I typically enjoy articles that look back in time and tell you how much money you could have today if you invested $X in the past. In all seriousness, the simple arithmetic by Delong proves a few very important portfolio management tenets.
First off, being too leveraged can destroy your portfolio. As you can see below, the beta=4 portfolio goes bankrupt during the Depression and the beta=3 was a few breaths away as it bottomed at $3 from $3,000 prior to the Depression. Another real world example of this is the spectacular blow up of Long Term Capital Management. And of course, who can forget the recent financial crisis; which was a liquidity event caused by over leveraged balance sheets.
On the other hand, not being leveraged at all leaves valuable dollars on the table. For instance, the beta=1 portfolio (this would be investing in an S&P 500 index fund) ended the time series at $16k. Compared to the beta=2 portfolio at $380k. Not to mention the optimal portfolio (well, optimal in this simplistic and unrealistic example) of beta=2.42 had an ending value of $528k.
This is generally important because the last 8 years has been glorious for stock portfolios. Even more glorious for stock portfolios with leverage.
I want to end with a thought on the “you would have to be a real idiot to put any of your money in bonds relative to a diversified index of stocks” line from Delong. First off, this is a dangerous statement. Sure, from a pure performance perspective this would be true. But if we consider risk metrics such as VaR and standard deviation the story would change. Not to mention, a growing proportion of the population cares more about income (getting a check in the mail) than the value of their portfolio.
Now, an argument can be made that a basket of dividend stocks of top rated companies will yield more than a top rated bond portfolio and have an appreciation component. But, remember, dividends are not guaranteed. Sure, technically, bond interest is not guaranteed either, but the risk of a bankruptcy from a top rated company is pretty close to zero.