7

I understand the theory behind dollar-cost averaging on the buy-side, and there are many articles about why it does (or doesn't work). But I can't find much, if anything, about its value when selling. So here's my situation:

Say it's Monday morning and I have a block of 500 shares currently worth $10000 that I must completely sell by the end of the week (therefore I can't set a market limit and try to sell on an uptick). I have no idea which way the stock price is going to move or what its volatility is. Is any one of these three strategies better than the other two?

  1. Pick a day at random and sell all 500 shares at noon on that day, regardless of the price.
  2. Sell 100 shares every day at noon, regardless of the price.
  3. Every day at noon, sell $2000 worth of shares, regardless of the price. On Friday, sell everything (assuming there is anything left).

With #1, I have one commission to pay; with #2 and #3, I pay 5 commissions. It seems to me that the only advantage to the latter two strategies is to reduce my risk a little, but doesn't increase my expected return. And the extra commissions would actually reduce it. Am I missing something?

3
  • 1
    On this short of time frame and this amount of $$$, I'd set a limit order at your price during the week, then on FRI, if the limit order hasn't hit, take what you can get. Mar 23, 2015 at 4:00
  • One thing worth realizing is that buying vs. selling is not the real issue here. After all, buying is the exact same thing as selling: you're exchanging one thing of value for something else of value. The real problem you're asking about here is what to do when you need to buy/sell within a certain time frame. And while the point of dollar-cost averaging is usually to space out transactions over a long period of time, there's no reason why you can't make it work over a week. Your option (2), for example, sounds reasonable (if you ignore commissions).
    – dg99
    Mar 23, 2015 at 21:37
  • 2
    A corollary to this is "when should a retired person make IRA/401(k) distributions: beginning of the year, or each month?
    – RonJohn
    Aug 9, 2021 at 20:51

4 Answers 4

4

None of your options or strategies are ideal. Have you considered looking at the stock chart and making a decision? Is the price currently up-trending, or is it down-trending, or is it going sideways?

As Knuckle Dragger mentions, you could just set a limit price order and if it does not hit by Friday you can just sell at whatever price on Friday. However, this could be very damaging if the price is currently down-trending. It may fall considerably by Friday.

I think a better strategy would be to place a trailing stop loss order, say 5% from the current price. If the stock starts heading south you will be stopped out approximately 5% below the current price. However, if the price goes up, your trailing stop order will move up as well, always trailing 5% below the highest price reached. If the trailing stop has not been hit by Friday afternoon, you can sell at the current price. This way you will be protected on the downside (only approx. 5% below current price) and can potentially benefit from any short term upside.

9
  • I should have been a little more clear. Mar 23, 2015 at 17:52
  • I personally use market limits to maximize my gain. This is a hypothetical question, derived from a conversation I had with a colleague who insisted that Option #2 will, over the long term, generate a better outcome than Option #1, and I can't figure out how -- considering the extra commissions. But it got me to thinking about the sell-side version of dollar-cost averaging (DCA). While the web is full of the pros and cons of DCA for buying, I couldn't find anything about the sell-side. If these are the only 3 options, is any one better than the others? Mar 23, 2015 at 18:00
  • 1
    @BobTreitman, why would #2 be better than #1? If the price is dropping during the week you would get a worse price overall and pay more commission. DCA doesn't work buying and it doesn't work selling. If you can't read a chart then just decide what price you want to sell at and sell it, although the trailing stop loss option would be the best way to go.
    – user9822
    Mar 23, 2015 at 20:28
  • Personally, I can't see how it could be better. But given all the interest in DCA on the buy side (not saying it works, but there are LOTS of people who claim it does), why doesn't anyone propose a corresponding sell-side strategy? Doesn't anyone ever sell? ;-) Mar 24, 2015 at 2:26
  • Using a trailing stop loss is the best option because if the stock continues going up all week you will capture most of that price rise, if the stock goes down all week you will only lose the amount your stop is from the current price. If you sell a bit each day to get the average, you will lose out if the stock rises all week compared to selling at end of week and prices drop all week you will lose compared to selling at start if week. The best you will get is average and pay alot more commission for it.
    – Victor
    Mar 24, 2015 at 20:33
4

Is there a sell-side version of dollar-cost averaging?

Yes, it's called scaling out of a position.

There's a simple answer to your question. The direction of price determines the benefit.

If share price is rising, scaling out is better than selling all shares immediately. If share price is dropping, selling immediately is better than scaling out.

If you knew what share price would be in the future, you'd know the best exit strategy. No one knows that.

4

If you frame buying and selling as exchanging asset A for asset B, I think the sell-side becomes obvious.

DCA is for buying (A is $, B is stock). The generalized concept would be that by exchanging a constant amount of A over time regardless of how much B that equates to, you'll smooth out volatility as you transition your portfolio to B.

So, the sell-side would have A as stock and B as $: Choose a fixed number of shares and sell them on a fixed schedule until they're all gone. When the price is high, you'll cash in, and when the price is low you'll avoid divesting too much.

So in your 1-week scenario, you'd count the shares not $ amount of the stock and divide it by 5 and sell each day. Notice you don't have the "whatever's left" problem like when you subdivide the $10k.

1
  • This is the correct answer. Dollar cost averaging allows you to buy a little stock when it's high, and buy a lot of stock when it's low. When selling you want to sell a lot of stock when it's high, and sell a little stock when it's low (the opposite of DCA), so sell based on the unit you're getting rid of. DCA trades dollars-to-stock per dollar unit. Inverse DCA means trade stocks-to-dollar per stock unit. So selling 1 stock every day is what you want, the inverse of DCA.
    – JJrodny
    Feb 3 at 17:20
1

To answer the title of your question.

Is there a sell-side version of dollar-cost averaging?

Yes, there is. There's actually a spreadsheet out there that compares value averaging to dollar cost averaging for both buying and selling.

That spreadsheet is linked in this article.

https://seekingalpha.com/article/4285823-value-averaging-alternative-to-dca-lump-sum

Download an Excel copy to modify the highlighted parameters. You can select if you’re buying or selling, your time horizon, and how much you want to buy or sell.

1
  • Isn't it about value cost average? Can you explain when to sell with the DCA not looking at the price graph?
    – Psijic
    Nov 7, 2021 at 7:35

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .