Author Topic: Not One Person Here Believes in Dollar Cost Averaging!  (Read 12922 times)

Cycling Stache

  • Bristles
  • ***
  • Posts: 470
  • Age: 48
Not One Person Here Believes in Dollar Cost Averaging!
« on: August 28, 2017, 11:08:59 AM »
A slightly over-the-top title, but there are so many threads about when to put money in the market that it's time to blow up this irrationality.

First, statistically and rationally, you should put your money in the market as soon as it's available.  The market goes up over time, it is more likely to be up than down on a given day, and you can't time the market successfully to know which days are which, so play the odds and put your money in as soon as possible.

Second, nobody here actually believes in dollar cost averaging.  Many think they do, but because of a logical error that is easy to make but completely wrong.  The dollar cost averaging argument goes something like this: I have $100,000, and I will put in $10,000 per month for the next 10 months (or whatever time frame) so that if the market suddenly drops 50%, I won't lose 50% on all $100,000.

That is not probabilistically the best approach (see the first point), but the rationale behind it is clear--the idea that you're avoiding a major loss on all the money.  The real problem, though, is that nobody actually believes in the second approach.  Here's why.  Every day you have any money in the market, you can pull it out!  Every day, you are effectively making a decision whether to invest "all" of the money that you currently have in the market.

How many posts about dollar cost averaging have you seen that say after you complete the $100,000 investment (after 10 months or whatever), NOW pull the money out and start all over, $10,000 at a time.  Why not?  It's the exact same logic.  You have $100,000 that you can invest all at once (indeed, it's already in the market!), or you can pull it out and dollar cost average.

Yes, there might be slight tax consequences, but that's not the logical error here.  Every day that you have money in the market--that you KEEP in the market--you are making a decision to invest it rather than pull it out.  And unless you are constantly pulling the money out and dollar cost averaging it back in, you do not really believe in dollar cost averaging.

This is important because it highlights that the correct answer for everyone is to invest money in the market as soon as the money's available (consistent with asset allocation, etc.).  Even for those who instinctively lean towards the idea of dollar cost averaging, realize that not only is it irrational, it is not something you really believe in because you would not take out the money that's in the market and dollar cost average it back in.  So if you're not really committed to an irrational idea, there's no reason to do it in the first place.

Finally, this advice ignores the market timers.  But really, a market timer doesn't dollar cost average in either because there's no reason to do so if you can predict when the market is up or down.  You then just invest all at once when "the signs" tell you its time.

dandarc

  • Walrus Stache
  • *******
  • Posts: 5454
  • Age: 41
  • Pronouns: he/him/his
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #1 on: August 28, 2017, 11:26:24 AM »
I agree with you - pretty sure I said as much somewhere on this forum 2 or 3 years ago.

But, for better or worse, "Dollar Cost Averaging" has come to mean the same thing as "Regular, Interval Investing".  Kind of like how "literally" now can mean "figuratively", the definition has changed such that you have to take context into account to determine if someone is talking about what they do with their 401K, or an actual situation where there is a lump sum to invest to gather meaning.

What most people do with their 401K is a good way to do it - invest the money as soon as you have it.  Just happens that you have the money to invest in a very steady way.  While an academic analysis would call this Interval Investing or something similar, most people call this Dollar Cost Averaging now.

DCAing an actual lump sum is only useful as a way to get over fear for a reluctant investor.  As an investing strategy, it is just dumb for the reasons you've listed.

Cycling Stache

  • Bristles
  • ***
  • Posts: 470
  • Age: 48
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #2 on: August 28, 2017, 11:32:43 AM »
But, for better or worse, "Dollar Cost Averaging" has come to mean the same thing as "Regular, Interval Investing".  Kind of like how "literally" now can mean "figuratively", the definition has changed such that you have to take context into account to determine if someone is talking about what they do with their 401K, or an actual situation where there is a lump sum to invest to gather meaning.

You're exactly right and I should have clarified.  I'm referring to "real" dollar cost averaging.  Investing on intervals (when you receive your paycheck, etc.) is not dollar cost averaging.  It's just investing whatever money you have to invest at the time you receive it.

The only place where it blurs a little bit is 401(k) investing.  Technically, that should be front loaded too, but some people (like me) have employers who match a percentage of each contribution, so those people have to invest in the 401(k) over the entire year to get the full match.

Eric

  • Magnum Stache
  • ******
  • Posts: 4057
  • Location: On my bike
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #3 on: August 28, 2017, 12:23:52 PM »
I agree with the OP that DCA is dumb.  It's a crutch to attempt to assuage some investing fear.  Better to get over the fear by investing rationally and trusting the process.  Giving into the fear, even something as trivial as this, is a great way to continue to invest with your "gut" and royally fuck up at some point in the future, by thinking Here Comes Red Dow or similar.

Kind of like how "literally" now can mean "figuratively", the definition has changed such that you have to take context into account to determine if someone is talking about what they do with their 401K, or an actual situation where there is a lump sum to invest to gather meaning.

I refuse to accept this and will fight against it until my dying breath.


sol

  • Walrus Stache
  • *******
  • Posts: 8433
  • Age: 47
  • Location: Pacific Northwest
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #4 on: August 28, 2017, 12:42:43 PM »
OP has misunderstood the argument for DCA.  It's not that it's financially optimal, it's that is psychologically optimal.  Historically, the DCA approach has produced better results because it keeps people from pulling their money back out of the market for a loss.

But as was pointed out above, we usually just mean "invest regularly".

Car Jack

  • Handlebar Stache
  • *****
  • Posts: 2141
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #5 on: August 28, 2017, 01:05:01 PM »
Sol got it!  It's not mathematical.  It's psychological.  Just like why people use snowball method over avalanche.  Sure, snowball gets rid of small loans quickly, so you feel like you accomplished something.  If instead of seeing the loan disappear, you were presented with a "total owed" every month, this advantage would disappear and nobody would do it.

NoraLenderbee

  • Handlebar Stache
  • *****
  • Posts: 1254
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #6 on: August 28, 2017, 02:38:09 PM »
Most people don't have large lump sums to invest and thus aren't faced with this issue.

Quote
Every day that you have money in the market--that you KEEP in the market--you are making a decision to invest it rather than pull it out.  And unless you are constantly pulling the money out and dollar cost averaging it back in, you do not really believe in dollar cost averaging.

Unless you churn your account daily, you don't believe in DCA? That's just silly. Transaction costs alone would eat up most of the gain you might realize.

MrMoogle

  • Handlebar Stache
  • *****
  • Posts: 1136
  • Age: 38
  • Location: Huntsville, AL
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #7 on: August 28, 2017, 02:51:26 PM »
I agree with you - pretty sure I said as much somewhere on this forum 2 or 3 years ago.

But, for better or worse, "Dollar Cost Averaging" has come to mean the same thing as "Regular, Interval Investing".  Kind of like how "literally" now can mean "figuratively", the definition has changed such that you have to take context into account to determine if someone is talking about what they do with their 401K, or an actual situation where there is a lump sum to invest to gather meaning.

What most people do with their 401K is a good way to do it - invest the money as soon as you have it.  Just happens that you have the money to invest in a very steady way.  While an academic analysis would call this Interval Investing or something similar, most people call this Dollar Cost Averaging now.

DCAing an actual lump sum is only useful as a way to get over fear for a reluctant investor.  As an investing strategy, it is just dumb for the reasons you've listed.
From: http://www.investopedia.com/terms/d/dollarcostaveraging.asp
Quote
Dollar-cost averaging (DCA) is an investment technique of buying a fixed dollar amount of a particular investment on a regular schedule, regardless of the share price.

That fit's the "regularly investing" from a paycheck.  Does it exclude money from a paycheck?  I've never heard DCA is different than what I'm doing with my paychecks.

WhiteTrashCash

  • Handlebar Stache
  • *****
  • Posts: 1983
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #8 on: August 28, 2017, 03:04:27 PM »
Nah, I'm timing the market right now because it's so freaking easy at the moment. All I have to do is watch the news and wait for Trump to do or say something stupid. Then, the market drops and I buy. I've made a bunch of money that way so far. It's so predictable.

dandarc

  • Walrus Stache
  • *******
  • Posts: 5454
  • Age: 41
  • Pronouns: he/him/his
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #9 on: August 28, 2017, 03:19:10 PM »
From: http://www.investopedia.com/terms/d/dollarcostaveraging.asp
Quote
Dollar-cost averaging (DCA) is an investment technique of buying a fixed dollar amount of a particular investment on a regular schedule, regardless of the share price.

That fit's the "regularly investing" from a paycheck.  Does it exclude money from a paycheck?  I've never heard DCA is different than what I'm doing with my paychecks.
Quote investopedia, I'll counter with wikipedia:

https://en.wikipedia.org/wiki/Dollar_cost_averaging

Quote
Dollar cost averaging is not the same thing as continuous, automatic investing.

If you don't have the money in hand, there is no lump sum vs. DCA decision to make.  All I was pointing out that the even though the research is clear - DCAing instead of lump sum investing is worse, we're often not actually talking about what the research is looking at when we use the term.  The term DCA has come to be equivalent with continuous, automatic investing, which if we're analyzing it in an academic sense is more explicitly described as a series of lump sum investments.  If you don't have the money on hand to invest, there is no lump-sum vs. DCA decision to make.

The problem arises when we say something like "the research says DCA costs you money", and a reader doesn't understand the context and takes it as a reason to try and time the market, or to not invest at all.  So it is important that the context the term is used in is clear.

MrMoogle

  • Handlebar Stache
  • *****
  • Posts: 1136
  • Age: 38
  • Location: Huntsville, AL
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #10 on: August 28, 2017, 03:27:24 PM »
Well I learned something new today.

dandarc

  • Walrus Stache
  • *******
  • Posts: 5454
  • Age: 41
  • Pronouns: he/him/his
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #11 on: August 28, 2017, 03:35:07 PM »
Not to mention the way your typical financial advisor will use the term "Dollar Cost Averaging".  As a way to talk a reluctant investor into investing with him/her.  When this use happens, any kind of technical definition of the term or discussion of research is secondary to getting you to invest something today, possibly with a commitment to invest more later.  AUM, however they get there, help the advisor make money.

dycker1978

  • Pencil Stache
  • ****
  • Posts: 768
  • Age: 45
  • Location: Regina, Saskatchewan, Canada
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #12 on: August 28, 2017, 04:03:17 PM »
Nah, I'm timing the market right now because it's so freaking easy at the moment. All I have to do is watch the news and wait for Trump to do or say something stupid. Then, the market drops and I buy. I've made a bunch of money that way so far. It's so predictable.
If this were truly the case the market would never go up.  Trump always says something stupid...

Cycling Stache

  • Bristles
  • ***
  • Posts: 470
  • Age: 48
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #13 on: August 28, 2017, 04:48:39 PM »
OP has misunderstood the argument for DCA.  It's not that it's financially optimal, it's that is psychologically optimal.  Historically, the DCA approach has produced better results because it keeps people from pulling their money back out of the market for a loss.

This is exactly what I'm trying to address.  The appeal of DCA is completely a behavioral economic error.  But what I'm trying to show is that even the people tempted to make it don't take it to its logical conclusion.  Once you see the logical flaw (I'm not taking the $100k out once it's invested), it's easier to see that DCA in the first place makes no sense.

I'm obsessed with behavioral economics.  One way to address it is to model around it (i.e., make the default option investing in 401(k)).  Another is to show that the erroneous approach is logically ridiculous.  That's the typical $5 latte adds up to real money over time argument.

Here DCA is admittedly very tempting.  Look at me not losing tons of money if the market drops immediately after I invest my large sum.  But given that nobody is (or should be) pulling the large sum out of the market once it's invested, it makes it obvious how logically flawed the first decision is.  I'm hoping that if people see that, they will realize that DCA makes no sense in the first place, and invest accordingly (all at once).

Serial lump sum investing is completely different.  And I picked $100k (an admittedly large amount) as an example because it was the subject of a recent DCA post.

lifeanon269

  • Pencil Stache
  • ****
  • Posts: 566
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #14 on: August 28, 2017, 06:09:39 PM »
I think making a blaket statement that dollar cost averaging is irrational without regard to the investment itself is actually irrational.

Dollar cost averaging is an investment technique that is best used for highly volatile investments. So making a statement that dollar cost averaging doesn't work and then only looking at that technique mathematically against non-volatile investments isn't a fair assessment at all toward DCA.

Obviously, over the long haul, if you're simply investing in the stock market with an index fund, then just put your money all in as soon as possible. So long as you don't pull out at the wrong time, then you're likely to beat any returns compared to if you had dollar cost averaged your money into the market.

However, dollar cost averaging was meant to be an investment technique against highly volatile investments. For example, bitcoin is a highly volatile investment at the moment where dollar cost averaging makes a lot of sense. On a day to day basis it can easily flucate 10% or more often with no apparent reasoning and it is pretty much impossible to time. Dollar cost averaging makes a lot of sense in this case if you're investing a large amount of money since it will minimize your risk of a huge loss.

Also, while OP states that the argument for dollar cost averaging is to limit the downside risk to a large financial loss on the lump sum investment, you also can't forget about the upside that dollar cost averaging provides as well. Dollar cost averaging not only limits the risk of the downside market movement shortly after a lump sum investment, it also can increase the benefit of an upside movement in the market by continuing your investment as the market moves upward again and providing you an opportunity to buy much lower than you otherwise would've.

Again, if you're just putting your money in a non-volatile (or already diversified) investment, then dollar cost averaging doesn't really make much sense. But in investment situations where you're dealing with a highly volatile or risky investment that is extremely difficult to time, then it is mathematically wise to take advantage of dollar cost averaging to limit your risk.

I really don't understand the argument against it so long as it is used in the right situations. If you're for index funds and any other types of diversification, then dollar cost averaging is just one more method of diversification and has many good uses to limit one's risk. You risk missing out large lump sum gains, but you protect yourself from large lump sum loses. That's what any diversification strategy is all about.
« Last Edit: August 28, 2017, 06:16:48 PM by lifeanon269 »

bigchrisb

  • Handlebar Stache
  • *****
  • Posts: 1237
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #15 on: August 28, 2017, 06:54:41 PM »
I'm of the rational, numerical optimiser type.  I used to believe that if you had a large lump sum, you should get it invested on day 1, due to a higher expected value over time.

However, having been faced with lump sums twice in my life, I've changed my mind on approach.  This is because usually if there is a lump sum, it is due to a significant change in circumstances.  For me, it was the sale of my business, and the payout of a trauma / critical illness insurance policy.  Both of these resulted in changed future circumstances than in the time before the lump sum. 

A change in circumstances is a good reason to re-evaluate asset allocation, and is usually accompanied by a degree of psychological adjustment too.  Having been through it twice, I'm a fan of DCA, not because it results in a better mathematical outcome, but because it gives time to consider if your old plan suits your new circumstances.  For me, each of these events saw a lowering in leverage and a more risk adverse investment profile.

When I think about why people may have a lump sum to invest, the main ones I can think of are:
1. They have saved and never invested before.  There is a whole psychology of what investment looks like in asset classes other than cash going on at the same time.
2. Business sale.  Often this will be tied in with ending a business related job.
3. Inheritance.  Dealing with emotional issues of losing family.
4. Insurance payout. Usually associated with dealing with some sort of misfortune.
5. Lottery.  Lucky you.  However, as you have been spending money on poor expected value outcomes (buying lottery tickets), an argument about improving expected values probably isn't going to resonate!


Cycling Stache

  • Bristles
  • ***
  • Posts: 470
  • Age: 48
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #16 on: August 28, 2017, 07:29:42 PM »
A change in circumstances is a good reason to re-evaluate asset allocation, and is usually accompanied by a degree of psychological adjustment too.  Having been through it twice, I'm a fan of DCA, not because it results in a better mathematical outcome, but because it gives time to consider if your old plan suits your new circumstances.  For me, each of these events saw a lowering in leverage and a more risk adverse investment profile.

Bigchrisb, these are good points, but asset allocation is a different question and a different analysis.  The response to your point is that once the money (how ever much you allocated) was invested in the market, then what?  Surely the answer was that it stayed in until you needed it (or it's still there).  To some extent, I would also argue that times of emotional stress or changed circumstances are the worst times to try to make rational investment decisions, because you're most likely to make behavioral errors.  At that point, having a clear logical rule to follow is important.

But in investment situations where you're dealing with a highly volatile or risky investment that is extremely difficult to time, then it is mathematically wise to take advantage of dollar cost averaging to limit your risk.

Lifeanon269, you're correct that I was referring to the typical situation on this forum of investing in broad-based index funds.  That said, mathematically, volatility makes no difference to the analysis.  If you believe the market that you're investing in goes up over time, and you can't time it, then you should always invest at the earliest possible moment.  There is no exception to that rule.   (Of course, if you don't believe the market you're investing in goes up over time, then it's just gambling or market timing.)  The idea of limiting downside is always irrational if you're talking about money that you can afford to lose.  Not sure about that?  Take the 4% rule.  When in the 4% rule is it recommended that you pull all the money out and dollar cost average back in? 

Ultimately, I sought to make a simple point clearly.  There seems to be a lot of angst in the threads of when to start investing, how to invest a large sum, etc.  I believe that indecision leads people to delay investing.  While Sol is correct that DCA is often used as a crutch to help people get past that indecision, I think the next step is to identify the easy logical error in the thinking.  Because I believe that none of us are thinking about pulling all the money back out after dollar cost averaging in just to start over again.  Once you realize that, it's easy to see that dollar cost averaging doesn't make sense in the first place.

WhiteTrashCash

  • Handlebar Stache
  • *****
  • Posts: 1983
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #17 on: August 28, 2017, 08:02:17 PM »
Nah, I'm timing the market right now because it's so freaking easy at the moment. All I have to do is watch the news and wait for Trump to do or say something stupid. Then, the market drops and I buy. I've made a bunch of money that way so far. It's so predictable.
If this were truly the case the market would never go up.  Trump always says something stupid...

Well, Trump hasn't been very successful in making his agenda actually happen, so the cycle is generally 1.) Trump says or does something stupid, 2.) The market drops, 3.) Trump's subordinates/Congress/Supreme Court/Other government officials tell the public that Trump's not actually going to be able to do what he just said/wrote, and 4.) The market goes back up again. It's been going on this way for months now and it's gotten fairly predictable. I've made some nice cash.

NoStacheOhio

  • Handlebar Stache
  • *****
  • Posts: 2136
  • Location: Cleveland
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #18 on: August 29, 2017, 06:33:15 AM »
Nah, I'm timing the market right now because it's so freaking easy at the moment. All I have to do is watch the news and wait for Trump to do or say something stupid. Then, the market drops and I buy. I've made a bunch of money that way so far. It's so predictable.
If this were truly the case the market would never go up.  Trump always says something stupid...

Well, Trump hasn't been very successful in making his agenda actually happen, so the cycle is generally 1.) Trump says or does something stupid, 2.) The market drops, 3.) Trump's subordinates/Congress/Supreme Court/Other government officials tell the public that Trump's not actually going to be able to do what he just said/wrote, and 4.) The market goes back up again. It's been going on this way for months now and it's gotten fairly predictable. I've made some nice cash.

Realized or unrealized gains? Have you compared your results to regular interval investing (and holding) over the same time period?

maizefolk

  • Walrus Stache
  • *******
  • Posts: 7400
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #19 on: August 29, 2017, 07:16:20 AM »
I'm obsessed with behavioral economics.  One way to address it is to model around it (i.e., make the default option investing in 401(k)).  Another is to show that the erroneous approach is logically ridiculous.  That's the typical $5 latte adds up to real money over time argument.

DCA is a tool that lets a person trade a reduction in expected average return for a reduction in the size of the distribution of expected returns around the average. In this way it functions a lot like folks who keep more bonds or cash in their portfolio than is logically optimal. One can argue either approach is logically erroneous, but that only works if you discount the role of human emotion in decision making.

Short of hanging out with Data or Mr Spock from Star Trek, I have found it makes more sense to try to understand what emotional responses drive people to make decisions that seem logically flawed and then try to come up with the least expensive options that will still address the same emotional needs, rather that trying to use reason and logic to try to convince them to decide to change their emotional needs.* For example, in this case a person DCA into a high stock market position is going to suffer a much smaller penality in terms of lower expected return than if that same person decided to invest more of their money in cash/bonds.

*it took me an embarrassingly long time to learn this. Is important in relationships as well as in giving investment advice.

lifeanon269

  • Pencil Stache
  • ****
  • Posts: 566
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #20 on: August 29, 2017, 07:36:29 AM »
Lifeanon269, you're correct that I was referring to the typical situation on this forum of investing in broad-based index funds.  That said, mathematically, volatility makes no difference to the analysis.  If you believe the market that you're investing in goes up over time, and you can't time it, then you should always invest at the earliest possible moment.  There is no exception to that rule.   (Of course, if you don't believe the market you're investing in goes up over time, then it's just gambling or market timing.)  The idea of limiting downside is always irrational if you're talking about money that you can afford to lose.  Not sure about that?  Take the 4% rule.  When in the 4% rule is it recommended that you pull all the money out and dollar cost average back in? 

Ultimately, I sought to make a simple point clearly.  There seems to be a lot of angst in the threads of when to start investing, how to invest a large sum, etc.  I believe that indecision leads people to delay investing.  While Sol is correct that DCA is often used as a crutch to help people get past that indecision, I think the next step is to identify the easy logical error in the thinking.  Because I believe that none of us are thinking about pulling all the money back out after dollar cost averaging in just to start over again.  Once you realize that, it's easy to see that dollar cost averaging doesn't make sense in the first place.

I absolutely agree on the statistically better outcomes with regard to lump sum investing versus DCA.

However I think it is important to clarify the difference between "mathematically" and "statistically" better outcomes. You keep saying that lump sum investing produces mathematically better outcomes, but that isn't true when I can clearly show results where dollar cost averaging would produce mathematically better financial outcomes in hindsight. I don't mean to argue semantics, but it is an important difference. Because, while statistically lump sum investing is favored over DCA historically, mathematically there are many times where DCA produces much better results.

The reason why I point out this difference is because this difference is where one's risk appetite is born. There probably wouldn't be too many people here who would argue for someone to alter their personal risk appetite in order to shift them from a portfolio balance of 60/40 stocks/bonds to say 80%/20% stocks/bonds.

One's investment portfolio is a direct reflection on one's risk appetite and the potential financial outcomes are similar enough that it would be more difficult to alter one's risk appetite than to simply adjust to the difference in financial outcomes between the two portfolio balances. While we can look back and say that one performs statistically better than the other, there is obviously more risk in one over the other in order to receive those better financial outcomes. That also means that mathematically there are periods in time where the less risky portfolio balance performs better than the riskier one. Saying there is no exception to the rule is completely false. Tell that to someone who retired in 1965 and by all means thought they statistically saved enough for retirement. You may try all you want to claim that historical statistical analysis removes inherent risk with investing, but it doesn't. The same thing can be said about DCA with regards to risk.

I think it is also important to point out that since you're talking here about long term investment which is where lump sum investment statistically beats DCA, then it is also important to point out that in the long term, the difference between lump sum investment and DCA is almost negligible if DCA is performed over the short term (say <1 year). So if utilizing DCA over the short term to minimize short term risks means sacrificing an almost negligible amount over the long term, then to many people with more aversion to risk, it might be worth it to them. Again, DCA is just another form of diversification that is used to minimize risk at the cost of upside. That being said, like I said earlier, DCA mostly makes sense with more volatile (and hence riskier) investments, so I'd question why someone with risk aversion is investing in riskier investments in the first place!!  :)

I also don't understand your 4% analogy. The 4% rule was born out of the fact that averages are meaningless to the individual retiree. A retiree who retired in 1965 could careless about what averages say. That's why it is more important to stay flexible so that you can beat averages.

If you're using the 4% rule, then that means you're already FIRE'd which means that the 4% rule only applies if you are invested and committed to being invested in the stock market. The 4% rule doesn't apply to other investment classes which is why it wouldn't make sense to exchange your stock investment for a currency investment (USD) at any time during your retirement, which is what you'd essentially be doing.

My point is, DCA is just another investment tool or methodology that can be used to minimize risk. Minimizing risk will always have both upsides and downsides. Focusing merely on the downsides of risk mitigation while ignoring the upsides in my opinion misses the point of risk mitigation entirely.

Cycling Stache

  • Bristles
  • ***
  • Posts: 470
  • Age: 48
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #21 on: August 29, 2017, 07:55:50 AM »
Short of hanging out with Data or Mr Spock from Star Trek, I have found it makes more sense to try to understand what emotional responses drive people to make decisions that seem logically flawed and then try to come up with the least expensive options that will still address the same emotional needs, rather that trying to use reason and logic to try to convince them to decide to change their emotional needs.*

Maizeman, I needed one of your cool charts to help make the math point!

There are two emotional needs in play, I believe.  As I tease them out, I realized that I was more strongly responding to the second one. 

First, fear of loss if the market drops shortly after investing.  That is addressed in part by realizing that's exactly what you're going to do once you've finally invested all the money.  Leave it in and it's all subject to the ups and downs.  If you're not going to keep pulling the money out trying to DCA back in, then why DCA in the first place. 

Second, and a little more unstated, is that it feels like a really big decision to invest a large sum of money all at once.  The key insight for me is that every single one of us with money in the market is making that exact same decision every day.  We don't think about it as a big deal because of course we're going to keep the money in and not try to market time.  But once you realize that the decision to invest a large sum is logically identical to the decision being made to keep the money in the market, it's a lot less scary decision.  I think that's the fundamental point I was trying to get at.

Last night, I made the decision to lump sum a very large sum of money into the market.  Not because of market timing, not because of reading the tea leaves, but because I decided to keep my investments in the market another day.  It's the same decision I've made for the last few years.  Once people realize that's exactly what they're doing when "deciding" to keep the money in the market, it's easier to move forward to with the initial decision to put in the market.  That's the attempt to address the underlying emotional issue.

A final point, and interesting parallel.  I try criminal cases.  One of the key points I've learned to address with the jury in closing is that the case they're hearing is just like criminal cases all across the country every single day.  That is to address the behavioral error that the case they've been selected for must be special in some way because they're there, and therefore it's going to be a really hard decision.  It's not special.  It's just the case they happened to be selected for.   That's important to put jurors back into the rational decision making process of guilt and innocence, rather than feeling like something special or different is happening that makes this case much harder than any other criminal case being tried to any other jury.  In the same way, lump sum investing feels like a really hard, unprecedented decision (because it's new to that person), when in fact it's the same decision that every single person with money in the market is making every single day (even if they don't think about it because the decision to stay invested seems so easy to make).

talltexan

  • Walrus Stache
  • *******
  • Posts: 5344
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #22 on: August 29, 2017, 08:05:02 AM »
DCA is about maintaining an allocation less than 100% stocks for a certain amount of time.

Investing Policy Statements are about maintaining that same allocation forever.

So why is it different to say, "I will be 25% in stocks for one year" than "I will be 25% in stocks forever?"

It's the time horizon that's the variable we need to be tracking.

J.Milly

  • 5 O'Clock Shadow
  • *
  • Posts: 33
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #23 on: August 29, 2017, 09:07:01 AM »
I think there's 2 large factors. One if you have a large lump you should dump it in immediately. However for those of us that take money from our pay every 2 weeks and auto invest, technically we are DCA by not letting a pile build up then going in, so technically i do both. Just my 0.02

lifeanon269

  • Pencil Stache
  • ****
  • Posts: 566
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #24 on: August 29, 2017, 09:29:13 AM »
DCA is about maintaining an allocation less than 100% stocks for a certain amount of time.

Investing Policy Statements are about maintaining that same allocation forever.

So why is it different to say, "I will be 25% in stocks for one year" than "I will be 25% in stocks forever?"

It's the time horizon that's the variable we need to be tracking.

That's exactly it. Which is why I think it is unfair of OP to make the assessment that if you wouldn't continue to DCA over the entire lifetime of your investment, then DCA doesn't make sense at all.

It is an extremely different investment choice between being 25% in stocks for 1 year versus being only 25% invested over the long term. Saying that someone who chooses to DCA over one year is no different than someone who dollar cost averages in perpetuity is completely wrong. It may seem like logically the choices are similar, but they couldn't be further from being different. Even if you were to DCA back out of the investment, that is still a 50% allocation between the two exchanged investment classes which is probably drastically different than what the investor would likely be looking for long term. Again, there is often a big difference between short term and long term risk appetite for each individual.

It is clear that OP is simply rationalizing his own recent personal decision in how his/her own lump sum was invested. I'm not arguing against that decision. I would've made the same choice myself. However, it is important to realize that 30 years from now the difference between making the decision that was made versus if the decision was to DCA over the course of a year will likely (statistically speaking) turn out to be negligible. However, don't discredit the use cases/benefits of DCA simply because historical statistics may not favor it when compared to investing in one lump sum. I think you're trying too hard to over analyze historical statistics in order to weigh one option more heavily than the other without taking it in context of the individual. Ultimately the more important decision is that one has an investment strategy to begin with and sticks to it.

History may well look back on this post and show you how you would've fared better had you dollar cost averaged your investment in over the course of a year. However, would that really matter to you since you know at the time you made your decision as part of an investment strategy you put forth and committed to? I think the inverse of this would likely be true for anyone utilizing a DCA strategy over the course of the next year.
« Last Edit: August 29, 2017, 09:33:02 AM by lifeanon269 »

maizefolk

  • Walrus Stache
  • *******
  • Posts: 7400
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #25 on: August 29, 2017, 11:23:18 AM »
There are two emotional needs in play, I believe.  As I tease them out, I realized that I was more strongly responding to the second one. 

First, fear of loss if the market drops shortly after investing.  That is addressed in part by realizing that's exactly what you're going to do once you've finally invested all the money.  Leave it in and it's all subject to the ups and downs.  If you're not going to keep pulling the money out trying to DCA back in, then why DCA in the first place. 

Second, and a little more unstated, is that it feels like a really big decision to invest a large sum of money all at once.  The key insight for me is that every single one of us with money in the market is making that exact same decision every day.  We don't think about it as a big deal because of course we're going to keep the money in and not try to market time.  But once you realize that the decision to invest a large sum is logically identical to the decision being made to keep the money in the market, it's a lot less scary decision.  I think that's the fundamental point I was trying to get at.

That's a good way of looking at the breakdown of the different emotions at play. However, I'd summarize #1 a little differently. People have a disproportionate emotional aversion to making "bad" decisions. This can cause decision paralysis (even when doing nothing is empirically worse than any of the choices they are trying to chose between). DCA lets such a person deal with the fear that they'd otherwise pick the single worst day to invest their money (even if the consequences of picking that worst day are still a lot less bad than leaving their money sitting on the sidelines in cash) and get their money into the market.

Completely agreed on the second emotion, and I think your approach (every time you decide not to take all your money out of the market is an equivalently big decision) is a good one that has a reasonable shot at getting people to change their perspective on that particular source of emotional resistance.

I thought I actually had a graph of the trade off between lump sum vs DCA over a year in terms of reduced average return vs deviation of the return, but I'm not having any luck digging it up. May call for a redone analysis.

lifeanon269

  • Pencil Stache
  • ****
  • Posts: 566
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #26 on: August 29, 2017, 12:27:41 PM »
Since we're on the subject of investment psychology, could the same emotional fears be said of lump-sum investments?

By that, I'm talking about FOMO or "fear-of-missing-out"?

Let me start with an example. I invest regularly in bitcoin, but my investment strategy is direct, deliberate, and also unique. I say unique in the sense that, every paycheck I receive, I move it over into bitcoin (not all, but most of it). From there I allocate some of the bitcoin that was purchased as long-term savings (ie, some for retirement), and then some of it is used as my spending money each month. However, rather than taking my paycheck every other Friday and immediately purchasing bitcoin with its full amount, I dollar cost average my purchase out to certain smaller amounts every week. Now, you could say that this is more like interval investing, but I say it is more like dollar cost averaging because I am choosing to not deliberately invest the lump-sum payment every other week into bitcoin immediately. While it may not be on the same scale, I still consider it a deliberate choice to dollar cost average.

The reason I do this is because I feel bitcoin is a highly volatile investment and having a purchase every week as opposed to a large purchase every other week gives me more granularity in my investment purchases and would likely smooth out the noise of bitcoin in the short term. "Short-term" is the key term here. Over the long term, there is likely no recognizable difference in either strategy. So I avoid the impulse of "FOMO" with each paycheck in order to better provide myself short-term stability. This is what I am referring to with regard to short-term versus long-term risk analysis. I am very long on bitcoin, but that doesn't mean I am blinded to the short-term risks in that investment.

I realize this situation is vastly different than a much larger lump-sum investment over the long term that is compared to dollar cost averaging that same investment over a year. But my point is that the psychological impulses surrounding FOMO could be persuasive into making one think that they better get their money in there today or else they're going to gravely cripple their investments over the long haul. Making a claim that there are no exceptions where DCA would be of benefit to anyone here or that it is illogical because of past historical performance ignores all aspects of true risk analysis for the individual making the decision. The reality is that if you're talking about long term performance, the difference between the two over that same long term cycle is likely negligible. Therefore, the decision for short-term risk mitigation is up to the individual to make based on their own situation.

This is why I believe that there are certainly use cases where dollar cost averaging makes sense and there are cases where investing the lump sum make sense, but I don't think blanket statements that determine that one is better than the other is sound investment advice in either case.

talltexan

  • Walrus Stache
  • *******
  • Posts: 5344
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #27 on: August 30, 2017, 08:47:19 AM »
Wow, I didn't expect the bitcoin twist there.

Suppose your first action upon receiving your paycheck were converting immediately to gold. Would anyone here think that was weird?

Dicey

  • Senior Mustachian
  • ********
  • Posts: 22319
  • Age: 66
  • Location: NorCal
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #28 on: August 31, 2017, 10:55:35 AM »
Wow, I didn't expect the bitcoin twist there.

Suppose your first action upon receiving your paycheck were converting immediately to gold. Would anyone here think that was weird?
Dicey raises hand.

talltexan

  • Walrus Stache
  • *******
  • Posts: 5344
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #29 on: September 01, 2017, 12:00:45 PM »
I understand that some people like gold, or are suspicious of our Federal Reserve System.

Mr Mark

  • Handlebar Stache
  • *****
  • Posts: 1229
  • Location: Planet Earth
  • Achieved Financial Independence summer 2014. RE'18
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #30 on: September 01, 2017, 12:49:50 PM »
I understand that some people like gold, or are suspicious of our Federal Reserve System.

This is why I like owning a piece of every traded company in the U.S. stock market rather than just pieces of paper or bits and bites in a bank account.

I also fully agree with maizemans psyc point. People hate losses more than they like equivalent  gains.  Yes, statistical optimum on average says put that big lump sum in. But the emotional pain of imagining doing that before a big correction is too much.

It's human. That's why we're mustachians. we (try our best to)  use math and logic to overcome our harmful emotional reactions.

zephyr911

  • Magnum Stache
  • ******
  • Posts: 3619
  • Age: 45
  • Location: Northern Alabama
  • I'm just happy to be here. \m/ ^_^ \m/
    • Pinhook Development LLC
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #31 on: September 01, 2017, 12:52:17 PM »
Wow, we are seriously splitting hairs.

DCA is simple: keep buying, regardless of market fluctuations, for a lower average cost of shares.

Unless you panic and avoid the market when it's down, you're doing DCA. I've bought stock with literally every paycheck for the past decade or more. That is DCA, broadly. Whether I park a windfall in the market or do installments, I am still a DCA investor because I'm buying the market all the time, whether it's up, down, or sideways.

On the opposite side:
The saddest investment fact I've heard in the whole last year was that the average TSP C Fund investor (think S&P 500) only earns about 2% despite the fund averaging 10% historically because they so often chase the highs and panic-sell the dips.

Cycling Stache

  • Bristles
  • ***
  • Posts: 470
  • Age: 48
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #32 on: September 01, 2017, 05:53:56 PM »
Wow, we are seriously splitting hairs.

DCA is simple: keep buying, regardless of market fluctuations, for a lower average cost of shares.

I've bought stock with literally every paycheck for the past decade or more. That is DCA, broadly.

Just to maintain the integrity of the original point, let me be clear on what DCA is as I've argued it.  Otherwise the post doesn't make any sense.

DCA is the decision to take a set amount of money that you have in your possession and want to invest, and invest it in chunks over time rather than all at once.

Although I can see from this thread it has become a popular alternative definition of DCA, investing your paycheck every two weeks is actually periodic lump sum investing because you're investing the money you have available to you as it becomes available.  That is of course mathematically ideal.   The original analysis doesn't deal with that situation (other than to applaud it).  I too invest most of my paycheck every two weeks.

The decision to invest a sum you currently have in chunks rather than all at once is what triggers the fear of loss if you invest right before a big downturn.    That is the point I sought to address.  There's very little equivalent fear when you invest an amount that is relatively small (paycheck) compared to the amount you already have invested (although I bet many of us check to see what the market does the day after it goes in!).

maizefolk

  • Walrus Stache
  • *******
  • Posts: 7400
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #33 on: September 01, 2017, 07:28:35 PM »
Not my favorite graph ever, but hopefully it makes the point.



Using 1,000 randomly samples start months from the Shiller dataset I either ran the scenario forward for 60 months (5 years) based on an initial investment of $100k for the lump sum method adjusting for changes in stock prices and dividend payments each month. For comparison, I also ran a scenario from the same start month where I started the portfolio as $0 and increased it by $8,333.33 each month for the first 12 months, then ran the scenario as before for the remaining 48 months. As you can see the results are largely comparable, but the lump sum method does come out ahead by an extra 6.5% of growth over five years (based on average return) or 4.1% (based on median returns). That translates to a loss of $6,500 or $4,100 over five years, respecitvely.

That loss of expect return over five years buys you a decrease in the standard deviation of end value from ~$60,000 to ~$54,500.

In 299 of 1000 start months, DCAing in over a 12 month period resulted in a better return than investing it all in the first month, in the remainder of cases lump sum provided the better return.

In the violin plot on the left hand side the blue bar marks the average value across all 1,000 simulations and the green bar marks the median value across all 1,000 simulations.

So as in many cases, data analysis tells us what we already know, in the average year, lump sum investing is going to beat dollar cost averaging, and something I didn't have a good sense for before, that the chances of out ahead from DCAing over a 1 year period are about 1 in 3.

moof

  • Pencil Stache
  • ****
  • Posts: 809
  • Location: Beaver Town Orygun
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #34 on: September 03, 2017, 08:26:29 PM »
I dollar cost average, by buying as soon as I have un-committed cash.  It it pretty even over the year.

I don't understand the front loading crowd, as this means you ended the prior year with cash sitting around waiting for Jan 1, WTF?  You should ideally buy Roth or taxable shares (higher value when sold after taxes, so you want more appreciation here than in IRA/401k shares bought in the same year) first and back load maxing out of any deductible traditional IRA and 401k's beyond matching minimums but the opposite usually gets advocated.  You can also better optimize by back loading tax deductions, but that is playing with fire if you mess up.  Doing either "optimized strategy" takes effort, so I just set it on auto pilot with equal deductions and buy extra taxable account shares when spare cash builds up.  The difference in net value for any of the three approaches is pretty minimal compared to the damage I could do by messing up.
« Last Edit: September 05, 2017, 10:44:18 AM by moof »

steevven1

  • Stubble
  • **
  • Posts: 150
  • Location: Florida
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #35 on: September 03, 2017, 09:20:30 PM »
Thanks, OP, for posting this argument. I have been making the exact same argument to others for a while now.

zephyr911

  • Magnum Stache
  • ******
  • Posts: 3619
  • Age: 45
  • Location: Northern Alabama
  • I'm just happy to be here. \m/ ^_^ \m/
    • Pinhook Development LLC
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #36 on: September 06, 2017, 09:03:49 AM »
Just to maintain the integrity of the original point, let me be clear on what DCA is as I've argued it.  Otherwise the post doesn't make any sense.

DCA is the decision to take a set amount of money that you have in your possession and want to invest, and invest it in chunks over time rather than all at once.

Although I can see from this thread it has become a popular alternative definition of DCA, investing your paycheck every two weeks is actually periodic lump sum investing because you're investing the money you have available to you as it becomes available.  That is of course mathematically ideal.   The original analysis doesn't deal with that situation (other than to applaud it).  I too invest most of my paycheck every two weeks.

The decision to invest a sum you currently have in chunks rather than all at once is what triggers the fear of loss if you invest right before a big downturn.    That is the point I sought to address.  There's very little equivalent fear when you invest an amount that is relatively small (paycheck) compared to the amount you already have invested (although I bet many of us check to see what the market does the day after it goes in!).
The first part of your post supports that clarification, but you invited my specific objection here:
Quote
And unless you are constantly pulling the money out and dollar cost averaging it back in, you do not really believe in dollar cost averaging.
The idea of trying to sell and re-buy at lower cost is the ANTITHESIS of DCA. The thesis is: market timing is difficult, even experts suck at it, and the best route is just buying all through peaks and valleys. Because you don't actually know which one the market is in until later, you just buy as cash comes right now. DCA is 100% about how you buy, and virtually every flavor of it is staunchly B&H, so saying only constant churners believe in it is extremely distracting.

If your focus is really just the question of lump-sum vs. installments for a windfall, I'd suggest you remove the other part of the OP and change the clickbaity thread title to enhance that focus.

ooeei

  • Handlebar Stache
  • *****
  • Posts: 1142
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #37 on: September 07, 2017, 07:01:01 AM »
DCA as you've defined it is a way to convert from cash to stocks. It's done with the intention of being all in on stocks (or whatever your asset allocation is) at the end of a certain time period. To say that it doesn't make sense unless you keep pulling money back from stocks into cash and then going back is pretty odd. I get what you're saying, but I don't think most people see DCA as an optimal long term investment strategy, they see it as getting into the pool by the stairs rather than off the diving board. In either case you plan on getting in the pool, I wouldn't expect the stairs person to get out and walk up and down the stairs over and over.

Most people know lump sum is mathematically the better choice, and that being in the market is better than being out of it. They want to avoid the psychological trauma of investing $100,000 and then watching it drop the next day. While mathematically they're hedging their bets wrong, psychologically they're probably hedging their bets right (for themselves anyway). There was a rise in suicides in men after the 2008 financial crisis. Is this all due to investing? Probably not, but psychology is a real and powerful thing. If you're the type of person who'd be up all night if your parent's inheritance lost $25k the day after you jumped in on the stock market, especially if maybe that's the majority of your net worth, I think it's reasonable to DCA your way in for your own sanity.

JohnSteed

  • 5 O'Clock Shadow
  • *
  • Posts: 15
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #38 on: September 11, 2017, 02:15:31 PM »
Historically, the DCA approach has produced better results because it keeps people from pulling their money back out of the market for a loss.



Do you have proof or are you just making stuff up?  How can anyone actually measure the performance based on proclivity to pull the money out? 

NoStacheOhio

  • Handlebar Stache
  • *****
  • Posts: 2136
  • Location: Cleveland
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #39 on: September 12, 2017, 05:52:26 AM »
Historically, the DCA approach has produced better results because it keeps people from pulling their money back out of the market for a loss.



Do you have proof or are you just making stuff up?  How can anyone actually measure the performance based on proclivity to pull the money out?

If you look at fund performance versus individual investor returns it's not uncommon for investors to underperform their own funds.

JohnSteed

  • 5 O'Clock Shadow
  • *
  • Posts: 15
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #40 on: September 12, 2017, 11:21:06 AM »
Historically, the DCA approach has produced better results because it keeps people from pulling their money back out of the market for a loss.



Do you have proof or are you just making stuff up?  How can anyone actually measure the performance based on proclivity to pull the money out?

If you look at fund performance versus individual investor returns it's not uncommon for investors to underperform their own funds.

Sure but that doesn't mean that DCA had anything to do with it.

NoStacheOhio

  • Handlebar Stache
  • *****
  • Posts: 2136
  • Location: Cleveland
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #41 on: September 12, 2017, 12:21:02 PM »
Historically, the DCA approach has produced better results because it keeps people from pulling their money back out of the market for a loss.



Do you have proof or are you just making stuff up?  How can anyone actually measure the performance based on proclivity to pull the money out?

If you look at fund performance versus individual investor returns it's not uncommon for investors to underperform their own funds.

Sure but that doesn't mean that DCA had anything to do with it.

No, it's the behavioral economics side (pulling money out at the bottom, buying in at the top). It shows underperformance based on the proclivity to pull money out. Which I think was the question?

zephyr911

  • Magnum Stache
  • ******
  • Posts: 3619
  • Age: 45
  • Location: Northern Alabama
  • I'm just happy to be here. \m/ ^_^ \m/
    • Pinhook Development LLC
Re: Not One Person Here Believes in Dollar Cost Averaging!
« Reply #42 on: September 13, 2017, 09:43:25 AM »
I wish I could unsubscribe for semantic quibbling.

Historically, the DCA approach has produced better results because it keeps people from pulling their money back out of the market for a loss.



Do you have proof or are you just making stuff up?  How can anyone actually measure the performance based on proclivity to pull the money out?

If you look at fund performance versus individual investor returns it's not uncommon for investors to underperform their own funds.

Sure but that doesn't mean that DCA had anything to do with it.

No, it's the behavioral economics side (pulling money out at the bottom, buying in at the top). It shows underperformance based on the proclivity to pull money out. Which I think was the question?

I posted a specific example from a retirement class I attended recently, showing that TSP investors in the C fund, which tracks the S&P 500 and tends to match its historical ~10% total return, chase highs and panic sell at lows so pervasively that the average individual only earns a return equal to inflation. I don't have the raw data, but it was well documented by the financial advisory group that provides pre-retirement training to federal employees.

The default investing method for federal employees is equal payroll contributions (DCA) so any deviations (underperforming the fund) are the direct result of not doing that.