Can someone help me out here? I’ve been wanting to reduce my market exposure, and to me, the most sensible approach is to balance portfolio weights so that the weighted average of beta for assets = 0 (vs S&P 500). If this is achieved, then the portfolio returns should be independent of the return on the S&P500. But many people talk about being “dollar neutral”, which is when the dollar value of longs = dollar value of shorts. If you have more than two assets, you can, in theory, be both dollar neutral and beta neutral. However, what is the advantage of being dollar neutral? If you had two assets and they were dollar neutral but not beta neutral, how would this affect things? As far as I can tell, dollar neutral is perhaps a “poor man’s substitute” for beta neutral (meaning that you don’t have to do as much math to get there), and that it means that you are neutral for small changes in the S&P, but not for larger ones. Is there anything else?
Sorry, one more related question. I see some ETFs that are designed to have a beta=2. This makes them substantially more volatile because they are leveraged. The main advantage I can see is that if I short them, it will take about 1/2 as much money in a margin account to achieve a Beta=0 portfolio as it would to short a regular index ETF. But how does this implied leverage affect the risk? As an individual investment, I can see how the beta makes the fund riskier than a regular index fund, but in a portfolio context, it seems to be an efficient risk-reducer, assuming a goal of beta=0. Am I analyzing the leverage risk correctly?
Dollar weighting is usually a bad idea but it comes from a management problem - there are oodles of not especially competent managers who “discover” that there isn’t a unique beta so when you become beta neutral, well that’s just the opinion of some bonehead quant. Now dollar neutral, that’s something real.
Got it! Thanks Joey. Makes a lot of sense. How about the leverage question? If you are trying to get to beta=0, assuming some consistent definition of beta, is it riskier to do it with an index ETF that is implicitly leveraged 2x than it is to do with a regular index ETF? The only risk I can think of is that it will take a smaller movement to generate a margin call, but I think I’ll have enough dry powder to cover that, and the leverage isn’t all that enormous (not like a future, for example).
My old man is in a hedge fund that I’m guessing is dollar neutral. They play long/shorts with a pair of companies, long an outperformer, short an underperformer in an industry, and play the spread as the market moves. The theory being in an up market the outperformer will rise more than the underperformer, and conversely in a down market the underperformer will drop more, and more gains accrue to the short position than loses on the long. It sounds good on paper… I suppose it comes down to how well you can pick your pairs. By the way, on the TSX at least (in my neck of the woods) they offer levered ETF’s that go against the market. Don’t know if they have this anywhere else, it’s a relatively new product here, but you could potentially do something with that if you have it.
I’ve seen some inverse index ETFs and levered index ETFs (see ProShares short and UltraShort: symbols SH and SDS, respectively). I’m not 100% sure on which exchange, but one could check. These have some nice qualities, like not having margin interest or worries about margin calls. The management costs are higher though, perhaps to account for dividend payments. The interesting question is what would the difference be between shorting an index ETF and buying a short index ETF on margin. I suspect the only difference is that you don’t need to worry about making dividend payments with the short fund. Otherwise, the difference would be that buying an inverse index ETF will eat up more of your cash than shorting a regular index ETF.
Dimes27 Wrote: ------------------------------------------------------- > My old man is in a hedge fund that I’m guessing is > dollar neutral. They play long/shorts with a pair > of companies, long an outperformer, short an > underperformer in an industry, and play the spread > as the market moves. The theory being in an up > market the outperformer will rise more than the > underperformer, and conversely in a down market > the underperformer will drop more, and more gains > accrue to the short position than loses on the > long. It sounds good on paper… I suppose it comes > down to how well you can pick your pairs. > > By the way, on the TSX at least (in my neck of the > woods) they offer levered ETF’s that go against > the market. Don’t know if they have this anywhere > else, it’s a relatively new product here, but you > could potentially do something with that if you > have it. Dollar neutral pairs trading would really irk me. If a fund does that then they can express overall market views in their trades. I definitely don’t want to pay anyone 2/20 to bet that the stock market will go up or down.
bchadwick Wrote: ------------------------------------------------------- Otherwise, the > difference would be that buying an inverse index > ETF will eat up more of your cash than shorting a > regular index ETF. I’ve never shorted a stock, but I think it would be more or less the same. On the TSX you would have to cover 150% of a short less proceeds. That would be the same as the maximum margin loan of 50% when buying. JoeyD, can you explain what you mean by “If a fund does that then they can express overall market views in their trades.” For what it’s worth, my old man is a little skeptical as well.
Dimes27 Wrote: ------------------------------------------------------- > On the TSX you would > have to cover 150% of a short less proceeds. Only if you are doing it with no money in your retail brokerage account where they really like it when people go short. > JoeyD, can you explain what you mean by “If a fund > does that then they can express overall market > views in their trades.” For what it’s worth, my > old man is a little skeptical as well. So when you invest in a hedge fund, you need to be really careful that you are taking on a risk that you can’t take on somewhere else much more cheaply and easily. If I was doing due diligence on a “dollar neutral” pairs trading shop, they would need to do a lot of convincing (it wouldn’t work though because they don’t have a quant on staff). Pairs trading is tough - the idea is to eliminate all variance except that due to the specific company. Dollar neutral on the pair ignores the leverage of the companies and possibly other aspects that cause them to have different betas. So if I’m always long the more levered of the pair, I have a bullish bet and vice versa. It’s really easy to have your backtesting do this to you even if you don’t intend it. (The hallmark is that when I ask them how they chose dollar-weighting, they will say our backtests suggests this is the best way). If you’re trading in stocks, you are usually better off net long.
Thanks
Went short SSO this morning (S&P 2x levered ETF), and feel much better for it. I didn’t end up net zero, but I reduced my exposure substantially.