Wednesday, July 11, 2007

Gotta love free dividend reinvestments

Yesterday I noticed my first free dividend reinvestment in my TD Ameritrade IRA. My $2.10 dividend from AGG was converted to 0.021 shares and added to my position. Nice! I was surprised that the dividend was so much, considering I have less than $500 in it. If it keeps up, I'll get about 5.14% per year. Not too bad for a low risk investment. Of course, I'll lose almost all of the interest when I sell it. I'm going to really miss the introductory free trading period when it's over! Maybe by the time I need to pay commissions I'll transfer this account to Zecco.

I was thinking about an asset allocation strategy and I realized two things. First, I probably don't need to worry about it until I have a lot more money in the market. Second, I bet the performance of the strategy depends more on the length of time between re-balancing than anything else. Most of the things I've read say to pick a time period of half a year to a year. It makes sense to use a longish time period because if you re-balance too often you miss out on big moves in price.

For instance, say I have a huge $100k portfolio with 10% in bonds and 90% in stock. Every month for a year the stock goes down 5%, so after 1 year I've got 0.95^12 = 0.54, or 54% of my original investment in stocks (I lost 46%). We'll assume we have no dividends and that the bonds haven't changed. I don't know how realistic that is, because I've heard bonds usually go up if the stock market goes down a lot. Anyway, that leaves $10k bonds + $48k stock. Now the allocation is 20% bonds, 80% stock. Time to re-balance! So I sell $5k of bonds and put that money into my stocks.

Now in a few years, who knows how long, the stocks slowly climb back up to their original position. Maybe they went down even more in the meantime, or bounced around a lot, but we can assume that eventually they get back up there. We still have $5k in bonds, but how much stock do we have? Well at this price level we used to have $90k in stocks, but we added 5/48 = 10.4% to our position while it was low. So now we have $90k * 1.104 = $99k! Yay. This is why asset allocation is good. It forces you to buy low and sell high (we had to sell bonds and buy stocks while they were low to get them back into proportion).

If we had re-balanced too quickly and too often, we would have had a smaller overall return, because the money we took out of bonds would have fallen along with the original stocks. But if the downturn had been shorter, say only 6 months, we would have missed out on the low stock prices. What to do?

I guess the best solution is to ignore time-spans and establish numeric rules for re-balancing. For instance, rather than say you have a 10% bond allocation, you could say you have 5% to 20% bonds with a 10% target (so the limits are half and double, not a fixed number either direction). You only re-balance your bonds if they fall under 5% or rise to over 20% of your portfolio. If it happens in 2 months you do it then. If it takes 2 years, you wait until then.

That range might be too big since basically your stocks would have to fall by half or double to get out of balance. Maybe 7% to 14% would be better? I don't know. It's definitely an art, not a science.

1 comment:

stealthy said...

I would say rebalance by numeric rules like you mention near the end. Ignore time spans because you can miss out during opportunities.

I plan on using this theory for balancing my portfolio with company stock. My plan is to keep about 10% of my portfolio company stock. If it falls below 5% I'll buy enough to get it back up to 15% then as time goes on that percentage will get smaller due to my growing 401k and other investments. I'll let it shrink back to 5% then redo back to 15%

*lemme just know that this was my ORIGINAL plan. It's not working out right now because I max out my ESPP because I get a match and its free money. I invest $3K in my ESPP a year so that means I would have to put back $30k to keep it at a 10% proportion. I think it will work out better as I grow the rest of my portfolio.

So for now I'm just going to keep maxing out my ESPP @ $3K and keep an eye on it. Reevaluate maybe at age 30.