That makes absolutely no sense. LTCM had about 30 to 1 leverage looking to make as little as a few basis points per trade. They too thought that by getting bigger and more diversified they were actually lowering this risk. This mathematically only works when correlation is low. When correlation is high, the diversification becomes one large synthetic position levered to the moon. And we know how that ended.
"With many market strategies, risk goes down as leverage goes up" I would argue that statement is inaccurate in a way that is quite important. All things being equal, leverage should not have impact on the safeness or riskiness of a strategy. Unless you use too much of it, in which case risk of ruin becomes unacceptable. The purpose of leverage is to amplify a potential return, at the cost of also amplifying potential drawdowns. If an unlevered strategy provides a 10% return and 3% historical drawdowns, then trading it at 10X would create a profile of 100% return and 30% drawdowns, or what have you. Point being here, leverage is an amplifier in both directions. Availability of leverage should not change the risk / return profile of the strategy itself, independent of "tap out" considerations (how much money you can lose on the downside before being wiped out). Of course, real world factors make the leverage question interesting. What Long Term Capital Management did was essentially argue their "hoovering up nickels" strategy had a drawdown risk of zero, allowing them to take a 50 basis point return (or whatever it was) and lever it to ridiculous orders of magnitude. We know how that turned out. Anyway, I think what you are trying to say is that a certain amount of leverage is needed in order for XYZ market neutral strategy to be implemented properly. But what you are really saying is that, a certain amount of AMPLIFICATION is needed to make the market neutral RETURNS worth gunning for in the first place, because unlevered market neutral is just not worth doing. Market neutral is about units of risk in a certain ratio to each other. So if you have X units of risk bullish, you want to be sure you can also have X units of risk bearish, and so on. (A gross oversimplification, but useful for the purpose of this explanation.) The thing is, it is possible to run a market neutral strategy in a completely unlevered account. You would just have to trade very, very small. This would not impact the ability to EXECUTE the strategy properly... but it WOULD impact the return. So I would dispute your argument that "higher leverage results in a higher sharpe ratio and smaller drawdowns." What you are really saying (I think) is that to run XYZ strategy at X size, to produce X return profile, a certain amount of leverage elbow room is required, and without that elbow room you are constrained. But the logical conclusion from this is not that more leverage = safer execution. It is to ask the following questions: * What are my realistic returns for running this strategy unlevered * How much do I want to amplify those returns with leverage, given the trade-off of amplified drawdowns as well Keeping in mind, too, Michael Milken's post-crisis observation that "leverage is not a business model." Meaning, if a strategy produces tiny returns on an unlevered basis, the light may not be worth the candle in terms of levering it up significantly. So all in all, it is not correct to say that increased leverage lowers risk. If one has a strategy in which more leverage is needed just to implement the strat properly, then the initial position sizes were too big in the first place. And if reducing position size to properly fit the available capital also reduces the return below the cost of funding, well, that is an issue with the strategy itself.
Doh, Mav and I are on the same page (I was finishing my reply even as you posted). Is that a sign of the Mayan apocalypse or what
I'm not knocking the idea of first loss programs. To the contrary, I think the idea sounds fairly interesting. Mike M, I agree with some of your previous posts; these allocators fill a much needed hole in the hedge fund capital raising ecosystem. Getting back to my original question though, if the leverage I can obtain through the allocator's program is comparable to what can be achieved in a Portfolio Margin account, then access to leverage is not a differentiator of "First Loss" type programs. In fact, the same leverage can be obtained with less risk to the Portfolio Manager's capital simply by using a Portfolio Margin account. The "less risk" part comes into play since with my Portfolio Margin account I have a similar "first loss" provision, but I also have a provision that grants me 100% of the gain (as opposed to a gain split provision with the allocator's program). We can therefore safely assume that the main benefit of such a program is the institutional exposure, the networking opportunities, the contacts gained, the mentoring provided on the capital raising side, and the future possibilities of ramping up capital based on track record. The value of this stuff cannot be overstated. Speaking as someone who has had experience going through a fund of funds due diligence process for months and months and months, the first loss program may well be worth it simply to expedite the entire process. Mike M, do you agree with my general statements above?
p.s. Final addendum, the above is not meant to knock the use of leverage. Sometimes leverage CAN and SHOULD be used... when the unlevered return profile is strong enough. This leads to a further interesting point, though: Participants in a first loss program are generally better off reducing leverage in % terms, rather than increasing it, because they are making a worthwhile trade-off: smaller % returns for higher absolute dollar returns across a much larger capital base. Take the hypothetical example of a trader with a historical track record of 36% returns and 12% drawdowns on his own 500K capital base. If this trader went into a first loss program, and began trading $5MM instead of 500K, he might cut his notional leverage by 50%, thus gunning for 18% returns with 6% drawdowns. In such case, his % return would be lower, but the absolute dollar return would be meaningfully higher (36% of 500K = $180K; 18% of $5MM=$900K). And the above, of course, cuts to the heart of institutional AUM in the first place: Seeking to generate higher absolute dollar returns, across a larger capital base, by dialing back the leverage on a strategy as such that % drawdowns are tolerable and acceptable to a risk averse capital base of high net worth investors. If said trader was successful in generating 18% returns w/ 6% drawdowns on his new 5MM base, he would then have the ability to get a traditional investment, more AUM, etcetera, to further press this idea. So the line of discussion brings up the following useful clarification - the best strategies for such are the ones that can be dialed down and still have attractive risk/return profiles, not the ones that need to be juiced up. This was something we already knew intuitively, but had not spelled out in discussion.
Hold on here guys, we have what is called a "failure to communicate". You guys are all talking about different things. Portfolio margin has nothing to do with leverage except in the cases in which it can be abused. Portfolio margin was created to give offsets in most cases were no risk was being added to a portfolio but just notional exposure. Adding 10 to 1 risk to what the "standard" risk parameters of portfolio margin would be suicidal. You will lose your 500k and quickly at that. That money will be gone. You will NOT see it again. There are no refunds in this game. This goes back to my very first complaint on this thread. It was not meant to be directed at dark horse but rather the idea that very few people are going to understand this program. You can't add leverage to leverage. You simply can't do it. I don't care if all you're going to do is crack two eggs and make an omelet. The way you need to think about this program is through notional exposure. You put up 500k, the allocator puts up 4.5 million and you are trading a 5 million dollar account. That is the account value. You have a 10% threshold before they cut you off. If the fund is down 10% on 5 million, you lose everything. Game over. Don't even use the word leverage as it only complicates things. As dark hose said, you might want to trade the account as if it's a 2.5 million dollar account allowing yourself a better drawdown threshold. What your strategy is and what you want it to do is not relevant to the allocator. You have to understand the math. You can believe whatever you want in terms of risk, just understand that once this fund is down 500k, it's time for you to move on.
Final quick observation: The portfolio margin question is an interesting one and deserves more investigation. It is not clear to me that portfolio margin accounts inherently allow more leverage than Reg-T accounts. They will under certain circumstances, but under other circumstances they would actually allow less leverage. Via IB (bold emphasis mine): One of the main goals of Portfolio Margin is to reflect the lower risk inherent in a balanced portfolio. Depending on the composition of the trading account, margin requirements under Portfolio Margin could be lower than under the Reg T rules. This translates to greater leverage (note that trading with greater leverage involves greater risk of loss). Conversely, for a portfolio with concentrated risk, the requirements under Portfolio Margin may be greater than those under Reg T, as the true economic risk behind the portfolio may not be adequately accounted for under static Reg T calculations. Therefore, one should not assume that Portfolio Margin availability is a license to ramp up risk. It is simply a more enlightened way of looking at total portfolio exposure. And it really is no substitute for the potential absolute dollar return amplification of 9 to 1 - the main restriction of which, at least as far as a first loss program goes, is the trader's willingness to consider his allocated dollars to the program as risk capital, while executing on a program in which notional leverage is set as such that max drawdowns should fall at less than 10%.
Right, the whole point of the first loss program is to structure the risk / return profile so that odds of losing one's committed stake are acceptably low. The way to do this is to adjust notional leverage so that max expected drawdown is well below the tap-out point. At 9 to 1, the tap-out rate is 10%, so you want your max expected drawdown on the dialed back strategy to be less than 10%. The benefits to doing this are that, if you have a good year as expected, or better yet a string of good years as expected, your own returns are substantially amplified and you are on your way to millions or tens of millions in AUM. It is like the trading equivalent of a seat at the WSOP main event, except with a hell of a lot better odds, because all you have to do is have a decent year to get a seat at the final table. As for "very few are going to understand this program"... yes, and even fewer are actually going to implement it. But nobody is going to go in blind, and nobody is accepted INTO the program without rigorous vetting in the first place. We are not handing out keys to Indy 500 cars here -- at least, not to anyone who has failed to prove beyond a shadow of a doubt, through a deep and rigorous vetting process, that they know how to drive the thing and have had good experience on previous tracks. As far as the individual trader accidentally blowing himself up, there is far more risk of that in simply opening up a run-of-the-mill forex or futures account and not understanding the world he is stepping into.
Please remove the words portfolio margin from all posts on this thread. You are going to drive this discussion over a cliff. You are taking something that is complicated and adding complexity. Albert Einstein is rolling over in his grave right now. "Everything should be as simple as possible, but not simpler." How this guy wants to trade is his own problem. He puts up 500k, you give him 4.5 million. Total account value is 5 million. Once the account is down 500k, the account is shut down and trader loses 100% of his investment. The trader also needs to understand that his percent return will be measured on what he makes on "5 million". Keep it simple guys.
that so right.. i don't fucking get how people don't see that! yeah just keep adding leverage till things have no risk.. haha really! GEEZ thats America right there.. "just put it on my credit card"