historic returns

so, i always hear that the stock market roughly returns 10% annually over the long haul, but i started thinking about the basis of this figure, which i had previously pretty much accepted blindly. i did some research and found the following data on http://www.econlib.org/LIBRARY/Enc/StockPrices.html:

The average compound rate of return on stocks from 1802 through 1991 was 7.7 percent per year: 5.8 percent from 1802 to 1870, 7.2 percent from 1871 to 1925, and 10.0 percent from 1926 to 1991.

this means that this fabled 10% figure is actually based on data from 1926 until now (i'm assuming that from 1991 until now, the return has been roughly 10%). why do we throw out the earlier data? i'm willing to accept that the financial markets have grown and matured and that the nation and the world is on more financially stable ground today versus the 1800s, but there are several things that are still reasons to be wary about the magical 10% number.

baby boomers. this is a topic everyone has probably heard something about, so i won't spend a lot of time on it, but the fact of the matter is that at least part of the more recent returns can be attributed to the huge baby boomer population spike that has some impact on the financial trends of the past few decades. they have worked, spent, saved, and invested. and at some point, they will retire and tap into their investments. the question is will this reversal have a similar fall in the markets as money moves out of their investments to support their retirement?

investment vehicles. there are a lot more families with some level of investment in the stock market today versus even 20 years ago. in fact, today more than 50 percent of americans own stock up from 15.9% in 1983 (http://www.ici.org/pdf/rpt_05_equity_owners.pdf). i have to think that part of the reason for that is because the investment vehicles that are available today have made it easy for people to invest. 401ks, for example were 'invented' in 1978 and have provided a medium for the average worker to very easily invest in the stock market. not only that, with employer matching, there is good deal of money that is naturally going into the market on a day-to-day basis. now, i'm not saying that this is creating any artificial demand or anything of the sort, but the fact of the matter is that i don't have any data that speaks to what the effects of these investment vehicles have been over the past 20 years.

time line. the main point that i want to consider is not whether a retiring generation of 401k investors is going to trend the market in other directions, but whether a timeline of 80 years is long enough to safely contend that we should continue to expect 10% annualized returns over the long haul. if you look backwards, it's obvious that the market had significantly lower returns the 80 years prior.

the reality of the situation is this -- no one really has all the answers. and while i may not achieve a 10% return, i'm still stashing away my investment cash in the market. and while historic returns may not be an indicator of future performance, it still is better than stashing my money away in a mattress.