Profit Taking

[marko]

Profit Taking Question/Help

Here’s an investment portfolio question that may be hard to fully describe. I hope my friends here can help.

With the markets so volatile, DW has been wondering why we don’t ‘lock in’ our portfolio profits from time to time. That is, when the market has a nice run (like now) why don’t we take the PROFIT/GROWTH and turn it into a bond fund or cash and then if/when the market takes a dive we’d at least have that much in the bank. (yes, I’m accounting for ‘missing a rally’ and so on…I’m just looking at the general logic here).

I built an Excel model that shows there is little difference between banking profit or not and I’m wondering if there is some flaw in my model.

Essentially, if you bank the profit, you drop your ‘base’ back to the original level. Then in the event of a tanking, you drop from that lowered base level to a level lower so that even after adding in the ‘banked’ amount you’re at just about the same level than if you did nothing.

EX:
“Bank scenario”: 1000 in January X 5% = 1050. Bank 50; base goes back to 1000
Feb through Dec: Market loses 5%: 1000 X -5% = 950. 950 + banked 50 = 1000

“Do nothing scenario”: 1000 in January X 5% = 1050. Feb through Dec: Market loses 5%: 1050 X -5% = 998….Bank vs Do nothing: 1000 vs 998

In this example (and dozens of others that I’ve run) there is minimal benefit to warrant 1) the effort and 2) missing a rally.

Is there something wrong with my logic assumptions?

 

[ronocnikral]

so....you are simply trying to time the market? which, I have found to be a masochistic exercise and generally a loosing proposition. your assumptions are correct (omitting any tax considerations). however, you are essentially pencil whipping a scenario which benefits your bias. what about a 5% market increase and then a 10% decrease - easy enough to find a time in history where that happened. and if we could only guess what the future holds...

you would undoubetly benefit from an investment plan, agreed to by both you and your wife.

a couple of things to consider:

-are these in tax sheltered accounts? if not, the tax man loves your wife's suggestion.
-how much are you paying for a trade? if anything, the broker loves your wife's suggestion.
-do you have the time to check your accounts daily/multiple times per day? I track my net worth monthly (just for data purposes and to keep me focused) and that is a lot of work. and I certainly don't need DW calling me at work pestering me to sell this or that.
-are you investing long term or short term? if long term, rebalance once a year. if short term, why do you have your money in equities? it's a rollercoaster ride, but develop your plan, stick to it and enjoy the ride (if that is even possible).

 

[marko]

Thanks for the insight!

I was just trying to see if the assumptions/logic made sense and/or if there was some flaw in my model. More of a technical question than an investing one.

Fully aware of the perils of timing the market. Long term (several generations) investor..."Never touch the principal!"

Ran dozens of scenarios with plus/minus 2,5,10,15, and 20% variations. (Lots of free time to do this) All come out with relatively same result until you get crazy with 40% losses.

Checking your net worth should take about 10 seconds: (consider putting your portfolio on Morningstar...free subscription to high net worth customers via TR Price/Fidelity etc)

 

[Animorph]

It is some work for a small gain, but I do something similar for wider swings.

My version is to sell for cash the excess of portfolio equities over my retirement planning forecast of equity value. Since I'm retired, I can spend the cash if I don't get the chance to reinvest it. But if the portfolio goes down more than 20% from its peak, I start to reinvest any excess cash. For each 5% or so below that 20% I invest a little more of the cash in a number of steps that depend on how low I think the market might dip. Typically 5 steps.

It worked very nicely in 2007 - 2009. I retired in 2007 and hedged quit a bit with cash, gold, and BEARX to ensure that I didn't get hit with that dreaded market drop right after retirement. As the market dropped, I reinvested the cash and hedges back into the normal portfolio. The last step was at 50% down, with mostly money borrowed from a HELOC and enough cash left for only a few months of expenses. I've also already hit a couple of steps of reinvestment in the last/current dip as well. I did fully recover my 2007 peak portfolio value after the dip, with no bonds to speak of in my portfolio.

The steps are pretty mechanical, just following steps that I have preplanned. But you have to keep buying as you watch your portfolio sink to half its value. And it may not be as practical in earlier accumulation years when you're not going to spend any extra cash in the next few years.

The other thing I like to do that is similar to what you're thinking of is to buy additional shares of ETF's that are at least 10% below the basis cost of shares I already hold. When the shares finally increase in price to above my old basis cost, I sell the old shares. Just a way to take advantage of price drops, lower your cost basis, and avoid wash sales. A fun thing to do if you end up with excess cash and time on your hands.

 

[Chuckanut]

In the long run will this plan give better results than simply re-balancing the AA once or twice a year? I don't know but I wonder if in the long run the plan would give up a $ for every $ saved. And, I imagine, it would increase costs, perhaps even resulting in the more highly taxed short-term gains. One might end up churning one's own account and therefore raising costs, taxes, etc.

I still like the idea of re-balancing once a year, using low cost index funds, putting tax efficient funds in my personal money, and tax inefficient funds in IRA's, and generally keeping things simple.



Again, I don't know for sure. I wonder if there is an academic study of this plan versus yearly re-balancing.

 

[MichaelB]

Here’s an investment portfolio question that may be hard to fully describe. I hope my friends here can help.

With the markets so volatile, DW has been wondering why we don’t ‘lock in’ our portfolio profits from time to time. That is, when the market has a nice run (like now) why don’t we take the PROFIT/GROWTH and turn it into a bond fund or cash and then if/when the market takes a dive we’d at least have that much in the bank. (yes, I’m accounting for ‘missing a rally’ and so on…I’m just looking at the general logic here).

I built an Excel model that shows there is little difference between banking profit or not and I’m wondering if there is some flaw in my model.

Essentially, if you bank the profit, you drop your ‘base’ back to the original level. Then in the event of a tanking, you drop from that lowered base level to a level lower so that even after adding in the ‘banked’ amount you’re at just about the same level than if you did nothing.

EX:
“Bank scenario”: 1000 in January X 5% = 1050. Bank 50; base goes back to 1000
Feb through Dec: Market loses 5%: 1000 X -5% = 950. 950 + banked 50 = 1000

“Do nothing scenario”: 1000 in January X 5% = 1050. Feb through Dec: Market loses 5%: 1050 X -5% = 998….Bank vs Do nothing: 1000 vs 998

In this example (and dozens of others that I’ve run) there is minimal benefit to warrant 1) the effort and 2) missing a rally.

Is there something wrong with my logic assumptions?

Sounds like portfolio rebalancing to me. We do that at least once each year.

 

[Trawler]

Marko,
I to have contemplated doing the very same thing with signifcant assets many times especially when I am up in my fun money tradng account. Then reality sets in when my fun money fund is not doing so well. Thus will not take the risk of timing with the retirement portfolio .

Have come away with timing in the long run is impossible for me.
Plus I would like to experience one big bull in my life like the 90s or 50s or early 20s and do not want to miss it. After putting together what we have during the past tough decade it would be nice. If it happens I will be able to sell all equities and live the rest of my days with just good old cash. Not going to risk missing it.

Currently looking at a portfolio model that invests in 16 different asset classes for diversifiscation to mitigated risk and setting rebalance percentages that make for selling best performing assets (selling high) and buying underperforming assets (buying low). Some years rebalance once others rebalance many times. This is a Rick Edelman model he speaks about in one of his books.

My thoughts anyway good luck with what you decide.

 

[Lsbcal]

...(snip)...
I built an Excel model that shows there is little difference between banking profit or not and I’m wondering if there is some flaw in my model.
....

I've done Excel modeling and it takes a long time to get this stuff together. How far back did you go to use market data? I'd recommend going back to at least 1969 (where my model starts). You can get SP500 data from Yahoo but must add in SP500 dividends. For bonds you could use the 5yr Treasury constant maturity and convert to monthly gains using that rate data. I'd recommend using the data for the last day of the month.

I think this is a very useful exercise because having the proper model let's you test assumptions quickly. I love to do modeling, it's an analytical (partial) substitute for not working . Most people would not want to go anywhere as far as I've taken this stuff.

At least a backtest should work for any rational methodology before putting it in place with real money into the future.

 

[grasshopper]

I have some tax free LTCG to take this year since this opportunity may be lost in 2013.

 

[LOL!]

Here is a good paper to read on rebalancing which hasn't been linked in a while:
http://www.tdainstitutional.com/pdf/Opportunistic_Rebalancing_JFP2007_Daryanani.pdf

 

[audreyh1]

marko - why don't you explain the concept of owning stocks, bonds and cash, and then rebalancing between them to your wife. And explain that you only need to rebalance once a year and that way you don't have to worry about timing or trying to evaluate how "overvalued" or "undervalued" the market might be at any given instant. She might like that as it does reduce portfolio volatility as well as automatically "take profits" in the fastest growing assets each year.

Audrey

 

[Nords]

With the markets so volatile, DW has been wondering why we don’t ‘lock in’ our portfolio profits from time to time.

Another way to do that would be to sell call options and buy puts. It can be profitable, but it's painful enough that she'll soon opt for plain ol' periodic rebalancing.

 

[justplainbll]

Here’s an investment portfolio question that may be hard to fully describe. I hope my friends here can help.

With the markets so volatile, DW has been wondering why we don’t ‘lock in’ our portfolio profits from time to time. That is, when the market has a nice run (like now) why don’t we take the PROFIT/GROWTH and turn it into a bond fund or cash and then if/when the market takes a dive we’d at least have that much in the bank. (yes, I’m accounting for ‘missing a rally’ and so on…I’m just looking at the general logic here).

I built an Excel model that shows there is little difference between banking profit or not and I’m wondering if there is some flaw in my model.

Essentially, if you bank the profit, you drop your ‘base’ back to the original level. Then in the event of a tanking, you drop from that lowered base level to a level lower so that even after adding in the ‘banked’ amount you’re at just about the same level than if you did nothing.

EX:
“Bank scenario”: 1000 in January X 5% = 1050. Bank 50; base goes back to 1000
Feb through Dec: Market loses 5%: 1000 X -5% = 950. 950 + banked 50 = 1000

“Do nothing scenario”: 1000 in January X 5% = 1050. Feb through Dec: Market loses 5%: 1050 X -5% = 998….Bank vs Do nothing: 1000 vs 998

In this example (and dozens of others that I’ve run) there is minimal benefit to warrant 1) the effort and 2) missing a rally.

Is there something wrong with my logic assumptions?

Your analysis for taxable accounts should likely be different than for tax deferred and tax exempt accounts.
In the current tax environment, I'm inclined to harvest some long term capital gains from equities in which I have deep positions in order to provide additional funds for deepening positions in more thinly held equities or purchasing securities not currently owned. It's tough to resist selling 1/3 of of the shares in a stock when the proceeds cover the cost of the remaining 2/3rds.
With tax advantaged accounts, I bimonthly review for sale positions that are more than 4 months old and have annualized returns of more than 50%.
My current average holding period is 4.8 years and I try to limit my exposure in each equity position to < 5% of my total equity investment.
Your bank scenario has the advantage of giving you the opportunity to buy $50 worth of stock at a 5% discount.

 

[FinanceDude]

I only reblance once a year myself, and I do investing for a living..........

 

[ziggy29]

If you have a "hardwired" mechanism that you automatically and mechanically follow when it is triggered -- where something occurs and automatically moves some hot money into other areas -- it's basically a form of asset allocation with rebalancing.

If the "mechanism" for deciding when to lighten up on the "hot money" is just a gut feeling, then it goes more into the realm of "market timing."

Though I add the usual disclaimer of "past performance is no guarantee blah blah blah," historical backtesting suggests that rebalancing too much is counterproductive. I believe Bill Bernstein commented that the optimal "rebalance period" was around 12-18 months, but I don't know if that has changed as a result of the crash and the volatility of the last several years.

 

[Midpack]

OP sounds like rebalancing to me too.

I've never understood rebalancing annually, I only want to rebalance as necessary, the less frequent the better. I update and review all our holdings quarterly and only rebalance if we fall outside the 5/25 rule **. I have gone more than a year without a transaction, and other times more than once a year but never more often than quarterly. YMMV

** When a major asset moves more than 5% off target, it’s time to rebalance. Example: Your asset allocation is 70% stock/30% bond. If equities grow to be 75%, then it’s time to rebalance.

If the an equity asset is 25% or less than total equity, then you use the 25% trigger. For instance, if you have 10% REIT, then the rebalance points would be plus or minus 25% of 10, which = 12.5% and 7.5%.

 

[FinanceDude]

EVERYONE has a BUY discipline, almost NOONE has a SELL discipline.........

 

[ziggy29]

I've never understood rebalancing annually, I only want to rebalance as necessary, the less frequent the better. I update and review all our holdings quarterly and only rebalance if we fall outside the 5/25 rule. I have gone more than a year without a transaction, and other times more than once a year but never more often than quarterly. YMMV

Either way works because both accomplish the following two objectives:

(1) Selling more richly valued assets for assets getting relatively cheaper;

(2) Avoiding "gut feeling" decisions that are influenced by the two twin portfolio-killers, fear and greed.

They may not yield identical results, but IMO either way is superior to both pure buy-and-hold and gut-feel market timing. I don't believe there's "one true way" to AA -- getting the two points above correct are the most important.

 

[Midpack]

I suppose. But if at year end your equity holdings were 76% vs a 75% target, is it worth it? Or if you wait until year end and you're at 90% vs a 75% target?

Seems like rebalancing based on some threshold would be useful, that's my logic for using 5/25, I don't recall where the methodology originated but it makes sense to me. Again, annual is certainly not too often regardless.

 

[ziggy29]

I suppose. But if at year end your equity holdings were 76% vs a 75% target, is it worth it? Or if you wait until year end and you're at 90% vs a 75% target? Seems like rebalancing based on some threshold would be useful, that's my logic for using 5/25, I don't recall where the methodology originated but it makes sense to me. Again, annual is certainly not too often regardless.

Maybe, but I'm not going to get into that debate -- I'd sooner revisit "rent or buy" or "pay off debt or invest" as it might be more productive.

Seriously, though -- I hear what you are saying, and my rebalancing has been every 12 months, but only if the AA is enough "out of whack" between stocks and bonds to justify it. My rule is 3% out of whack. So if my target was 60/40 and I was at 61/39 or 58/42, I would not rebalance. If it were 63/37 or 57/43 (or farther out of whack from the target), I would.

And even when the stock/bond allocation isn't too far off, if the allocations in the stock class are too far out of line (as happened in 2011 when international funds tanked relative to US stocks), I rebalance within the stock buckets even if I don't change the AA to do it.

Still, Bernstein's research tended to show that you didn't want to rebalance a rapidly rising asset too quickly, as hot money tends to remain hot in the short term. Nevertheless, I'm sure there are other ways to approach it. Again, just get the big picture right -- buy lower, sell higher and avoid bad emotional decisions. Either mechanism for triggering a rebalance does that.

 

[Midpack]

No debate intended, thanks...

 

[NW-Bound]

To me, the term "profit taking" is only meaningful when you invest in individual stocks, and have had a huge run up. Kind of like this poster, when he said he watched a $50K investment ballooning up to $600K, then crashing down to 0. See this and this.

When you are investing in MFs, you are only rebalancing, or in the extreme case, market timing.

I have never had a stock that went up 12 times like the above poster to know what I would do, although I have had stocks that ran up 2X, or even 4X. Additionally, I am so conservative that the single stocks I would put $50K in are established companies that would never go up that much.

About taking profit, I remember reading a book by John Neff. His claim-to-fame was as the manager of the Vanguard Windsor from 1964 to 1995, where his annualized return of 14.8% soundly trounced the S&P500.

Anyway, John Neff said one should not be too greedy. He said one should always "leave something for the next guy", and that he was never "smart enough to get out at the very top". He said he'd rather sell too soon than too late.

 

[ejman]

I don't like to re balance too often for two reasons 1) I don't like the hassle and 2) I don't think that frequent re balancing such as annually does much from a total return standpoint.Quite frankly I can not understand the yearly rebalance - what does a date in the calendar have to do with the relative valuations?

My approach is to use wide bands of 10% so that basically I do nothing until the 10% band is exceeded. Since I ER'd in December of 2002 I've re balanced twice. My equity allocation exceeded the 10% band in early 2007 (At that time my equity midpoint was 60%) so I sold out to my equities midpoint which was nice come 2008. In early 2009 I was getting ready to sell bonds and buy equities but as I was getting ready to pull the trigger in March 2009 equities started going up. I sold again in 2010 to get my equity balance to 50% on account of my age (I turned 60 in 2010) and I intend to keep the 50% equity allocation indefinitely. I normally do my reallocation selling/buying is in tax advantaged accounts.

The above approach is my version of sell high and buy low and so far so good.

 

[Lsbcal]

...(snip)...
Anyway, John Neff said one should not be too greedy. He said one should always "leave something for the next guy", and that he was never "smart enough to get out at the very top". He said he'd rather sell too soon than too late.

In 1987 the market peaked in August with a fantastic run, then went down and really cratered in mid-October (down 22% in October). Tough to get that timing right. Any rebalance scheme should consider periods like this.