In a previous article, I expanded on Peter Lynch’s “high and low analysis.” That view looked at potential returns had you bought at the high and low price each and every year. This commentary takes the opposite view: seeing what happens if you sold at the high and low price each and every year. In a previous article , I expanded upon Peter Lynch’s “high and low” analysis. This involves looking at what would have happened if you invested at the very best and worst times (the high and low price, respectively) each and every year. The process was simple, but the takeaway is enormously instructive: “Investing at the high or low, especially over the long term, is not the difference between positive and negative returns. The difference between perfect timing and miserable timing over lengthy time periods is perhaps a couple of percent. Once you figure this out, it becomes clear that you should be focusing on the amount you can contribute and utilizing a long time frame, rather than concerning yourself with unknowable short-term fluctuations.” In Lynch’s example, the difference between perfect yearly purchases and dreadful annual timing was about 1% per annum. In my example, the difference was slightly larger, but the basic conclusion remained intact: “it’s not about timing the market, it’s about time in the market.” These demonstrations were based on the purchase side: “what happens if I bought each and every year?” For this article, I wanted to focus on the selling side. To make the process simple, we can rely on the same high and low price information as generated by the SPDR S&P 500 Trust ETF (NYSEARCA: SPY ). Here’s a look at the low price from each year dating back to 1996: Year Low Date Price 1996 1/10/96 $59.97 1997 4/11/97 $73.38 1998 1/9/98 $92.31 1999 1/14/99 $121.22 2000 12/20/00 $126.25 2001 9/21/01 $97.28 2002 10/9/02 $78.10 2003 3/11/03 $80.52 2004 8/6/04 $106.85 2005 4/20/05 $113.80 2006 6/13/06 $122.55 2007 3/5/07 $137.35 2008 11/20/08 $75.45 2009 3/9/09 $68.11 2010 7/2/10 $102.20 2011 10/3/11 $109.93 2012 1/3/12 $127.50 2013 1/3/13 $145.73 2014 2/3/14 $174.17 Additionally, we have the high price available as well from the same article: Year High Date Price 1996 11/25/96 $76.13 1997 12/5/97 $98.94 1998 12/29/98 $124.31 1999 12/29/99 $146.81 2000 3/24/00 $153.56 2001 2/1/01 $137.93 2002 1/4/02 $117.62 2003 12/31/03 $111.28 2004 12/29/04 $121.36 2005 12/14/05 $127.81 2006 12/14/06 $143.12 2007 10/9/07 $156.48 2008 1/2/08 $144.93 2009 12/28/09 $112.72 2010 12/29/10 $125.92 2011 4/29/11 $136.43 2012 9/14/12 $147.24 2013 12/31/13 $184.69 2014 12/29/14 $208.72 Let’s imagine that you want to supplement your dividend income , such that you take all of the dividends and also begin selling some shares along the way. Now, assuredly it is the goal of a great deal of people to never have to sell a share. However, that doesn’t mean that everyone must follow this route. Nor does it mean that we can’t think about the process. For illustration, let’s imagine that you have a portfolio balance of $1,000,000 back in 1996 (the number isn’t important, just the underlying math). Your idea is to buy shares in an assortment of holdings (in this case an index fund), collect the dividend payments and supplement this by selling an amount of shares each year. Let’s imagine that you want to sell $25,000 worth of shares beginning in 1997, followed by a 2% larger amount in the subsequent years. Here’s what your sold shares would need to amount to through the years: Year Sold Shares 1997 $25,000 1998 $25,500 1999 $26,010 2000 $26,530 2001 $27,061 2002 $27,602 2003 $28,154 2004 $28,717 2005 $29,291 2006 $29,877 2007 $30,475 2008 $31,084 2009 $31,706 2010 $32,340 2011 $32,987 2012 $33,647 2013 $34,320 2014 $35,006 Note that I’m taking a bit of a shortcut here. In reality, instead of focusing solely on the amount of shares you want to sell each year, you’d like to focus both on dividends received and the amount of shares needed to sell in a given year depending on your income requirement. However, with the high and low prices varying dramatically in dates, it’s a rather manual process to figure out the dividends received. It’s not uniform such that sometimes you have more than a year between the high or low price in two consecutive years and sometimes the time frame is just a few months. However, for our purposes, the illustration of selling a certain dollar amount of shares each year will work nicely. Let’s begin by seeing what happens in a “best case” scenario. The low price in 1996 was just under $60 per share. If you bought at this time, you would have been able to purchase 16,675 total shares. Additionally, your expected annual dividend income would have been about $21,000. We can now move on to the process of selling. In 1996, you were able to purchase shares at the best possible time. Let’s presume that you’re also able to sell shares at the best possible time – the highest price of every single year. In 1997, the highest share price was just under $99. In order to generate $25,000 in additional income, you would need to sell roughly 253 shares. As a result, your share count would go down to 16,422 or thereabouts. And so the process continues; in 1998, the highest share price was just over $124. In order to generate $25,500 in supplemental income, you would need to sell 205 shares, bringing your total share count down to 16,217. Here’s a look at your year-end share count from 1996 through 2014 if you kept selling at the high each year: Year Shares 1996 16,675 1997 16,422 1998 16,217 1999 16,040 2000 15,867 2001 15,671 2002 15,436 2003 15,183 2004 14,947 2005 14,718 2006 14,509 2007 14,314 2008 14,100 2009 13,818 2010 13,561 2011 13,320 2012 13,091 2013 12,905 2014 12,738 At first glance, this looks like pretty bad news. You started with 16,675 shares, and nearly two decades later, you’ve sold almost 4,000 shares, bringing your total share count down to “just” 12,700. Yet, it’s important to remain cognizant of what this indicates. Your share count has been reduced, that much is obvious. What’s not as apparent is the idea that you would actually be getting richer over time. Your $1 million beginning portfolio balance would now be worth nearly $2.7 million. Further, in the last 12 months, this amount of shares still would have generated $50,000 worth of dividend payments. Contrary to what many suppose, selling shares doesn’t have to be an exhaustive process to zero. I’m not necessarily personally advocating for this method, but it’s instructive to know that this option exists nonetheless. In this particular case, you would have collected hundreds of thousands in dividends, sold over half a million in shares through the years and still ended up much richer. Of course, this was also a best-case scenario – buying at the low and selling at the high every single year. Let’s take a look at the “worst case”: buying at the high and selling at the low every single year. In 1996, shares of the index traded as high as $76. Had you purchased shares at this price, you would have begun with 13,135 total shares – noticeably lower than the “best” case of buying at the low. Additionally, your expected annual dividend income would be about $17,000. (We could adjust for this in the example, but the illustration is the important part). In 1997, the low share price reached about $73. In turn, in order to reach your $25,000 supplemental income goal, you would need to sell about 341 shares. This would bring your total share count down to fewer than 12,800 (basically where the other example ended). Wash, rinse, and repeat. Here’s a look at your year-end share count each year after selling shares to reach your additional income goals: Year Shares 1996 13,135 1997 12,795 1998 12,518 1999 12,304 2000 12,094 2001 11,816 2002 11,462 2003 11,113 2004 10,844 2005 10,586 2006 10,343 2007 10,121 2008 9,709 2009 9,243 2010 8,927 2011 8,627 2012 8,363 2013 8,127 2014 7,926 Once more it’s obvious that your share count will be reduced each and every year. Moreover, it’s also apparent that this situation is markedly worse than the “best case” scenario. Yet, the output provided here is perhaps even more instructive. You began with a $1 million balance that was purchased at the worst possible price point. Then you went on to make sale after sale at the worst possible time each and every year. However, those 7,900 shares would now be worth almost $1.7 million. Further, over the last 12 months, these shares would have provided over $31,000 in dividend income. Expressed differently, even if you bought and sold at the worst possible moments, you would still get richer over time (with a larger cash flow to boot). The reason this works is due to “selling in moderation.” Obviously, you can’t go out and sell 15% of your portfolio each and every year and expect to end up with a higher portfolio balance over time. However, when done purposefully, selling shares and getting richer do not have to be opposite notions. The thing of it all is that you’re not going to complete either exercise. You’re not going to have perfect timing and you’re not going to have the worst possible timing year-in and year-out. Mathematically it just won’t occur. Yet, even if you did, at least in this illustration, it has been no great tragedy. Regardless of the situation you still ended up with a higher balance. Much like the previous example of consistently buying, it seems that having an underlying plan – rather than figuring out the “best” timing – is a much more important factor. Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. (More…) I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.