6 ms·

Its insane that you think this is the case. The markets are an incredibly important part of the economy and this technology does not only simply value a compan

Its insane that you think this is the case. The markets are an incredibly important part of the economy and this technology does not only simply value a company.

zzleeper 10y ago
I thought it was satire. Although you might make an argument for why reducing time-to-trade from 50ms to 100ms would not be the end of the world
- witty_username 10y ago
  That would make markets less efficient and thus allocate capital in a worse manner.
  - aleeds 10y ago
    Less efficient for literally 50 ms. Then it all balances out.
    - zeveb 10y ago
      The world GDP is about $76 trillion; that means that the global economy produces about $120,000/ in 50 ms. Reducing the time to work on a trade from 100 to 50 ms gives a trading algorithm that much of an edge, which ultimately helps increase the size of that $76 trillion pie.
- eej71 10y ago
  I didn't think it was satire as I believe it to be a popular viewpoint both in the larger world as well as among many folks here on HN.
jbob2000 10y ago
GS is an investment firm, their job is to find things that are under valued and cash in on the value left on the table. What else would they do with this technology? They aren't an agri-business, looking for a drought resistant strain of wheat. They're not a biotech company, folding proteins to cure cancer. They aren't an engineering firm, looking for a device to clean and desalinate water. They program computers to eek out another 0.1% on their portfolios.
- minipci1321 10y ago
  > What else would they do with this technology? Why, the same as all other hi-tech companies -- remove the human from the chain, replace them by the computers, automate processes. Will we still be able to speak about "greedy bankers" once it is all computers everywhere trading?
simonh 10y ago
Though even if all it did was value companies, a 1% improvement in accuracy on the NYSE would correctly allocate $200 Billion dollars of miss-allocated capital to do more work for the economy. And that's just in the US. Globally it could route half a trillion dollars into more deserving companies, creating new jobs, services and products and pushing down unnecessarily high prices. This is why a well functioning finance industry is so vitally important.
- Bartweiss 10y ago
  > a well functioning finance industry I'm fully prepared to argue we don't have that, though. Traders are throwing many millions of dollars at projects without even a pretense of improving market accuracy, or adding meaningful liquidity. Algorithmic trading hasn't been especially good for accuracy - we suddenly have flash crashes with markets dropping 30% and recovering, based on no change in fundamentals. High speed trading hasn't been much good for it - paying to move your trading servers slightly closer to a T1 backbone doesn't improve liquidity meaningfully, or accuracy at all. And we already have serious, systematic fraud: was the money Goldman Sachs spent buying insider leaks from the Fed really good for the industry? I take your point, and I don't begrudge finance its enormous economic role. But suggesting that the industry's growth in the last ~decade is about improving accuracy seems pretty questionable.
  - simonh 10y ago
    > But suggesting that the industry's growth in the last ~decade is about improving accuracy seems pretty questionable. Finance companies overwhelmingly make money by more accurately allocating capital whether that's channeling investment, arbitrage, executing a trade before the other guy, whatever. In every case that's money that went to a better place to do more economic work faster. It's what finance companies are for. Yes there's the occasional dodgy deal or rip off, but to read your post you'd think that is the only thing financial companies ever do. In reality trillions of dollars spent on research, technology, infrastructure, business growth and jobs got there via financing. Better financing means more and better quality of all of those things.
Bartweiss 10y ago
Honestly, there's still a point there. Having a well-priced, liquid stock market is obviously important. Mises even set it as his threshold for socialism - a country that has an independent stock market is not socialist. But that's not the same as appreciating the current state of stock trading. Sinking millions into millisecond improvements does not improve pricing. Identifying large trades and running pump-and-dump just before they clear doesn't improve liquidity. Running years-long fraud to get early leaks of Fed decisions helps no one. A lot of what happens in high speed trading today is actually harmful to the market as a pricing and liquidity tool. Insider leaks from the Fed weaken market confidence for everyone. Flash crashes are a new horror born entirely of algorithmic trading. The list goes on. I think it's entirely reasonable to appreciate free markets, even libertarian economic policies, and still consider the start of trading today alarming and detrimental to the economy.
zhte415 10y ago
Indeed it is, and indeed getting it all wrong because on is chasing 5 basis points on a quote from LIBOR, and creating synthetic upon synthetic based on this quote. The quote being a single analyst calling up a trading desk for a market rate, sometimes getting a reply, sometimes not, if getting a rate often from not the stated markets contact, and if not getting any response just inputting the rate from the day before. Organisation setting LIBOR was doing no wrong, those playing it were, but those depending on it diverting any accountability to a black box of interest rates which was clearly not a black box. Absconding selves of responsibility. Markets are indeed allocations of capital. But only when markets dont' get wound up in local optimals. As LIBOR. And bring havoc to everyone.

Its insane that you think this is the case. The markets are an incredibly important part of the economy and this technology does not only simply value a company.

zzleeper 10y ago

I thought it was satire. Although you might make an argument for why reducing time-to-trade from 50ms to 100ms would not be the end of the world

witty_username 10y ago
That would make markets less efficient and thus allocate capital in a worse manner.
- aleeds 10y ago
  Less efficient for literally 50 ms. Then it all balances out.
  - zeveb 10y ago
    The world GDP is about $76 trillion; that means that the global economy produces about $120,000/ in 50 ms. Reducing the time to work on a trade from 100 to 50 ms gives a trading algorithm that much of an edge, which ultimately helps increase the size of that $76 trillion pie.
eej71 10y ago
I didn't think it was satire as I believe it to be a popular viewpoint both in the larger world as well as among many folks here on HN.

jbob2000 10y ago

GS is an investment firm, their job is to find things that are under valued and cash in on the value left on the table. What else would they do with this technology? They aren't an agri-business, looking for a drought resistant strain of wheat. They're not a biotech company, folding proteins to cure cancer. They aren't an engineering firm, looking for a device to clean and desalinate water. They program computers to eek out another 0.1% on their portfolios.

minipci1321 10y ago
> What else would they do with this technology? Why, the same as all other hi-tech companies -- remove the human from the chain, replace them by the computers, automate processes. Will we still be able to speak about "greedy bankers" once it is all computers everywhere trading?

simonh 10y ago

Though even if all it did was value companies, a 1% improvement in accuracy on the NYSE would correctly allocate $200 Billion dollars of miss-allocated capital to do more work for the economy. And that's just in the US. Globally it could route half a trillion dollars into more deserving companies, creating new jobs, services and products and pushing down unnecessarily high prices. This is why a well functioning finance industry is so vitally important.

Bartweiss 10y ago
> a well functioning finance industry I'm fully prepared to argue we don't have that, though. Traders are throwing many millions of dollars at projects without even a pretense of improving market accuracy, or adding meaningful liquidity. Algorithmic trading hasn't been especially good for accuracy - we suddenly have flash crashes with markets dropping 30% and recovering, based on no change in fundamentals. High speed trading hasn't been much good for it - paying to move your trading servers slightly closer to a T1 backbone doesn't improve liquidity meaningfully, or accuracy at all. And we already have serious, systematic fraud: was the money Goldman Sachs spent buying insider leaks from the Fed really good for the industry? I take your point, and I don't begrudge finance its enormous economic role. But suggesting that the industry's growth in the last ~decade is about improving accuracy seems pretty questionable.
- simonh 10y ago
  > But suggesting that the industry's growth in the last ~decade is about improving accuracy seems pretty questionable. Finance companies overwhelmingly make money by more accurately allocating capital whether that's channeling investment, arbitrage, executing a trade before the other guy, whatever. In every case that's money that went to a better place to do more economic work faster. It's what finance companies are for. Yes there's the occasional dodgy deal or rip off, but to read your post you'd think that is the only thing financial companies ever do. In reality trillions of dollars spent on research, technology, infrastructure, business growth and jobs got there via financing. Better financing means more and better quality of all of those things.

Bartweiss 10y ago

Honestly, there's still a point there. Having a well-priced, liquid stock market is obviously important. Mises even set it as his threshold for socialism - a country that has an independent stock market is not socialist. But that's not the same as appreciating the current state of stock trading. Sinking millions into millisecond improvements does not improve pricing. Identifying large trades and running pump-and-dump just before they clear doesn't improve liquidity. Running years-long fraud to get early leaks of Fed decisions helps no one. A lot of what happens in high speed trading today is actually harmful to the market as a pricing and liquidity tool. Insider leaks from the Fed weaken market confidence for everyone. Flash crashes are a new horror born entirely of algorithmic trading. The list goes on. I think it's entirely reasonable to appreciate free markets, even libertarian economic policies, and still consider the start of trading today alarming and detrimental to the economy.

zhte415 10y ago

Indeed it is, and indeed getting it all wrong because on is chasing 5 basis points on a quote from LIBOR, and creating synthetic upon synthetic based on this quote. The quote being a single analyst calling up a trading desk for a market rate, sometimes getting a reply, sometimes not, if getting a rate often from not the stated markets contact, and if not getting any response just inputting the rate from the day before. Organisation setting LIBOR was doing no wrong, those playing it were, but those depending on it diverting any accountability to a black box of interest rates which was clearly not a black box. Absconding selves of responsibility. Markets are indeed allocations of capital. But only when markets dont' get wound up in local optimals. As LIBOR. And bring havoc to everyone.