A New Game Plan for Young Investors? FPAnet article

[Telly]

This article in the August FPA Journal caught my eye. It's main purpose was:
"While the debate concerning dollar cost averaging (DCA) versus lump sum (LS) investing rages on, we believe the debate should center on how to accurately measure investors’ true returns."

However, this is what caught my eye:
"We show there is a decrease in ending wealth variability for a more balanced approach to investing. This implies that even young investors might consider a balanced asset allocation to reduce the likelihood of such a divergent outcome"

The usual advice has been that a young investor go 100% on equities, then taper down the EQ % as they get considerably older. This article suggests otherwise. It also reminded me of a Scott Burns article some years ago where he looked at "balanced" funds, and pointed out that the return over long periods of some balanced funds were just a small amount lower than, say, the S&P 500, with much less volatility.

Even if you don't want to read the whole article, at least read the Executive Summary, and the Summary and Conclusion. That may then entice you to read/contemplate the rest.

For ER'd folks, it's water over the dam. But it may help your kid's water


Return Measures and Dollar Cost Averaging

 

[caninelover]

Interesting article, thanks. I only read the exec summary, and the summary/conclusions but the one question on the assumptions I would have is that it seems to assume that people invest the same annual amount in periodic contributions. I know when I started contributing to my 401k plan, I started at 6% and increased annually by 1 or 2% per yr until I maxed out. Also, when I started w*rking in the 1990's I was certainly making much less money than I am now, so the % employer match is also not constant. I do think the dollar weighted method is a more accurate way of measuring total return but our 401k administrator (as do most I think) uses the simple averaging method, so I'm not sure how I'd go about doing the calculation with my own portfolio? Still interesting to think about.

 

[FIREd]

My portfolio has always been much more conservative than what experts typically recommended for people my age exactly for that reason.

If you put all your money in safe investments (like treasuries or CDs), your money won't grow much BUT you can pretty much pinpoint with good accuracy when you will reach your financial goal. "If I save $50,000 a year and my 30 year treasuries yield 5%, I will have 1.5M in savings 20 (or whatever) years from now".

If you put all your money in speculative investments, the returns could potentially be much greater but there is no telling when you will reach your financial goal because of the volatility. It could be much sooner than planned if you hit a strong bull market. But if the markets hit the skids, it could be much, much later than planned too. "If I invest $50,000 a year in the stock market, I will have $1.5M in investments in years". Your guess is as good as mine...

I don't want this kind of uncertainty though I still want a bit of growth. So I favor a 50/50 (+/-) portfolio.

 

[ziggy29]

If you put all your money in safe investments (like treasuries or CDs), your money won't grow much BUT you can pretty much pinpoint with good accuracy when you will reach your financial goal. "If I save $50,000 a year and my 30 year treasuries yield 5%, I will have 1.5M in savings 20 years from now".

If you put all your money in speculative investments, the returns could potentially be much greater but there is no telling when you will reach your financial goal because of the volatility. It could be much sooner than planned if you hit a strong bull market. But if the markets hit the skids, it could be much, much later than planned. "If I invest $50,000 a year in the stock market, I will have $1.5M in investment in years". Your guess is as good as mine...

To me, it's basically the difference between people who invest to maximize the average-case value of their portfolio over time, and those who invest to minimize the chances of falling short of their goals over a period of time.

For example, over the next 20 years it may well be that close to a 100% equity portfolio would maximize my expected asset value at the end of that period. But I'm more concerned about what I think I will need in 20 years and choosing an AA that minimizes my chances of falling short.

 

[LOL!]

Otar described how folks just starting out should have significant fixed income in their asset allocation. If one includes an emergency fund in your portfolio, you sort of get this automatically. As you become financially independent or retired the concept of an emergency fund is kind of moot since your entire portfolio could be used as an emergency fund and "6 months of expenses" is probably less than 2% of your assets anyways.

 

[FUEGO]

Interesting article. I have a number of issues with it however.

One thing that it completely ignores (best I can tell, I skimmed the middle of it) is a consideration of your "Human Capital" as covered in numerous other JFP articles by Milevsky and others. In a basic sense, even if your actual DCA'd investment returns are highly divergent from your geometric mean investment returns, so what? After only a 20 year period of investing (say, starting at age 25), you likely still have another 10-20 year period of human capital that can be converted into cash flow by working and earning an income. By this I mean to say that the individual investor can typically decide to continue converting their human capital into cash flow if the particular time series of stock investment returns has been unkind (as it has 1990-2009 as shown by their study). Wait a couple years and it will probably look rosier for the 100% equities investor versus the 50/50 balanced investor.

Why look at a 20 year period instead of a 30 year period? Or 35 year period? That seems to be a more typical investment horizon for the average working Joe. My guess is that over longer periods of time, DCA average returns converge with geometric mean returns since you get more of the broad market moves and economic cycle.

I ran a quick spreadsheet on the hypothetical $12000 a year invested each year from 1990-2009, and up until the last 2 years, you come out ahead with 100% equities instead of 50/50 virtually every year from 1991-2007 (except 2002). Them's odds I can live with! I can just work an extra couple years to offset that risk.

In my thinking, I would rather have a 70-80% chance of retiring earlier, potentially much earlier with a high equities portfolio. Which guides my current decision to maintain 100% equities in my portfolio. I don't have a concrete retirement date set, like most folks I would guess. If I get to my target amount at age 36, I'll get there. If at age 35 the market tanks and it takes me another 4-5 years to retire, I'll get there. Add in greater tax efficiency for equities vs fixed income and it makes the high equities portfolio an even better bet.

This nugget toward the end of the article is worth a lot however: "If one does not believe a more conservative approach is prudent from the very beginning, there is definitely a need for the downside protection once you have a sizable portion of your eventual ending wealth at stake."

Those who have high equities portfolios always have an option to lock in any gains from equities. During times of high valuations, one can permanently change asset allocations to a more conservative asset allocation to essentially "lock in" those gains from a high equities portfolio. As the article points out in Figure 6, for a 40 year accumulation period, the last half of your target wealth is acquired in the last 8 years of investing. It would seem prudent that at some point when you have accumulated a sizable portion of your target portfolio value to reduce your exposure to return volatility (by going to a more conservative asset allocation, for example). I'm not sure what the magic number is. 50% of your target portfolio value? 75%? I'm at roughly 50% of my low end target portfolio value and I'm starting to think about fixed income exposure. The fixed income yields don't really interest me at this point though. And I'm still waiting to get those returns that 100% equities have historically provided (on average) so that I can lock them in...

 

[Nords]

I think DCA was debunked long ago as a returns-boosting mechanism. Its main value has been enforcing the routine discipline of a "set and forget" approach. And you all know that we all talk about "saving your pay raises" yet somehow that approach gets skipped here too.

The annoying aspect of many of these studies is that they simulate a flawed subset of reality and then attempt to derive life lessons from it. For example, the "normalized distribution returns" bell curve isn't even normal, and its returns exhibit both skewness and kurtosis. Yet 99.9% of the studies (with the exception of Taleb) gloss right over this point and start talking about what the mathematics prove we should do. Well, sure, the math works great as long as the stock market matches that artificial mathematical construction. Oops.

Then, as has been pointed out, it's assumed that we invest the same amount in an identical fashion for decades, followed by withdrawing a fixed amount in a similar manner for 30 years. Someday I'd like to meet this fictional investor.

I think the significant points of this study are drowned out by the effect of the "simplifying" assumptions. Its main value is to force investors to reflect upon the reasons behind their asset allocation and their [-]market timing[/-] rebalancing criteria, especially after retiring.

I think all financial researchers should post five years of tax returns with their papers... let's see who's been swimming naked!

 

[Telly]

Then, as has been pointed out, it's assumed that we invest the same amount in an identical fashion for decades, followed by withdrawing a fixed amount in a similar manner for 30 years. Someday I'd like to meet this fictional investor.

Did I miss a page of the paper? I didn't see anything about a withdrawal method.

I think the significant points of this study are drowned out by the effect of the "simplifying" assumptions. Its main value is to force investors to reflect upon the reasons behind their asset allocation and their market timing rebalancing criteria, especially after retiring.

I still don't see any reference in the body of this paper to any actions after the Terminal Portfolio Value. Nor in the Executive Summary, nor in the Summary and Conclusion. Am I that dense that I can't find it? I looked twice again for it. I sure hope the article at the link you are seeing, is the same as what I am seeing... if it's not, then I have no idea.

 

[Nords]

Did I miss a page of the paper? I didn't see anything about a withdrawal method.
I still don't see any reference in the body of this paper to any actions after the Terminal Portfolio Value. Nor in the Executive Summary, nor in the Summary and Conclusion. Am I that dense that I can't find it? I looked twice again for it. I sure hope the article at the link you are seeing, is the same as what I am seeing... if it's not, then I have no idea.

Sorry, not trying to be obscure, that was just a general comment about the lack of analysis of "variable" amounts in these types of studies. I didn't mean for the comment to apply solely to that study.