Log in

No account? Create an account
Previous Entry Share Next Entry
The dot.com bubble...
Here's an image for you to consider:

Go to the Google Finance page for the S&P 500 index. On the chart, select the "settings" tab, and select the logarithmic vertical scale. Fiddle with the controls on the bottom so that you are looking at the timeframe from 1985 to present (essentially, the last 20 years).

It's possible that this link will take you to the chart in question.

What I noticed is that from 1985 to octoberish 1996, the graph is roughly linear. From octoberish 2002 to present, the graph is roughly linear. The dot.com bubble is clearly visible as an excursion between two periods of fairly constant growth. Even more interesting, pre- and post-bubble periods on the log graph appear to be on the same line, suggesting that the dot.com bubble was overlayed on top of a two-decade period of constant growth of the fundamentals.

Has anyone else heard of this sort of long-term analysis?

I looked it up when I heard of news reports claiming that from 31 December 1999 to now, the S&P lost money, and investors were calling the 2000's the "lost decade". From the perspective of the straight-line I'm seeing, this decade wasn't lost, but rather something unusual was plonked on top of the late 1990's. The comparison point they are choosing, 31 December 1999 was within a month or two of the dot.com peak, and was within the top 5 highs of the S&P500 during the bubble. Seems a somewhat cherry-picked date for a pessimistic report. For perspective, $10000 invested in the S&P500 on 31Dec1999 (and not reinvested) would be worth $8645 today. $10000 invested on 4Oct2002 would be worth over $15000 today. But no one comments on the 8+%APY average for the past 6 years.

  • 1
I don't think you are taking into account that the S&P constituent stocks change over time, with some added and others dropped. Looking at that might change your results. All the indices get rejiggered, with losers getting the boot and winners added. How that skewing affects returns is something the casual investor doesn't usually know about. Mutual funds often gloss over this, BTW.


I'm sure you'll agree that it is simplistic to say "If you invested $xxx in the S&P500 in such-and-such year, it would be worth $yyy today" for a number of reasons. Dividend reinvestment, index restructuring, etc. Yet people, even the press, do it.

I don't have the time nor initiative to evaluate how an investor, dollar-cost-averaging over the same time periods into a tax-advantaged account would have fared. That would involve tracking weekly index values and hunting down dividend payments for all the companies in the S&P500 for the period in question. Luckily, since the indices try to maintain continuity across constituency changes, and since a tax-advantaged account doesn't have to pay capital-gains, I wouldn't have to pay too close attention to the constituency changes, except to track dividends.

You're missing the point I was trying to make.

In your post, you were talking as though the earlier S&P was the same as the later S&P -- an "apples to apples" comparison. The problem is that it's really not, even though it's marketed as such. The industries represented can change wildly in their proportion of the overall indices used in mutual fund benchmarking (like the S&P, Russel and Dow).

The indices are mostly chosen with respect to their market capitalizations, so that tends to favor the "popular" stocks and disadvantage the less-than-popular ones. This is one of the reasons why your actually comparing apples to oranges, not to apples. The industries represented in these benchmark indices have changed quite a bit over the timespan you indicated, with more technological industial proportional representation at the end of the timespan than there was at the beginning. A lot of those companies got "shaken out" during the dot-com boom/bust.

You are also ignoring the effects of world events, namely 9/11 and the start of the war in Afghanistan and Iraq -- events that had a major effect in the year prior to your "boom close date" of October 20002. There was a huge selloff in the markets, across the boards, at that time, as investors fled to safer investments like Treasuries and dumped stocks -- the Dow dropped to around 7000 during that period, from well over 10,000 pre-9/11.

It was after your "boom close date" that stocks began their recovery as victory was declared in Iraq so that now they are returning to their pre-9/11 levels. That dumping of stocks is what makes the gains seem impressive, when really they aren't, when you take a longer, more historical look at them. That 8% figure you quote is really an artifact.


As a former economist at the Dept. of Labor, i did a *similar* study, though not the same. The National Bureau of Economic Research (www.nber.gov) defines a recession as "The NBER does not define a recession in terms of two consecutive quarters of decline in real GDP. Rather, a recession is a significant decline in economic activity spread across the economy, lasting more than a few months, normally visible in real GDP, real income, employment, industrial production, and wholesale-retail sales."

In that regard, the S&P Index - a mix of industrial and other funds of stocks designed to be a proper gauge of the overall health of the stock market- roughly follows our recessionary cycles. A stock market, by definition, is a market trading in risk. A person or fund invests in a stock because they think it represents a long term growth that is better than, say, municipal bonds or a CD. Since investing in a private company inherently involves risk (as opposed to a government), stocks have a higher potential return for higher risk.

So, if you invested in an index fund that followed the S&P, net of any expenses (though most index funds are no-load), yes, for the years Oct 2002-May 2007, you would make a very tidy return. There are many people who will say the best years for investing were the 90s or the 80s, often depending on their political views, but the truth is that all 3 decades, in rough concert with the strength of the economy, were pretty good. The stock market is generally considered, taken broadly, a leading indicator. (You may hear the term used as leading economic indicator). Employment, inversely, is considered trailing. In other words, there are certain things that go bad before an economy is in recession (stocks, industrial manufactures, for example) and some react (the labor market, typically). For example, the official recession was March - November of 2001. Stocks began to slide in late 2000, per the S&P. If you invest in Sept. 2000 and cash out in Sep. 20002 ( a 24 month period) - you'd lose roughly 40% of your investment, assuming your investment followed the S&P exactly. Similarly, if you invested in Sep. 2002 and cashed out in November of 2007 or thereabouts, you'd have made an 80% return approximately.

Since 1984, when the information/productivity economy began to overtake traditional manufacturing, you'd wind up with a 750% return, investing $1 then and holding it till today. So, yes, if you bought a hypothetical S&P dollar in Sep. 2000, and you held it till now, you'd not get a real good rate of return, because you're buying top of market to top of market, but if you go bottom-to-bottom over time, 1990-2002, you'd get 225% return approximately on the S&P data.

The reason why people call the 2000s a lost decade is because the decade started really high and then cut back, going back up to a prior high. But the prior S&P high wasn't sustainable. It wasn't just the dot coms, it was a series of poor investments in companies which sustained no profits - and these markets imploded. But it wasn't a lost decade, it just began with a correction and is now in one, again. Wait till it gets the same amount of time - and then make the evaluation of the "lost" nature of the decade. There's a big difference between the two economies -peace and war, internet as novelty vs. internet as mature, etc- and so comparing the two, in terms of their investment risks - is not really apples to apples.

  • 1