Stock Market Sell-Off Might Stem From Trader's Fat Finger

s122604 points out a CNBC story according to which "the catalyst for today's extraordinary price swing (at one point the Dow lost almost 9 percent in less than an hour) may have been because a trader entered a 'B' for billions instead of an 'M' for millions on a trade of Procter and Gamble: 'According to multiple sources, a trader entered a "b" for billion instead of an "m" for million in a trade possibly involving Procter & Gamble, a component in the Dow. (CNBC's Jim Cramer noted suspicious price movement in P&G stock on air during the height of the market selloff).' Unbelievable there are no safeguards to protect against this."

Comments

Slashdot, you missed the software part!

[Wrexs0ul]

CBC Story about software controls for selling on the market: http://www.cbc.ca/money/story/2010/05/06/tsx-markets.html

Nuts to fat finger keyboards, there are automated software controls in the industry that caught-on to the sale and snowballed this individual's mistake into something really big. The issue wasn't just in this guy's mistake, but the fact that potentially billions of dollars changed hands because of a trust relationship these systems have with market indicators.

Not that there's anything wrong with that: on a good day this could protect big firms from being the guy caught holding the bill, but I think we've discovered where the next upgrade in broker software might be :)

-Matt

Protections may be bypassed...

[tlhIngan]

So you implement some protection. Then some prima donna trader comes by and asks that they be disabled and his trades unquestioned. If the company makes good profit off the guy, down the protection goes.

Reminds me of this story on a commodities trader that not only didn't close his position, but actually ended up taking physical delivery of the commodity. Oops. Sure there were protections, but the guy had them disabled.

http://thedailywtf.com/articles/special-delivery.aspx

Hell, for all we know, this is exactly what happened - most traders can't enter in a "b", except a succint few well-trusted individuals. Just one of the "gods" managed to fumble it.

Institutional Traders Don't Enter Trades Like That

[Knara]

It may have been a system problem, that's quite possible. But institutional traders don't type in "b" or "m" next to some number they type in of stock they want.

But even in some strange world where they did, entering in a standard lot quantity that required an "m" (much less a "b") for the stock that is suspected to be the issue at hand (PG), would result in an order that exceeded the 30-day avg vol for PG by a factor of 10.

And that's not even considering that the firm's risk management would, in theory, have caught the issue already.

I am, obviously, doubtful of this explanation.

Market correction due and it's here!

[tekrat]

They've been saying for some time the market was due for a correction. Mind you, at the height of the financial meltdown, the Dow was at 6500, and has almost doubled value in about a year, it was rising too fast considering that the recovery still really hasn't come (i.e., there are still no jobs).

The only people making money are the same ones that are always making money -- the fat cats. Now it looks like the market will correct, and probably stablize around 10k, maybe 9. And even more people will lose their jobs and the cycle will continue until America admits that it is bankrupt.

Then will come some really hard times, but, once we address the real issues plauging the country, we'll come out of it stronger. But first, we need to start getting rid of all the lawyers....

I'm surprised, somewhat.

[fuzzyfuzzyfungus]

Not that there isn't some finance-clippy that pops up and asks "You appear to be tanking the Dow, would you like help with that?", or that people are allowed to do whatever stupid shit they want with the assets they have(the amount of stupid shit that people are allowed to do with assets that they don't have is somewhat concerning, however).

However, I am somewhat surprised that the guys who do UI design for financial systems don't design systems to make things like power-of-ten or million/billion errors very difficult. Having a 3 factors of 10 difference be just one key away(and phonetically not all that dissimilar) seems like a mess waiting to happen.

I've seen in doctor's offices(and I know pharmacists and pharmacy techs, especially ones where compounding and other tougher than "dispense stock pill" type activities go on get drilled hard on this) outlining acceptable and unacceptable notetaking protocols to reduce the risk of power-of-ten dosing errors(things like ".2 is wrong, there should always be a leading zero to clue you in to the decimal point, use 0.2.") Some of them are even domain specific conventions, specifically trading off other factors in favor of reducing the risk of error. In science, for instance, saying 2.0, or even 2.0000 if you have that much precision, instead of 2 is a good thing. It tells your reader how precise the value they are looking at is. In prescriptions and medical notes, "2.0" is dangerously close to "20", and is thus avoided.

One would think that, even if it meant making up arbitrary symbols, or using UI element sizes to convey magnitudes, or something, financial UIs would adopt a similar set of domain-specific tricks to head off the most common and dangerous errors.

Re:SELL!

[tgatliff]

It doesnt take a genius to figure out that the "typo" theory is BS... In 2008, it was a "computer fault"... Deflation is still very much in control at the moment, and it appears that we have only delayed it. As greece and many other sovereigns start to default on their debts, we will see the leg down... Acceptance is a b&^%*& sometimes...

Re:SELL!

[name*censored*]

It reminds me a little of a throwaway comment Stephen Hawking made in the recent series Into The Universe With Stephen Hawking - he was asked not to speculate on the end of the universe in a certain lecture series for fear that it would affect the stock market. Really? Even if the universe was going to end in our lifetime, and no-one had noticed before now (oops), what kind of fool would hear the news and immediately worry about his or her stock portfolio? What are you going to do with your money after the universe ends? You would think (if people behaved rationally) that the stock market would grind to a halt when every trader says "Screw this, I haven't got much time left and I'm not going to waste it here".

This should happen EVERY DAY, would be good

[lanner]

You read that headline right. This should happen ALL THE TIME. It would be good for the markets.

Speculators would be driven out, or driven insane. Emotionally driven traders would have heart attacks.

Sound judgments made based on factual data would not be affected.

Next week, people like me won't give a toot that this ever happened. However, a lot of day traders just pooped their pants. I'm buying men's underwear stocks.

The person who made the mistake will be punished dearly.

I think you guys are missing the actual point

[Anonymous Coward]

What's being talked about here isn't the general decline in the market today, but a very suspicious "blip" that occurred in a huge number of stock prices at 2:45 EST, followed by immediate recovery.

Look at the blip:

Adobe
Google
Westlake Chemical
Cabela's Incorporated
Apple
Microsoft
Titanium Metals
Fidelity IIS

This shit is across the board, with very few exceptions. You try explaining how something like that happens apart from some major fuckup somewhere.

look at the volume!

[je ne sais quoi]

Your comment is spot on. Look at the volume of shares traded for PG today. There is no statistically significant spike in volume today that correlates with the price drop. If the sell was staggered, the price drop should have been staggered. Since it isn't, either Google's volume is way off or this story is a crock. Based on the volume data, the sell-off started well before the major drop in stock price.

I suspect that something funny did happen though, in TFA they are quoting that PG was trading down at $30 per share at some point, so something definitely slipped. Fortunately, we managed to avoid another Black Monday, where the DOW went down and stayed down.

Correct, but also incorrect

[spun]

It wasn't a typo, it was Bernie Sanders speaking for an hour on the Senate floor today, pushing for a bill to audit the Fed. Everyone who is anyone knows what we will find if we audit the Fed, and it isn't good. Not just for us, but for the world. Which is why Obama threatened to veto this bill, citing national security. The dollar is the world's reserve currency. If all the plebeians of the world found out how utterly worthless our currency is, we would suffer a crash that would make the last one look like a cake-walk.

As for Greece, though, that crisis is actually pushing investors back to America.

Re:Correct, but also incorrect

[Antisyzygy]

What will we find? I suspect if we audited any government branch or large private company it would trace all expenditures to some rich asshats pocket.

Re:Correct, but also incorrect

[betterunixthanunix]

It is a matter of faith in those dollars, not simply how many of them are in circulation. If auditing the fed is considered a problem, that suggests to me that there is some faith-shaking problem with our economy...

Re:Correct, but also incorrect

[Sycraft-fu]

The problem is there are a whole lot of people who know just enough about economics to understand that money has nothing backing it, but not enough to understand that really, that's how money has generally always been. For some reason they see gold as a magical substance that cannot lose value, and that if we just had that behind currency there'd be no problems. This ignores, of course, the Great Depression, when currency was on the gold standard (and some would argue the inflexibility of it helped create the depression). They can't separate the large amount of value gold has due to its use as a financial reserve with the smaller value it has for industrial uses.

So they have no faith in current currency, but think that a currency backed by metal would be worth something. They don't understand that money is just a theoretical construct that facilitates trade.

It is one of the reasons that I think a basic economics course should be mandatory in high school. Too many people have their own half-assed ideas about how things work. Understanding the very basics is important, fundamentally that everything is just trade.

Re:Correct, but also incorrect

[twostix]

You've created a strawman, attributed it to "they" and seem to be completely oblivious to the actual argument, and certainly nobody say there would be "no problems", they say it would constrain the government to only spend money that the country actually has right now.

When money is backed by something physical it prevents the government from spending more than it has by printing (or adding ten zeros) to its bank account whenever it wants, therefore devaluing the money that is already in existence.

You can't print gold so the government has to try a lot harder to spend money it doesn't actually have.

That's the real point of a hard money currency system, the fact that you hear gold (and silver) all the time is that it's just a convenient, historic store of wealth, if there was something more convenient the argument would be made for that.

Also you conveniently leave out the fact that it was while on the gold standard that the US, the UK and France became world powers, and once they went to fiat currencies became mired in debt and devaluation and lost their prosperity and declined.

It's not about money, it's about an inbuilt restriction to preventing politicians writing blank cheques until the country is broke, like most western countries are - as we see the beginnings of with Greece, Ireland, the UK and soon The US if it doesn't get its spending under control and cut the services it can't afford.

Finally you make it out like the current Fiat money system has proved itself superior, yet every single fiat currency in history has imploded at around 50 years due to unrelenting printing. You do realise it's only been implemented in the US for 38 years and that government spending has been increasing exponentially over that time compared to revenue? Do you think that can continue forever? Or do you think they'll get to a point and say, that's enough spending?

Re:Institutional Traders Don't Enter Trades Like T

[khallow]

Financial engineers as a whole are a bunch of Dilettantes. They literally play guessing games disguised with fake knowledge. Any scientist would look at the markets as an optimization or stochastic problem. Not financial engineers. They look at indicators that have minimal mathematical basis and "psychological" levels.

They're also damn good at what they do. No offense to scientists, but anyone trading using the scientific method is just going to be giving money to people who use more effective methods. The simple explanation is that the scientific method is far from optimal for the problem of rapidly evaluating the price of a security in real time. Market trading also isn't an optimization or stochastic problem. Those are approximations for the real deal. As I see it, a seat-of-your-pants market maker is going to know more and make better trades.

Pardon me while I make a brief appeal to authority here. I have a PhD in math. It's not in financial mathematics, but I'm acquainted with what they do here. The math/computation part is in getting a good estimate of what things are worth and how they correlate with other securities over certain time scales. It enables the trading of complex derivatives and execution of automated strategies (especially hedging and arbitrage related trading). IMHO, there's no magic math algorithm that will trade well understood securities far better than current methods. That vein is probably almost mined out. There might be something there, but I doubt it. The current play seems more in those complex derivatives.

I see a lot of the current problems more as social engineering problems. For example, I bet every single bank and investment firm that collapsed in 2008 had incentives (and lack of accountability) in place for the traders and managers to accumulate highly leveraged risk. Guessing right on a highly leveraged strategy can get you excellent bonuses. A prudent strategy can lose you your job, even if you are right in the end. The last of the outcomes, guessing wrong with highly leveraged risk just loses your job again (the company might go belly up as well, but it's not your problem any more).

As I see it, the fundamental problem with most such businesses (and most publicly traded companies as a whole) is simply that the owners do not run the business. The people making the decisions risking the capital are completely divorced from the owners of the assets. The decision-makers only stand to lose their jobs.

Re:SELL!

[moeinvt]

" . . . in a matter of minutes the market had recovered . . ."

Yes, but think of the people in the following situations:

1. Had stop-losses in place
2. Got caught on the wrong side of "put" options at the wrong time
3. Had margin requirements and were automatically liquidated from their positions to meet those requirements

This whole thing stinks.

Re:SELL!

[zippthorne]

If you're worried about that kind of collapse, then yes, gold would go to zero, too. What can you do with gold other than look at it? You think you're going to arrange a deal with the typical large industries that actually put gold to a useful purpose? After a massive currency collapse? Good luck!

Baskets of metals are ok as a hedge, though all bets are off in a real crisis, and there is so much more out there you could be involved in that doesn't involve getting ripped off by a sketchy company with cheesy ads.

Re:SELL!

[fritsd]

I've never had economics in school, but I'll bite..

But it's worse than stocks, bonds, deeds, and a host of other things.

And why's that? It must be because you assume that your stocks, bonds, deeds etc. increase in value with time, whereas gold does not and costs money to store it safely. Am I right?
Gold is an inert metal that doesn't wear down with age (unlike the Dutch tulip bulb craze ;-)). This is a fact.
The paradigm that in general stocks increase in value with time is *not* a fact. It is an assumption based on past behaviour, explained by a world-wide exponential growth in economy, as we got more people, the people extracted more raw materials (including non-renewable and non-recyclable materials such as oil) from our planet's crust, and those resources in turn fueled the economy and consumption rate.

Our planet is an open system in the sense that we get free sunlight from our sun and ultimately this is the power source for all economic growth (except nuclear fission and, as mentioned in another discussion, tidal :-) ). But besides the sunlight, Earth is a closed system, and we're probably using up too much of the long-term concentrated chemical potential energy stores (oil) and should switch to weaker, more direct energy sources (solar photovoltaic, solar heating, wind, biogas, biodiesel) to sustain our civilisation longer-term than just this century.
How can you believe to have permanent exponential growth of power resources / manufacturing / consumption in a closed system? IT DOESN'T MAKE SENSE.

Listen to this:

• The greatest shortcoming of the human race is our inability to understand the exponential function Albert A. Bartlett, physicist
• Anyone who believes exponential growth can go on forever in a finite world is either a madman or an economist. w:Kenneth Boulding, economist

In a steady-state economic system, solar energy input == economic production. Your stocks and bonds are valuable but will not increase in value (I don't see how?). I haven't found any economics articles that I could understand about what such a world would look like, but the transition is probably painful (I'm not a doomer btw).

Just because Malthus (1798) and the Club of Rome (1972) were impopular doomers doesn't make them wrong. Our generation and the next few must *prove* them wrong.
For any serious (and really not always doomer-like :-)) discussion of the discrepancies between economy and reality, I refer you to The Oil Drum forum, where people much smarter and possibly more clear-minded and rational than me discuss these things :-).