4 ms·

I personally am against selling equity that you don't own. It creates so many failure modes for the equity markets (failure to deliver chief among them). And I

I personally am against selling equity that you don't own. It creates so many failure modes for the equity markets (failure to deliver chief among them). And I wouldn't be surprised if the net impact of short selling were actually positive for spot price, because shorts tend to sell into strength but they often squeeze into weakness. We've gotten accustomed to short selling and buybacks, but the equity markets would make far more sense if we were to outlaw both practices. Borrowing something to sell it is a recursive behavior, as is buying shares of yourself. In markets as in computer programming, we have to be careful about recursive behaviors.

gruez 5y ago
>It creates so many failure modes for the equity markets (failure to deliver chief among them). Has this actually caused issues? Stock prices usually go up when the economy is doing well, so when short sellers default the economy/banking system is well prepared to absorb the impact. This is as opposed to something like MBS which causes a downward spiral of "people losing their jobs -> default on mortgage -> banks pull bank loans -> business spending drops -> people losing their jobs". >the equity markets would make far more sense if we were to outlaw [buybacks]. Why? Buybacks are equivalent to paying dividends and then reinvesting them (which most people do).
- totalZero 5y ago
  > Has this actually caused issues? That was one of the issues for the clearinghouse with GME. When a market-maker sells short, they don't have to locate inventory prior to selling. Sometimes, they are unable to deliver to the clearinghouse, which technically forces the clearinghouse to remain short to the buyer. A "buy-in" is supposed to take place when a short seller cannot deliver, but this doesn't happen with market-makers because they get extra time under the rules, and also they tend to trade so much that the clock resets on their past undelivered shorts every time they buy and sell (this is called a "reset transaction" and it is illegal for any other market participant to do). If a market-maker gets caught wrong-way on a position where they sold lots of shares that they neither own nor have the solvency to buy at market, they can collapse and that puts the clearinghouse itself at risk. This is why a firm like Citadel (one of the biggest market-makers) had an incentive to capitalize Melvin Capital with $2B to attenuate a squeeze. IMO it was self-preservation, not charity nor opportunity. The SEC makes FTD data available twice monthly. You can find it here: https://www.sec.gov/data/foiadocsfailsdatahtm https://www.sec.gov/data/foiadocsfailsdatahtm MBS wasn't a problem because of the defaults themselves. It was a problem because the correlation of defaults was underpriced, so banks took heavy losses on positions that they thought were pristine. > Why? Buybacks are equivalent to paying dividends and then reinvesting them (which most people do). This isn't true. Buybacks do two things that make them highly attractive to executives. First, buybacks allow option holders (including many insiders) to generate capital appreciation on those options without exercising them. Dividends, on the other hand, distribute only to shareholders and actually reduce the price of the stock (share price drops by the amount of dividend on ex-date). Second, buybacks provide a perpetual bid to the stock that mitigates the downward impact of selloffs, which is particularly nice for executives that receive performance incentive bonuses based on share price. With something like an ASR (a more complicated buyback traded against a derivatives desk), buying may even get more aggressive (in terms of % of total volume) as the stock declines. Buybacks are rationalized as capital return but really they are a form of inducing asset inflation.
sigstoat 5y ago
> Borrowing something to sell it is a recursive behavior, as is buying shares of yourself. no, it isn't. this is nonsense, you're just trying to equate some reasonable market activities with something that sounds a bit scary and weird because you don't understand it. > ... in computer programming, we have to be careful about recursive behaviors. no, we don't. no more than of any other failure to terminate a loop.
- totalZero 5y ago
  I'll ignore the ad hominem parts of your comment. If I borrow a share and sell it, I can later borrow it again and sell it again. That is recursion. When it hits a base case -- eg, I get liquidated -- all of those shorts have to be simultaneously covered. Doing so can create a stronger bid than the float of the asset. > no, we don't. no more than of any other failure to terminate a loop. A loop that can be done in-place, or in constant space, is less space-complex than a recursive function that cannot. Surely you have had situations where writing a top-down recursive function requires more memory than achieving the same result with a bottom-up recursive function, or vice versa. That doesn't happen with an iterative loop where I can better characterize the space required a priori. There is no rule in Regulation SHO preventing market-makers from selling a greater number of shares than the float, and there is no rule preventing any short-seller from borrowing and selling the entire float multiple times over.
  - fighterpilot 5y ago
    This recursion concern is a theoretical one. It's exceedingly rare in practice for the short float to exceed the public float, and there's an inbuilt negative feedback which stops this from happening. The larger the short float %, the lower EV shorting becomes, ceteris paribus, and eventually the marginal benefit becomes negative and no rational actor would add to a short. That's one of the reasons (among others) why we almost never see a short float above 100% of the public float. Additional regulations aren't needed for this, because it's not a problem.
    - totalZero 5y ago

gruez 5y ago

>It creates so many failure modes for the equity markets (failure to deliver chief among them). Has this actually caused issues? Stock prices usually go up when the economy is doing well, so when short sellers default the economy/banking system is well prepared to absorb the impact. This is as opposed to something like MBS which causes a downward spiral of "people losing their jobs -> default on mortgage -> banks pull bank loans -> business spending drops -> people losing their jobs". >the equity markets would make far more sense if we were to outlaw [buybacks]. Why? Buybacks are equivalent to paying dividends and then reinvesting them (which most people do).

totalZero 5y ago
> Has this actually caused issues? That was one of the issues for the clearinghouse with GME. When a market-maker sells short, they don't have to locate inventory prior to selling. Sometimes, they are unable to deliver to the clearinghouse, which technically forces the clearinghouse to remain short to the buyer. A "buy-in" is supposed to take place when a short seller cannot deliver, but this doesn't happen with market-makers because they get extra time under the rules, and also they tend to trade so much that the clock resets on their past undelivered shorts every time they buy and sell (this is called a "reset transaction" and it is illegal for any other market participant to do). If a market-maker gets caught wrong-way on a position where they sold lots of shares that they neither own nor have the solvency to buy at market, they can collapse and that puts the clearinghouse itself at risk. This is why a firm like Citadel (one of the biggest market-makers) had an incentive to capitalize Melvin Capital with $2B to attenuate a squeeze. IMO it was self-preservation, not charity nor opportunity. The SEC makes FTD data available twice monthly. You can find it here: https://www.sec.gov/data/foiadocsfailsdatahtm https://www.sec.gov/data/foiadocsfailsdatahtm MBS wasn't a problem because of the defaults themselves. It was a problem because the correlation of defaults was underpriced, so banks took heavy losses on positions that they thought were pristine. > Why? Buybacks are equivalent to paying dividends and then reinvesting them (which most people do). This isn't true. Buybacks do two things that make them highly attractive to executives. First, buybacks allow option holders (including many insiders) to generate capital appreciation on those options without exercising them. Dividends, on the other hand, distribute only to shareholders and actually reduce the price of the stock (share price drops by the amount of dividend on ex-date). Second, buybacks provide a perpetual bid to the stock that mitigates the downward impact of selloffs, which is particularly nice for executives that receive performance incentive bonuses based on share price. With something like an ASR (a more complicated buyback traded against a derivatives desk), buying may even get more aggressive (in terms of % of total volume) as the stock declines. Buybacks are rationalized as capital return but really they are a form of inducing asset inflation.

sigstoat 5y ago

> Borrowing something to sell it is a recursive behavior, as is buying shares of yourself. no, it isn't. this is nonsense, you're just trying to equate some reasonable market activities with something that sounds a bit scary and weird because you don't understand it. > ... in computer programming, we have to be careful about recursive behaviors. no, we don't. no more than of any other failure to terminate a loop.

totalZero 5y ago
I'll ignore the ad hominem parts of your comment. If I borrow a share and sell it, I can later borrow it again and sell it again. That is recursion. When it hits a base case -- eg, I get liquidated -- all of those shorts have to be simultaneously covered. Doing so can create a stronger bid than the float of the asset. > no, we don't. no more than of any other failure to terminate a loop. A loop that can be done in-place, or in constant space, is less space-complex than a recursive function that cannot. Surely you have had situations where writing a top-down recursive function requires more memory than achieving the same result with a bottom-up recursive function, or vice versa. That doesn't happen with an iterative loop where I can better characterize the space required a priori. There is no rule in Regulation SHO preventing market-makers from selling a greater number of shares than the float, and there is no rule preventing any short-seller from borrowing and selling the entire float multiple times over.
- fighterpilot 5y ago
  This recursion concern is a theoretical one. It's exceedingly rare in practice for the short float to exceed the public float, and there's an inbuilt negative feedback which stops this from happening. The larger the short float %, the lower EV shorting becomes, ceteris paribus, and eventually the marginal benefit becomes negative and no rational actor would add to a short. That's one of the reasons (among others) why we almost never see a short float above 100% of the public float. Additional regulations aren't needed for this, because it's not a problem.
  - totalZero 5y ago