Dollar cost averaging is... a lie?

[Sisyphus]

Hi all. A few nights ago I spotted an article in a financial industry trade magazine that made the claim that dollar cost averaging (DCA) is a farce. They claim that lump sums outperform DCA the majority of the time. And, that many companies promote DCA to make more commissions. :

I don't want to spam you with links, so feel free to google "lump sum vs. dollar cost averaging" and you'll find a bunch of articles.

The interesting thing, I think, is that when you have a large sum already invested and you're leaving it in the market for the long term, isn't that something like "lump sum" investing... every month? DCA just seems like an easy way to "pay yourself first" and take the emotion out of it... two keys to any long term strategy.

Thoughts?

 

[Alex]

I've read these articles. What they essentially say is: If you have a large lump sum to invest, say from the sale of a home, statistically you should invest it all at once if you want to maximize return.

What they don't address is the normal situation that income earners are in. When you are earning money and investing a percentage of these earnings every month, DCA is the way to go.

 

[yakers]

Since the market goes up more days than it goes down then it is logical to invest a lump sum and as early as possible. For most folks, like me, with regular paychecks, DCA is the only way I can invest. And it is a good way, just not as good as a lump sum, in theory. Now from a defensive approach, one could DCA into the market if you felt it was overvalued and subject to a sudden decline. DCA should be done with an appropriate fund that does not charge for regular contributions, it would not be a good way to purchase a lot of small lots of individual stocks.

 

[sgeeeee]

Do a search on DCA and you will find this subject has come up before:

There has been quite a bit of research on DCA. If you have a lump sum, historically you have had a higher probability of maximizing return (in stock and bond investments) if you invest it as a lump sum as soon as possible. The reason for this is simple: there are far more periods of positive returns than there are for negative returns. Waiting to DCA the lump sum tends to reduce your overall return.

However, it is clearly possible to invest a lump sum just prior to a large decline in the markets. In this case, waiting to invest at least some of the money would avoid the losses associated with the decline in the markets. So there are no guarantees that a lump sum investment is best. What DCA does for you is reduce the risk of large losses. In a declining market, not all the money is losing at once.

So the choice of lump sum or dollar cost averaging a lump sum is simply another example of risk-reward trade-off.

There is another factor to consider. Where is the lump sum currently residing that you are considering investing? If the money is in cash, then the above arguments apply. If the money is in stock or bond investments and is simply being laundered into equivalent investments, the problem is different. Since the investments themselves are identical in this case, the movement of money provides no obvious investment advantages.

The main thing DCA accomplishes for most people is discipline. If you choose to invest x% of each paycheck into specific investments, you eliminate having to make a decision about what to invest in and when to do it. This is almost always a better decision than trying to time the market.

 

[LOL!]

The articles on DCA vs LumpSum that I have read, state that LumpSum beats DCA about 60-65% of the time,. So, yes, you have a higher probability of doing better with LS if you get a cash inheritence, win the lottery, get a big bonus, or otherwise need to invest a chunk of change.

But 60% is not too far off from 50%. So I'd rather hedge my bets and DCA. But don't take years to DCA your cash into your asset allocation, take few months at most.

 

[azanon]

Anyone who wont fully invest a lump sum if they have a lump sum to invest is either a market timer or should never be "fully invested". Of course a third choice could be you just dont know what to invest in yet due to being a novice investor. That aside, if a given market is what it is, then what difference does it make if you've been invested for years or you invest fully for the first time today? The past has already happened and the future is completely unknown in either case.

 

[mathjak107]

Anyone who wont fully invest a lump sum if they have a lump sum to invest is either a market timer or should never be "fully invested". Of course a third choice could be you just dont know what to invest in yet due to being a novice investor. That aside, if a given market is what it is, then what difference does it make if you've been invested for years or you invest fully for the first time today? The past has already happened and the future is completely unknown in either case.

i agree , they are gun shy and are afraid to committ.

 

[ladelfina]

A while back, brewer mentioned an author named Moshe Milvesky.
One of his books, "Wealth Logic" has a free chapter, which happens to be all about dollar-cost averaging, here:

http://www.captus.com/information/Wealth-flyer.htm

shorter version:
only do it if you don't have more to invest at the time, and even then make sure that your trading costs are as low as possible (preferably zero, like buying into a no-xaction fee fund you already own some of).

 

[boutros]

The worst part of DCA to me seems to be tracking the lot by lot cost basis. Great option for tax deferred assets though.

 

[ERD50]

It's funny, I'm a numbers guy, but I really do think the biggest advantage to DCA is to take the emotion out of it and turn it into a 'habit'.

For most investors, if you lump sum, and then the market goes down that next month, you feel bad, and it might affect your approach to investing. But doing it every month averages out those ups/downs. If you feel better about it, you are more likely to stick to it.

I advise DCA to people who ask.

And as Alex said, it really is rare to have a lump sum. For normal earners, just DCA as you go. Doing otherwise would mean putting a year's worth of savings in a MM, and THEN investing all at once. I don't think that is a good idea, or a good numbers game.

-ERD50

 

[sgeeeee]

The articles on DCA vs LumpSum that I have read, state that LumpSum beats DCA about 60-65% of the time,. . .

Well . . . that depends entirely on the number and timing of the DCA events you choose. Are you going to use two DCA contributions? three? six? twelve? and how often? weekly? monthly? . . .

But what is just as important as the probability of lump sum beating DCA is by how much? And how much do you risk by doing lump sum?

You can find DCA calculators on the web to explore these questions . . . historically. Here's one:

http://www.moneychimp.com/features/dollar_cost.htm

 

[FinanceDude]

When clients ask me when a good time to invest is, I usually say: : "When you have decided you WANT to invest"..........i.e. never a bad time.

The reason I say this is that I have seen more UNDERFUNDED "plans" than I thought was possible.........

 

[retire@40]

When clients ask me when a good time to invest is, I usually say: : "When you have decided you WANT to invest"..........i.e. never a bad time.

The reason I say this is that I have seen more UNDERFUNDED "plans" than I thought was possible.........

I wonder what the stats are for how many people that qualify to invest in a retirement plan actually invest.

If I had to guess based on what I've seen, I would say probably 10%.

 

[LOL!]

Well . . . that depends entirely on the number and timing of the DCA events you choose. Are you going to use two DCA contributions? three? six? twelve? and how often? weekly? monthly? . . .

But what is just as important as the probability of lump sum beating DCA is by how much? And how much do you risk by doing lump sum?

You ask questions. Here are some answers:

http://www.fpanet.org/journal/articles/2004_Issues/jfp0604-art11.cfm

http://www.fpanet.org/journal/articles/2006_Issues/jfp1006-art8.cfm

 

[azanon]

And as Alex said, it really is rare to have a lump sum. For normal earners, just DCA as you go. Doing otherwise would mean putting a year's worth of savings in a MM, and THEN investing all at once. I don't think that is a good idea, or a good numbers game.

Is it now? With all the life insurance policies out there, and estates? I have personally benefited from one of these and also know several others that have too.

 

[ERD50]

Is it now? With all the life insurance policies out there, and estates? I have personally benefited from one of these and also know several others that have too.

Well, 'rare' was a bad choice of words. Maybe better is - 'the average working stiff has many opportunities (once a paycheck for most) to DCA, and far fewer lump sum payments to deal with'.

-ERD50

 

[maddythebeagle]

It's funny, I'm a numbers guy, but I really do think the biggest advantage to DCA is to take the emotion out of it and turn it into a 'habit'.

For most investors, if you lump sum, and then the market goes down that next month, you feel bad, and it might affect your approach to investing. But doing it every month averages out those ups/downs. If you feel better about it, you are more likely to stick to it.

I advise DCA to people who ask.

-ERD50

very good points...that emotion thing is a killer I always find SG's comments on this issue good, also...Seems this topic comes up every few months.

 

[Ed_The_Gypsy]

I like ERD50's answer. Emotion (fear) is important. It must be taken into account.

I used to worry about DCA when I was a green investor, nervously moving money from one account to another. I slowly got the message that it is the long-term trends that are important. There is a discussion of this in "The Coffeehouse Investor" that is pretty clear. The way to win is to be invested, not be waiting to invest.

Having learned the value of diversity in assets (knowing ain't the same as doing, though ), if I had a windfall pot today available for the purpose I would invest it all at once--in a wide variety of assets. (My selection of exactly which assets is something I am always thinking about.)

One thing about doing it all at once is the act of commitment. If you do it a little at a time, it gets nerve-racking and the cash part of the pot is vulnerable to diversion to other projects.

 

[jazz4cash]

I wonder what the stats are for how many people that qualify to invest in a retirement plan actually invest.

If I had to guess based on what I've seen, I would say probably 10%.

Naw....not THAT low!

http://www.ebri.org/pdf/publications/facts/fastfacts/fastfact012406.pdf

 

[retire@40]

Naw....not THAT low!

http://www.ebri.org/pdf/publications/facts/fastfacts/fastfact012406.pdf

I guess those are the facts, but I still have a hard time believing 40% or more of the population contributes to a retirement plan.

Maybe only 10% of them contribute to their own IRA?

 

[samclem]

While 40-50% of folks may be contributing to 401Ks, an awful lot of folks withdraw all/most of their money when they change jobs.

" . . 42% of the 160,000 401(k) participants Hewitt surveyed in 2002 did just that. The cash-out rates were highest among workers in their 20s. Half of these workers raided their 401(k) accounts rather than rolling them over to IRA accounts or their new employers' plans."

http://articles.moneycentral.msn.com/RetirementandWills/InvestForRetirement/7MostCommon401kBlunders.aspx?wa=wsignin1.0



Disclaimer: In view of recent reactions to other posts which were deemed to be judgemental of the acts of others, it is worth noting that the above information is not intended to be judgemental in any way, or to imply that the actions taken by various non-FIRE-types are somehow "dumb." None of us can know that cashing out a 401(k) in order to trick out your car is dumb, and implying that might be is hurtful and mean.

 

[FinanceDude]

While 40-50% of folks may be contributing to 401Ks, an awful lot of folks withdraw all/most of their money when they change jobs.

" . . 42% of the 160,000 401(k) participants Hewitt surveyed in 2002 did just that. The cash-out rates were highest among workers in their 20s. Half of these workers raided their 401(k) accounts rather than rolling them over to IRA accounts or their new employers' plans."

http://articles.moneycentral.msn.com/RetirementandWills/InvestForRetirement/7MostCommon401kBlunders.aspx?wa=wsignin1.0



Disclaimer: In view of recent reactions to other posts which were deemed to be judgemental of the acts of others, it is worth noting that the above information is not intended to be judgemental in any way, or to imply that the actions taken by various non-FIRE-types are somehow "dumb." None of us can know that cashing out a 401(k) in order to trick out your car is dumb, and implying that might be is hurtful and mean.

I think most folks are doing that to pay off credit card debt and other bills...............

 

[jazz4cash]

I find the topic of participation rates for various benefit plans very interesting. Here is a link to a Vanguard Report on 401k demographics. As others have indicated the account average and median account balances are not as large as might be expected given participation rates > 40% and contribution rates of 7-8%:

(sorry for the long link if that sorta thing annoys)

https://institutional.vanguard.com/VGApp/iip/Research


MODERATOR EDIT: Shortened URL

 

[FinanceDude]

If DCA is a lie, then NOONE should do 401K contributions..........how does that compute??

 

[sgeeeee]

If DCA is a lie, then NOONE should do 401K contributions..........how does that compute??

Actually (common usage aside) when people make regular fixed contributions from their pay check they are not dollar cost averaging. They are making a lump sum contribution as soon as the lump sum is available. They are following the high percentage probability that lump sum beats DCA.

In order to DCA a regular pay check lump sum, they would need to make regular investments of the pay check surplus spaced out periodically during the interval between paydays. Of course doing that would probably eliminate corporate matching dollars they might otherwise be eligible to receive. So DCAing a regular paycheck would be stupid for at least two reasons.