Triple X returns, how do levered ETF work for (against) you!

One of my students in Finland (Aalto University) asked me a simple (enough) question this winter...

Why would any young person not invest in a levered ETF?

My first inclination was to agree (I still do)... I  did mention that because of compounding the long-term returns would not be exactly 2X or 3X those of the benchmark index. I was not sure how big of a discrepancy this would be so I decided to do some work. I chose UPRO, the well known ETF that aims to give 3X the return of the SP500. It began trading on 6/25/09. Conveniently that is right after the worst of the crisis.

The following chart shows the results of investing $1,000 in both, from 6/29/09 - 1/10/2013.
SP500=60%
UPRO=235%
The winner is... well, this is an easy one. More than XXX, UPRO is almost 4X!!! I don't think I need to point out anything about UPROs volatility

Data from yahoo finance

 So what if we had had a bad few years? I reversed the daily returns and added them in the chart below... The lines "BEAR" show what would have happened to an investor if every return had been the opposite of what it was (a 1% gain turns into a 1% loss). The "Bull" lines are the same as the chart above.
Toting up in this case


Bull SP500=60%
Bull UPRO=235%
Bear SP500=-44%
Bear UPRO=-89%

The winner is...hard to tell...

Data form yahoo finance

I will never say this in my classroom, (too afraid someone will sue me): I am still for the levered way... If you are 20 and have nothing to lose... had you started with $476 on 6/25/09 in UPRO, you would be in the same place as someone who invested $1,000 in SP&500. Young people can worry about re-balancing their portfolios later.

Data from yahoo finance

 It is a probabilistic game... This is from a previous posting on this blog... if the percentage of winners is loaded on my favor, why not take advantage of it and go XXX?

Data from Thomson Reuters