3 ms·

Let's say I'm not interested in spreads or liquidity beyond being able to within a day or two purchase or sell a share that pays dividends on company earnings.

Let's say I'm not interested in spreads or liquidity beyond being able to within a day or two purchase or sell a share that pays dividends on company earnings. What would make me want to go with a market that offered lower spreads and more liquidity? How would it affect a long term holder? Isn't there a cost to all of these transactions occurring? Like, why do I need the spread to be as small as possible? How much is actually saved? How much are the infrastructure costs for maintaing a marketplace that supports these high frequency trades? Isn't there room for a much simpler, low frequency marketplace that had fractions of the operational costs? Couldn't savings be passed on to retail investors with a cheaper marketplace?

tptacek 13y ago
Why would you go with a market that offered lower spreads? What a strange question. You'd pick the market that offered the lowest cost of trades, which absent some other fact you haven't named, is the one with the lowest spreads. There is the opposite of a cost to normal investors of "all these trades" occurring. The volume of trades makes the pricing of stocks more competitive and increases the odds that when you want to trade, someone who agrees with you about the price of that stock will be available to take the other side of the trade. This discussion keeps waving the term "spread" around like a totem, but in reality the term just captures that paragraph.
- williamcotton 13y ago
  The infrastructure required to offer incredibly low latency trades and super high volumes seems like it would cost a heck of a lot of resources. I'm not convinced that all of these marginal gains makes up for these costs. I don't get the jump from describing an illiquid real estate market to this hypermarket with millions of trades a second. Are you telling me that it is a linear function? We will always need more trades, and more volume? What about the underlying economic activity? How does this follow any sort of market fundamentals? Are you really saying there is no upper limit? How could that make any sense at all?
  - tptacek 13y ago
    No, there's surely a point at which the marginal utility of additional liquidity stops justifying the expense of providing it. We may even be far past it. But even if you stipulate that we are, it doesn't follow that interventions to suppress trading will do more good than harm, because we also know that prior to the advent of electronic trading, we had "not enough" liquidity. Surely the current system is imperfect in many ways, and I'm sure Nash has something to say about the competition among market makers of varying degrees of competence.

tptacek 13y ago

Why would you go with a market that offered lower spreads? What a strange question. You'd pick the market that offered the lowest cost of trades, which absent some other fact you haven't named, is the one with the lowest spreads. There is the opposite of a cost to normal investors of "all these trades" occurring. The volume of trades makes the pricing of stocks more competitive and increases the odds that when you want to trade, someone who agrees with you about the price of that stock will be available to take the other side of the trade. This discussion keeps waving the term "spread" around like a totem, but in reality the term just captures that paragraph.

williamcotton 13y ago
The infrastructure required to offer incredibly low latency trades and super high volumes seems like it would cost a heck of a lot of resources. I'm not convinced that all of these marginal gains makes up for these costs. I don't get the jump from describing an illiquid real estate market to this hypermarket with millions of trades a second. Are you telling me that it is a linear function? We will always need more trades, and more volume? What about the underlying economic activity? How does this follow any sort of market fundamentals? Are you really saying there is no upper limit? How could that make any sense at all?
- tptacek 13y ago
  No, there's surely a point at which the marginal utility of additional liquidity stops justifying the expense of providing it. We may even be far past it. But even if you stipulate that we are, it doesn't follow that interventions to suppress trading will do more good than harm, because we also know that prior to the advent of electronic trading, we had "not enough" liquidity. Surely the current system is imperfect in many ways, and I'm sure Nash has something to say about the competition among market makers of varying degrees of competence.