Index portfolios are theoretically a fairly good balance between volatility and return. At a general annual volatility of 15% (enjoyed by only a few world markets), and a mean return of over 10% (depending on how it's measured), they seem pretty safe. Over a year, we expect about a 2/3 chance of being between -5% and 15% return, and a 95% chance of being between -20% and 40%. [Students of stochastic calculus, don't worry, the refinement is coming up later]
One way to take advantage of this is to go to your broker and buy 100 SPY on margin. Assuming the S&P is at 1400, that'd take around $7,000 today. Over a year, you'd expect a 2/3 chance of ending up with between $6300 and $9100, with a mean of $8400. Your margin interest would be around $500, so you're averaging $7900, a 13% return, here without really doing anything.
Another way is to buy a year-out deep-in-the-money S&P call option. A 1050 call, for example, would be about $400. If you can afford it, there may be rewards. We're looking at a mean value of 1540 and a 2/3-chance range of 1330 to 1610; there's not a lot of risk of a worthless expiration on this option. Your $40,000 investment has a mean of $49,000 and a range of $28,000 to $56,000.
Even better is the leverage provided by S&P futures, if you want to try that. I don't know the exact requirements, but at 10% margin you might make as much as a 150% return, with a mean of 100%. The drawback here is that a short-term drop would hurt really, really bad.
As part of an aggressive portfolio, some margined index shares is a good diversification technique, especially for smaller portfolios that need to minimize commission (I try for no more than 2% to start out with). Nevertheless, I doubt many individual investors try for the super-leveraging techniques.
Still, let's think about the call-option technique. It gives us a leverage of around 2.25, and if we don't like that, we could probably go up a few strikes to widen it to 3 or more. But I imagine people have tried ideas like this before and had less than excellent results. This is because while the probability of a large price drop is low, the results it inflicts are devastating.
The manifestation of this in mathematical terms is that the constant drift rate of a stock price, mu, is not a measurement of expected return. To find that we have to apply Ito's lemma to the logarithm of the price. The result is that mu is reduced by half the squared of the annual volatility. That knocks down an assumed 10% "overall" upward drift of the S&P to 8.8%, which isn't really noticeable. But if we apply a constant leverage of, say, 3, this triples both mu and the volatility. Upon calculation, the mean return is not 30%, but 20%.