Does Re-Balancing Lock In Gains?

[heybro]

So, I make sure my account "auto" rebalances.

Is there some sort of magical locking-in benefit that occurs during rebalancing or is that just smoke and mirrors?

In other words, when you rebalance, you are selling shares that are HIGH and, with these dollars, buying shares that are LOW.  (Sell High; Buy Low).

Therefore, is this an argument for owning many different types of uncorrelated assets?  Because it sort of all dances with each other? hehe

[steveo]

I think re-balancing does imply buying low and selling high on a relative basis.

[gobius]

Yes, re-balancing is a simple way to lock in on gains and the key part of that is owning many different assets that don't correlate with each other (or at least don't correlate very much).  The premise is that you own broad-based index funds and, overall, markets will go up over time.  If you own 50/50 stock/bond and the stock market tanks 50% with no effect on bonds, you now have 33/66 stock/bond.  You sell bonds and buy stocks to get back to 50/50.  You just bought stocks at a low price, assuming that the stock market will eventually recover.  If you stick to this strategy it keeps you from bailing out of markets when they are (hopefully temporarily) low.  It also keeps you from being too greedy when a large bull market is going on.

You could do the same with individual stocks and bonds; there's just more risk since an individual stock/bond could be on its way to losing all its value due to bad accounting or losing its competitive edge.  With broad index funds you are mitigating the risk of individual company failure and relying on the stock/bond market to recover after it dips, which if it doesn't, we have a lot more problems on our hands.

The key is non-correlating assets.  If all your assets went up/down the same way it would be pointless to rebalance because your percentages wouldn't change.  Stocks and bonds typically go opposite ways, which is why they are used together a lot.  Some people use international stocks and REITs since they don't correlate 100% with the US stock market.  Some may use gold/silver too.

[Kaspian]

Rebalancing is awesome!  It's something those folks in the "100% equities" thread like to ignore.  There have been years where bonds go up 16%.  Moving that cream off into the US equity index (when it was selling at a discount) while rebalancing has really paid off well.  Moving cash to the underperformers pretty much always work.  Check out the attached chart.

I think the trick is, you can't try and time it.  Tell yourself you'll do it at the very start of January of and/or end of June.  I rebalance twice a year and ignore all emotional feelings while doing it.

[Le Barbu]

Rebalancing does lock return but the trigger has to worth it before you do rebalance. There's no gain within 5% delta, so don’t do it too often. My guidelines are to buy lagging asset with new money, rebalance when you are 5% off for biggest asset classes and 25% from itself target for smaller one.

Ex. if the target is 25% rebalances when it's under 20% or over 30%. If the target is 5%, rebalance under 4% or over 6%

Don’t forget, rebalancing can locks gains but it takes A LOT of volatility, negative correlation, discipline etc. to get for some real gain out of this. The purpose should be to keep your AA in line with your goal. So don't do it for extra gains !

[smilla]

Ex. if the target is 25% rebalances when it's under 20% or over 30%. If the target is 5%, rebalance under 4% or over 6%

Le Barbu could you further explain this.  Do you mean if you are down more than 4% or up more than 6%, in the case of your larger target allocations?  Or do you mean if it's within a 4-6% difference?

[skyrefuge]

Rebalancing should not be thought of as a way to increase your investment return, because in most cases it will not, and it will often reduce it.

See http://www.kitces.com/blog/is-passive-rebalancing-a-form-of-active-management/ for a detailed discussion and examples.

By "locking in a gain", while you have removed the possibility of future losses on those dollars, you have also removed the possibility of future gains. If stocks rise at a steady 10% per year and bonds at a steady 5%, all that rebalancing will do is trade your better-performing assets for lower-performing assets, reducing your return. Not that you shouldn't rebalance; there are other important reasons why you should. Enhancing gains is just not one of those reasons.

cFIREsim generally supports Kitces's analysis as well.  With a 50/50 portfolio and a 4% WR, the ending portfolio average and median are significantly higher when you do *not* rebalance, and the failures only increase from 12 to 13 (out of 115).

[Le Barbu]

When I mention 4% and 6%, it´s for an asset which target is 5%.

[smilla]

When I mention 4% and 6%, it´s for an asset which target is 5%.

Thanks.  I totally misread that.

[seattlecyclone]

cFIREsim generally supports Kitces's analysis as well.  With a 50/50 portfolio and a 4% WR, the ending portfolio average and median are significantly higher when you do *not* rebalance, and the failures only increase from 12 to 13 (out of 115).

Have you seen a good explanation for why this is? It seems intuitively obvious that occasional rebalancing causes you to trade some assets that are outperforming their historical average for other assets that are underperforming, using market cycles to your own advantage. "Buy low, sell high" as they say. But the data seem to indicate that the intuitive answer is wrong. I'd love to see an expert explain why that might be the case.

[Eric]

cFIREsim generally supports Kitces's analysis as well.  With a 50/50 portfolio and a 4% WR, the ending portfolio average and median are significantly higher when you do *not* rebalance, and the failures only increase from 12 to 13 (out of 115).

Have you seen a good explanation for why this is? It seems intuitively obvious that occasional rebalancing causes you to trade some assets that are outperforming their historical average for other assets that are underperforming, using market cycles to your own advantage. "Buy low, sell high" as they say. But the data seem to indicate that the intuitive answer is wrong. I'd love to see an expert explain why that might be the case.

I'd guess that because over the long haul, you're trading stocks for bonds to decrease your volatility, but of course that also decreases your maximum possible returns.  I don't think there's a 30 year period where stocks trailed bonds, right?

[skyrefuge]

Have you seen a good explanation for why this is? It seems intuitively obvious that occasional rebalancing causes you to trade some assets that are outperforming their historical average for other assets that are underperforming.

I thought the Kitces link explained it pretty clearly, and Eric summed it up nicely too.

Also, be aware that in order to be sold, an asset doesn't have to outperform its historical average, it just has to outperform the asset that it's being rebalanced to. Imagine you're expecting stocks to return 10% and bonds 5%, and want to maintain a 50/50 portfolio. Even if stocks underperform and return only 8% while bonds soar, returning 7%, you'll still have to sell stock "low" and buy bonds "high" to maintain your allocation.

[tj]

Nope - see TFB's article:

http://thefinancebuff.com/rebalancing-sell-high-buy-higher.html

[foobar]

cFIREsim generally supports Kitces's analysis as well.  With a 50/50 portfolio and a 4% WR, the ending portfolio average and median are significantly higher when you do *not* rebalance, and the failures only increase from 12 to 13 (out of 115).

Have you seen a good explanation for why this is? It seems intuitively obvious that occasional rebalancing causes you to trade some assets that are outperforming their historical average for other assets that are underperforming, using market cycles to your own advantage. "Buy low, sell high" as they say. But the data seem to indicate that the intuitive answer is wrong. I'd love to see an expert explain why that might be the case.

Because rebalancing doesn't cause your to sell high and buy low. It causes you to sell a high returning asset to buy one that hasn't returned as well.  In cases like switching between international stock and US (who over long periods of time will have roughly the same return), you are likely buying low and selling high. When doing it with a low return asset like bonds, you are pretty much just reducing risk at the expense of return. You tend to get much sharper drop in risk for only giving up a little bit of return. Please note that we are talking long term here (30+) years. Over short periods like 10 years, anything can happen.

[Kaspian]

I don't think there's a 30 year period where stocks trailed bonds, right?

I don't think so, but there are definitely decades where they've trailed.

I think the overall point of rebalancing AND diversification is not to have godlike powers and beat all the other financial possibilities.  It's actually to beat yourself.  It's been proven that investors are their own worse enemy and subject to incredibly stupid things (especially when equities drop).  Rebalancing and creating diversification are like a car seatbelt and safety on a gun.  It's not that the item itself is dangerous, but the human in control can do some amazingly stupid things.  Especially without any bulwarks in place.

[skyrefuge]

I think the overall point of rebalancing AND diversification is not to have godlike powers and beat all the other financial possibilities.  It's actually to beat yourself.  It's been proven that investors are their own worse enemy and subject to incredibly stupid things (especially when equities drop).  Rebalancing and creating diversification are like a car seatbelt and safety on a gun.  It's not that the item itself is dangerous, but the human in control can do some amazingly stupid things.  Especially without any bulwarks in place.

Yup, exactly.

[foobar]

  I don't think there's a 30 year period where stocks trailed bonds, right?

Long term bonds beat stocks between 1982-2011. You can google bonds beat stocks to see all the articles about it. Now long term bonds are normally ignored (they have volatility and returns like stocks) but for the people going 90/10 and the like they make sense.