Lump sum vs. DCA

[TomPie]

Hi All,

I need your help with the following question.

Imagine that you have 100k in cash right now. You have two options (to keep it simple). Invest it (global portfolio, index funds - to keep it simple again) as a Lump sum or dollar cost average over the next 1-3 years. Now the issue. Assume that the next crash will come in 1-3 years and the market will drop by 30%. The better option seems to be to keep cash and wait for the crash (basically time the market) since even if the market grew for the next 3 years by 10% and only then drop by 30% you're still almost even (ignoring the effect of compound growth).

Obviously, there are few built in assumptions in this scenario like:
- the next crash will come in 1-3 years (high probability, but as we know the market can stay longer irrational than we can stay solvent)
- the market will grow by 10% p.a.
- the market will drop by 30% it could be 15% but also 50% (or 80)
- trying to catch the falling knife problem - you never know when we've hit the rock bottom
- other???

So to summarize - I'm aware of the study that a Lump sum investment the better option in most cases (https://personal.vanguard.com/pdf/ISGDCA.pdf) but is it really now (10-years bull market, Shiller PE >32 etc)? When I'm looking at probabilities of each scenario and do couple simulations the best options seems to be: wait for crash -> invest or invest 1/3 (33k) every year for the next 3 years. I can provide some excel spreadsheets with my probabilities calculations. Interested, anyone?

Your thoughts?

thanks
Tom

[Dunkelheit]

IMHO the market doesn't care about the assumptions. You can build a zillion of different scenarios and in some years maybe none of them correspond to reality. Anyway, I wouldn't put all my money in the current market although John Bogle appears at my front door.

Another option is to invest a % of the total amount now (for ex: a 33%, a 50%...) and dca/value average from then or wait for a crash while you have a part invested because you never know when it will happen.

If you are sure a crash is going to happen soon, don't invest now, but then you will need to decide when to go in. Are you going to invest when the market is -30%? Or maybe better waiting for a -50% drop? Maybe 70%? Maybe never?

A 30% drop after 3 years raising 10% up makes a total portfolio loss of ~7%.

[arcyallen]

Historically lots of market gains have been missed by sticking cash on the side waiting for an "inevitable" dip. I don't know what will happen in the next three years, and despite what we all tell ourselves no one does. Short term fluctuations (a couple years qualifies) are based more on your fellow man's (Mr Market's) feelings than any actual fundamentals. If you can tell me how your fellow man will feel next year, you're doing better than me.

Statistically, it makes the most sense to invest it all, lump sum. I'm a huge fan of DCA when you can't lump sum it, but when you can "Time in the market beats timing the market". And if you're absolutely positive it's poised for a correction...or a bull run...it's time to challenge your viewpoint. Which is tough.

[jacob]

If you are concerned about value, you might want to look into value-averaging. It used to be a popular technique before the invention of mutual funds and the massive inflow of retail investors to the market which supported/promoted certain other behaviors such as DCA (via the monthly paycheck).

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

[Tyler9000]

It sounds to me like you're over-generalizing "the market". Not all asset allocations have the same level of downside risk as the stock market alone. Pick a portfolio that works for you in both good times and bad, and you'll probably be a lot more comfortable investing a lump sum.

As an example, I personally use the Golden Butterfly. Everyone is different and I'm definitely not suggesting you should invest the same way, but check out the Drawdowns chart and think about how that may change your assumptions.

[ThisDinosaur]

In the Intelligent Investor, Ben Graham advises always having between 25 and 75% of your money in stocks depending on market valuations. It's not all or nothing. This approach has a nice psychological benefit because you dont completely miss out on gains even if you're wrong. No FOMO.

[classical_Liberal]

...

[TomPie]

Thanks guys, this was helpfull. I should have noted that I have a portfolio of rental properties so I can afford to wait. Anyway, there’s a related question of diversificatiin vs concentration of your portfolio but I’ll raise it as a separate thread.

[TomPie]

And here is the answer to my question with data to back it up.

https://www.onefpa.org/journal/Pages/JA ... mance.aspx

TL;DR - if the Shiller CAPE 10 is above 31 DCA beats Lump sum investing 100% of the time.

[jacob]

Which is another way of saying that since future price appreciation is negatively correlated to CAPE, do not blow all your money (lump sum) at the high CAPEs (like now).

See fig 1 in TomPie's link for the graph. Also see
https://en.wikipedia.org/wiki/Price%E2% ... ock_market

[BRUTE]

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

brute was disappointed a few years ago when he discovered that Value Averaging does not, in fact, relieve the user from guesstimating the value (duh) of things. so it seems like just a sprinkle of chronological diversification on whatever value judgement has already been made.

[jacob]

VA requires a growth rate (say 6%) as an independent parameter input. If it's set too high (relative to market performance), one might not have enough cash to buy with. If it's set too low, one might end up with too much cash. However, results are always guaranteed.

I think it's more useful when horizons are longer and savings rates are lower and cash actually yields something.