6

I'm a young person with some money to start investing. I do believe in the superiority of low cost index fund over actively managed fund. But I still freeze up a bit about the timing of when to start my first buy in.

Sure, I understand that there's no way to time the market. But even if I only buy-and-hold instead of day trading, wouldn't the timing of my first investment still matter greatly? For example, a person who started to invest during the 2007 bubble only regained his net worth 6 years later in 2013.

Is this the case of something that does matter but I have no control over so I should not worry about? Or does the timing NOT matter since I'm supposed to invest part of my paycheck every month no matter the economic condition?

Heisenberg
  • 681
  • 7
  • 16

4 Answers4

7

If you're worried about investing all at once, you can deploy your starting chunk of cash gradually by investing a bit of it each month, quarter, etc. (dollar-cost averaging). The financial merits and demerits of this have been debated, but it is unlikely to lose you a lot of money, and if it has the psychological benefit of inducing you to invest, it can be worth it even if it results in slightly less-than-optimal gains.

More generally, you are right with what you say at the end of your question: in the long run, when you start won't matter, as long as you continue to invest regularly. The Boglehead-style index-fund-based theory is basically that, yes, you might save money by investing at certain times, but in practice it's almost impossible to know when those times are, so the better choice is to just keep investing no matter what. If you do this, you will eventually invest at high and low points, so the ups and downs will be moderated.

Also, note that from this perspective, your example of investing in 2007 is incorrect. It's true that a person who put money in 2007, and then sat back and did nothing, would have barely broken even by now. But a person who started to invest in 2007, and continued to invest throughout the economic downturn, would in fact reap substantial rewards due to continued investing throughout the post-2007 lows. (Happily, I speak from experience on this point!)

BrenBarn
  • 23,964
  • 6
  • 60
  • 80
  • 1
    Thanks for pointing out the fallacy in my 2007 example. That really clears my head. – Heisenberg Aug 19 '14 at 05:33
  • 1
    An auxiliary question: Does bogglehead investing ever talk about the fact that during boom years we tend to have more money to invest, whereas during recession there's high chance of being laid-off, etc. Thus, we are in fact only able to invest when it's less beneficial? – Heisenberg Aug 20 '14 at 18:25
  • 1
    @Heisenberg: Interesting question. I think the ups/downs that are meant to be smoothed out by steady investing are more ups/downs in the market than in the economy as a whole. If you find yourself losing your job every time the stock market goes down you have bigger problems than buying low. You could ask that question on the Boglehead forums and see what the true Bogleheads say. – BrenBarn Aug 20 '14 at 18:48
  • @Heisenberg If you did post in Bogleheads, could you link it here? – Calculus Knight May 01 '15 at 21:16
  • 1
    @Tachibanaian Here. Hope it's helpful to you! – Heisenberg May 01 '15 at 21:19
  • @Heisenberg Absolutely! You discuss very interesting points. – Calculus Knight May 01 '15 at 21:24
2

Yes timing does matter.

Using a simple Rate of Change indicator over the past 100 days and smoothed out with a 50 day Moving Average, I have plotted the S&P 500 since the start of 2007.

S&P 500

The idea is to buy when the ROC indicator crosses above the zero line and sell when the ROC indicator crosses below the zero line.

I have compared the results below of timing the markets from the start of 2007 to dollar cost averaging starting from the start of 2007 and investing every 6 months. $80k is invested in both cases.

For the timing the market option $80k was invested at the start of 2007, then the total figure was sold out when a sell signal was given, then the total amount reinvested when a new buy signal was given.

For the DCA option $5000 was invested every 6 months starting from the start of 2007 until the last investment at the start of July 2014.

The results are below:

Timing the Market vs DCA

Timing the markets results in more than double the returns (not including dividends and brokerage).

Edit

It has been brought up that I haven't considered tax in my Timing the Market option. So I have updated my timing the market spread-sheet to take into account both long-term and short-term CGT in the USA for someone on the highest tax bracket. The results are below:

Timing the Markets after CGT

The result is still almost a 2x higher returns for the timing the markets option.

Also note that even with the DCA option you will have to sell one day and pay CGT on any profits there. However, the real danger with the DCA option is if you need to sell during a market downturn and not make any profits at all.

Victor
  • 20,991
  • 6
  • 47
  • 85
  • By timing do you mean that I'm able to buy low and sell high every time? – Heisenberg Aug 20 '14 at 18:24
  • If I read you right, this simulation does not factor in taxes. It's important to note that doing this in a taxable account will reduce the gains. Although it may be possible for an IRA or other tax-advantaged account, most people cannot contribute $80k to such an account all at once. In addition, from a psychological perspective, buying and selling your entire stake in response to market signals is not for the faint of heart. – BrenBarn Aug 20 '14 at 20:35
  • 1
    @Heisenberg - by timing the markets I mean what I said in my answer, when you get a buy signal you buy and when you get a sell signal you sell. Basically the ROC indicator is a momentum indicator. Using a longer term look back (100 days compared to say 10 or 20 days) allows you to capture medium to long term trends in the market. A buy signal is usually given after the market has bottomed and has started to rise again. A sell signal is given after the market has reached a top and is starting to fall. – Victor Aug 20 '14 at 23:41
  • 2
    I would totally love to see 2 sides on this debate, especially since all the evidence I've found is that index fund performs better than managed fund. – Heisenberg Aug 21 '14 at 00:07
  • 3
    @Heisenberg - I am in no way talking about managed funds, in fact I would never recommend buying managed funds. I am talking about Dollar Cost Averaging into an ETF vs timing your buys into and sells out of the same ETF. And what I have shown is evidence. – Victor Aug 21 '14 at 00:11
  • 1
    @BrenBarn - Your right I didn't consider taxes, so I have updated my answer to include both short-term and long-term CGT for someone in the USA on the highest tax rate. Still it is almost double the returns as compared to the DCA option. And you will also have to pay CGT on the DCA option once you do eventually sell. Some people may have $80 to invest at once others don't - but if they don't they can start investing with what they do have and invest more as they save more on the next buy signal. – Victor Aug 21 '14 at 00:18
  • 1
    @BrenBarn - From a psychological perspective I would say that having thousands of dollars invested in a falling market would be more nerve racking than investing a large amount when you see your funds growing in a bull market and then taking your capital and profits out to cash when you see the market falling sharply in a bear market. No one ever has gone broke in taking a profit. – Victor Aug 21 '14 at 00:23
  • Could this strategy be improved by only considering buy signal when investing my monthly paycheck? That way there is no tax involved, plus it's less nerve-racking since I'm only holding to my paycheck temporarily instead of investing the day I get it. – Heisenberg Aug 21 '14 at 01:10
  • 1
    @Heisenberg - what are you on about? The whole point is to buy when the market is about to take off and sell and take profits when the market has become overheated and starts to fall. If you don't sell you will not make any profits. Simple, if you don't want to pay any tax don't make any profits ! – Victor Aug 21 '14 at 04:16
  • I don't want to pay short term capital gain tax plus frequent brokerage fee though, is that correct? – Heisenberg Aug 21 '14 at 04:19
  • 1
    @Heisenberg - In the Market Timing eg. above there are 7 transactions (4 Buys & 3 sells) over 7.5 years compared to 16 buy transactions with the DCA eg. So with the market timing strategy there is an average of one transaction per year - that is not frequent in my books. Also there is only one occurrence of short-term CGT and the tax paid is only $504.80. If you don't want to pay short-term CGT then don't make a profit - keep your money invested in a falling market and watch your profits quickly dry up. Paying tax is actually a good thing - it means you actually made money and not lost it. – Victor Aug 21 '14 at 04:30
  • 3
    Have you tried applying this strategy to another time period, such as 1980-2000? – Kaleb Brasee Aug 21 '14 at 16:08
  • 1
    @KalebBrasee - I didn't have data going back to 1980 but compared between 1988 to 2000 - the results were more comparable with DCA returning 140% and Timing the Markets 172%. However, the real power of timing the markets really comes in by avoiding and being out of the market during steep downtrends similar to those that commenced in mid-2000 and late-2007. And if you really want to boost your profits during these big bear markets instead of just cashing out you can also go short or buy into a bear ETF. – Victor Aug 22 '14 at 05:17
  • @victor how did you source this data and what software are you using to graph it? – Gramatik Feb 09 '18 at 21:28
  • @Gramatik - the charting software is Incredible Charts, but you could find the same indicators on most charting packages. The prices were extrapolated from the chart for each given date for the DCA case and again for the Timing The Market case were extrapolated from the chart whenever a buy or sell signal was given. So the data is actual price data for the days selected. – Victor Feb 10 '18 at 03:03
  • @Gramatik - also, the calculations are just done on a spreadsheet. – Victor Feb 10 '18 at 03:14
1

When you start investing makes a very large difference to the outcome, but that is on the time scale of what generation you were born into, not what week you choose to open your 401(k). As you note in your last sentence, there is nothing that you can do about this, so there is no point in worrying about it.

If you could successfully market time successfully, then that would make a difference even at smaller time scales. But you probably can't, so there is no point in worrying about that either.

As BrenBarn points out, your statement about not regaining their net worth until 2013 applies to someone who invested a lump sum at the 2007 peak, not to someone who invested continuously throughout.

By my calculation, if you started continuously investing in a broad market index at the peak (around Jun 4, 2007), you would have recovered your net worth (relative to investing in a safe instrument that merely kept up with inflation, a hard thing to find these days) around April 12, 2010. I've done the computation on each business day because that is easier, so it might be slightly worse if do the periodic investment on each payday which is much more realist for a 401(k).

(And of course if you need to preserve/recover you net worth in 3 years, you shouldn't be in stocks in the first place)

jjanes
  • 1,141
  • 5
  • 8
1

A much less verbose answer is. Don't worry about buying low. You have a whole lifetime to dollar cost average your retirement dollars.

JohnFx
  • 52,979
  • 12
  • 134
  • 245
Mark Monforti
  • 1,782
  • 10
  • 15