- at least for me: Prototype in, say, Python, and then translate to C (either the hotspots or everything, in case you need to have a solution in pure C only).
There are many sources of stock for stock loan. 25 years ago, the primary source was stock in the margin accounts of investors. Today, nearly every large holder of stock loans it out. The reason they loan it out is to make more money.
Virtually every broker/dealer (b/d) that holds their own accounts has a stock loan desk (small to medium sized firms frequently have their accounts an another firm's books on a fully-disclosed basis). Virtually all large index fund managers have a stock loan desk as well. The stock loan desk at a b/d will loan stock to the firm's customers from the available shares (more on that in a moment) or it will find another place to borrow the shares from on behalf of the customer. This can involve looking in a system called Loannet or merely calling up other participants in the market.
The original source of available shares was the margin accounts of customers. The amount of stock available for loan depends on the amount of funds loaned to the customers. The stock loan activity is completely invisible to the customer whose account the shares are taken. Don't want your shares loaned? Don't use a margin account. Stock loan is the financing mechanism that provides the funds loaned to you for your margin account.
Securities can also be loaned from fully-paid (non-margin) accounts of customers with the written consent of the customer. It's a pain in the ass from a regulatory and operational point of view. It's usually only done if the customer has a really large holding in a hard to borrow stock. The customer generally negotiates a share of the revenue from the transaction.
In the past 25 years, institutional investors have started to loan stock as well. The pioneers were index funds, but it has spread to most other fund types. The big institutional investors generally set up their own desk and participate in the market directly. Being a direct participant can improve their ability to borrow stock as well.
The borrowing party puts up collateral (100-110%) for the stock and the lending firm either uses the funds to finance the margin business or puts it in a limited class of interest bearing accounts (I forget the name and regulation) at a bank. The borrowing party gets the stock and promptly sells it. The amount of the collateral is trued up to the value of the borrowed shares on a regular basis, so the risk to the lender is small.
So it is clear that one reason to loan the shares is financing. The second reason is revenue.
The revenue comes from the interest earned on the collateral. The interest on the collateral belongs to the lender except for a negotiated "rebate". For most stock, the rebate is generally 10 to 25 basis points less than the overnight benchmark (fed funds). The lender keeps what they can earn over the rebate.
Notice that I said "for most stocks". Some stocks can be hard to borrow. The supply can be low because large amounts of the stock are held in non-margin accounts or by investors who don't loan it out. The demand can be high because there is a large amount of short interest in the stock already.
The negotiated rebate on hard to borrow shares can be negative, and not just a few basis points. The negative rebate for a really hard to borrow can be negative 10 percent and worse. And a negative rebate means that you're paying somebody interest to hold your money as collateral.
This can be very lucrative for index funds based on a broad index like the Russell. It's one reason index fund fees are so low.
At the other end of the scale is generally available stock. Known as GC (general collateral), loan transactions in this stock are usually initiated by a stock lender looking for financing.
From a b/d point of view, stock loan is one aspect of a business called prime brokerage. Prime brokerage is a bundle of custody, operational, financing, and loan services offered to hedge funds.
One note, the perspective I've provided is largely from the institutional trading side of the b/d business. Retail investors borrowing stock will generally see a tier of rebates (I think the most common rebate is zero).
I worked on Wall Street starting in the mid 80s and with and on trading desks since the mid 90s. I've built the technology for equity trading desks and program trading desks. I was the business manager for a couple of program trading desks. I've created trading products (most notably a dark pool that mixed retail and institutional order flow).
I am a big believer that individual investors can't beat CAPM and shouldn't be picking stocks. I think the ideal portfolio is a mix of treasuries, equity and debt index funds, and angel investments.
"I am a big believer that individual investors can't beat CAPM and shouldn't be picking stocks."
First of all, hasn't CAPM been shown to be a nice, but not terribly accurate, picture of how the market works? Among other things, it assumes that asset returns are normally distributed random variables, it assumes that investors have homogeneous expectations about the return of an asset (aka everyone has the same info at the same time and observes the same risk and expected return of any particular asset), and it doesn't explain variance in stock returns. In other words, it doesn't accurately mirror reality.
Secondly, while I agree that most people shouldn't be picking stocks, my hair still bristles when I hear someone say that individual investors can't beat the market. True, little retail investors are probably at a disadvantage to the bigger, faster funds out there, but this hasn't stopped a handful of people from beating the market. <brag>I've averaged 28% annual returns over the past 12 years. Maybe you would say I'm just lucky though? </brag>
Well, by definition, half the investors (by portfolio size) beat the market, and the other half fail. And everybody assumes that they are in the top 50%, which is unlikely since the winners stay and losers cash out.
So most investors can't beat the market. I guess you could also look at the risks - betting a small amount (and finding out if you have the chops) could be a reasonable bet. Leaving large sums in the hands of someone else (as opposed to say an index fund) ... less good.
I didn't say that CAPM precisely modeled the market. I said that individual investors can't beat CAPM, meaning they can't beat the reward/risk relationship predicted by the model.
The model predicts that you can get higher returns by accepting greater variance on those returns. Even at moderate risk levels, it's easy to predict that some investors will happen to get the kind of returns you describe. Were your returns a result of your agency? Who can say.
Let's sidestep the issue of luck. I don't believe it is likely that your prior returns are a meaningful indicator of your future returns. But I could be wrong.
For a given asset, (Jensen's) Alpha is usually quite small compared to total return. How do you isolate the Alpha you identify?
What I quoted is after transaction costs. If I go off from Jan 1, 1998 when the S&P was 975.04 to Dec 31, 2009 when it was 1115.0, that means it has averaged 1.12% compounded annually.
