Hacker News Reader

Comment by cortesoft | Hacker News Reader

That is what risk is.... not knowing. If you want high reward from an investment, you need to take high risks, which means you will invest without knowing something.

Now, some risks are smarter than others...

View on HN

That's not necessarily a good idea. You don't need to take on higher risk for a higher return. It's usually better to leverage risk-adjusted returns rather than chase high total returns with commensurate risk.

You can take on more risk that if you have the appetite, but if that's the case you could also just use levered beta (e.g. 3x levered S&P 500). This would significantly improve your portfolio while still being fundamentally much easier to understand than a novel asset.

If your portfolio has a decent risk adjusted return and very low volatility and beta exposure, it's safer to just leverage it up to the same risk as the S&P 500. This reduces the chance of you blowing up your capital in the long run.

I would argue there are very few investment goals for which extreme risk is sound (especially if it's unhedged!).

That's also ignoring the cost of capital, as leveraging risk adjusted returns has to take that into account. You don't get the same rate of return if you use margin, say, in order to leverage.

If you use a 3x or other leveraged fund, then you run into tracking issues (look at https://www.etf.com/etfanalytics/etf-comparison/SPXL-vs-SPY) where you see tracking break down), you can lose everything (remember XIV?), and you have other potential issues.

So it's not as simple as leveraging up a low-risk portfolio to assume a given risk/return ratio. There's also diversification to keep in mind.

That is a good point, but that all applies to riskier investments as well. Whether or not that specific example will beat out something like cryptocurrencies does depend on margin and transaction costs, this is true.

> You don't need to take on higher risk for a higher return

The lower the risk of an instrument, the more it will be saturated with investors, the more thinly per-investor share of profit will be spread. Therefore there is no such thing as "low risk, high returns", unless it is a scam. There is no unexploited profit opportunity that is risk free, if one thinks they have found one, they must have just missed accounting for the hidden risks.

I didn't say "low risk, high returns." It is a spectrum. What you've said in your first two sentences sounds fine as a textbook principle. But the real world is messier and opportunities don't just vanish: if you do the math on a basic risk parity strategy with the S&P and some uncorrelated ETF, you can see it will beat the market on a risk adjusted basis. Very often you can then leverage this up to a higher absolute return than SPY while keeping lower volatility and beta overall.

> There is no unexploited profit opportunity that is risk free

This is essentially encapsulated by a Sharpe ratio (among other things). On the contrary, it is not especially difficult to produce a relatively high Sharpe ratio, accounting for transaction and margin costs, if you don't have a large amount of money to invest (large means single digit billions or more). This is especially, but not exclusively, the case if you don't care to compound your returns.

> But the real world is messier and opportunities don't just vanish: if you do the math on a basic risk parity strategy with the S&P and some uncorrelated ETF, you can see it will beat the market on a risk adjusted basis.

Sure, there are asymmetries in real life and information takes a while to dissipate to all agents, but unless you're a robo-trader doing high frequency temporal arbitrage, I don't buy for a second that you can beat the market on a continuous basis. You might think you do, but that would be just some hot-hand fallacy.

> This is essentially encapsulated by a Sharpe ratio (among other things).

You're conflating two things. Sharpe ratio is about a profit opportunity with regards to its well established degree of risk. Unexploited opportunity is when you asymmetrically discover a new return opportunity upon what was already priced.

In both cases it comes down to the belief that "I am smarter than other investors, and can profit from an angle they haven't thought". I will not assert a strong efficient market hypothesis, but in overwhelming majority of the cases, no, you're not.

Like I said, run through a basic risk parity strategy holding SPY and an uncorrelated ETF. Calculate the beta, volatility and return over 10 - 20 years. It does beat the market continuously on a risk adjusted basis. With leverage it also beats on total returns with lower volatility and beta.

Beating the market is not mysterious, it's just difficult to do it by a lot or at scale. Just because well performing portfolios are well known doesn't mean they cease being effective in principle, the way you seem to think would happen. These days a risk parity portfolio isn't enough to solicit funds from savvy investors because it's well known and they won't pay management/performance fees if that's all you're offering. But it's been a staple since Dalio developed it 30 years ago for good reason.

You can find code for running through the kinds of things I'm talking about, as well as more in-depth discussion here: https://qoppac.blogspot.com.

if you leverage your risk for a higher return you _also_ increase your risk

Yes, but you can choose how much risk to take on using your leverage weight. You don't have to accept the baseline risk of the inherently riskier asset. It's easier to start with a less risky portfolio and weight it accordingly than it is to derisk a portfolio which is intrinsically riskier.

Reply on news.ycombinator.com